Ahead and get started. Very pleased to be here with the Neumora team at our healthcare conference. My name is Julian Pino, a member of Paul Matisse's team. I'm here with Josh Pinto, President, and Helen Rubinstein, Head of IR. Very excited to have you guys here. Josh, it would be great if you could just give an overview of Neumora, everything that you sort of want to highlight, and the state of the company right now. Thanks so much.
Absolutely. Thanks, Julian, for having us here. It's a really exciting time at Neumora. As we look across the next 12 months, we've got up to six clinical readouts across our four main clinical stage assets, which we just rolled out at our R&D day a couple of weeks ago. I'll maybe start with some of our earlier stage clinical programs and work our way up in terms of what to expect. NMRA-215 is our NLRP3 inhibitor, and we released class-leading weight loss data in the diet-induced obesity model at our R&D day a few weeks ago, really highlighting that NMRA-215 is the first NLRP3 inhibitor to show incretin-like weight loss, not only across a total study duration, but also from an induction phase.
We are very excited, given the high correlation between DIO models and clinical weight loss, to move this program into the clinic in the first quarter of 2026 and be able to deliver human clinical proof-of-concept data in monotherapy setting and the combination setting before the end of 2026 for that program. With NMRA-511 as well, we are moving quite rapidly with our phase 1b study in patients with Alzheimer's disease agitation towards our readout around the end of this year. Just as a reminder, that is a signal-seeking, randomized, double-blind, placebo-controlled study looking at assessing NMRA-511 versus placebo over eight weeks. The primary endpoint is the change from baseline to week eight on the Cohen-Mansfield Agitation Inventory. Our third franchise is really our M4 PAM franchise.
At R&D day, we announced that we had initiated a clinical study with NMRA-898 being our second M4 PAM to go into the clinic in the last few months, in addition to NMRA-861, which entered the clinic in July of 2026. Across that franchise, we have a strong degree of conviction that it really is M4 driving the antipsychotic benefit of the muscarinic class. To appropriately target the M4 receptor, we believe that the PAM approach is the right way to go. With this franchise, we're progressing two molecules, and we plan to provide a franchise update with data from the phase 1 SAD/MAD studies by the middle of 2026. Finally, our most advanced program, navacaprant, is our highly selective kappa opioid receptor antagonist.
We're continuing to make our way through the KOASTAL- 2 and 3 studies and are on track to read out KOASTAL- 3 in the first quarter of 2026 and KOASTAL- 2 in the second quarter of 2026. Julian, all of this is funded with our robust balance sheet. We announced last week that we ended the third quarter with over $170 million on the balance sheet, and that provides us cash runway into 2027. Fully funding the operational plan to deliver all six of the clinical catalysts we've just outlined.
Awesome. Amazing. That is a great overview. I guess just kind of going down the line in order would be great. Starting with the NLRP3, I know there are a lot of other programs out there, obviously some very interesting weight loss data that you showed. Just what about it makes it so exciting for you and gives you confidence that this is going to be able to show some weight loss in humans as well?
Yeah, absolutely. I think first and foremost, we just need to look at the model, Julian, and there's a high degree of correlation in weight loss between what is seen in DIO models and what is seen in the clinic. A recent Nature publication suggested the correlation could be even as high as 0.9 or higher. There's a large body of evidence looking at multiple mechanisms to show that the data does translate. I think that's the first piece that gives us confidence. Second is really just looking at why is our data differentiated relative to the rest of the NLRP3 inhibitors in the class from a weight loss perspective. Based on our work and work from other NLRP3 sponsors, it's clear that you have to achieve IC90 concentrations in the CNS ultimately to unlock the full weight loss potential for this mechanism.
In our DIO study, that was clearly our target, and we achieved that with our target dose of NMRA-215, where we achieved IC90 concentrations over a full 24-hour period. That is enabled by our best-in-class CNS penetrant pharmacology. We are a neuroscience company through and through. When we started our journey to unlock the potential of the NLRP3 inflammasome, we were very, very focused on developing novel chemical series that had best-in-class CNS penetration, ultimately to unlock the weight loss potential and the potential in neurodegenerative and other CNS conditions. We do not believe, Julian, based on the data we've seen, that a number of the other NLRP3 inhibitors out there do achieve IC90 concentrations in the brain, which is why we've seen some really exciting benefits on peripheral markers such as cardiovascular markers or inflammatory markers.
We have not seen other NLRP3 inhibitors generate clinical weight loss through studies.
Yeah, no, that totally makes sense. I think it's definitely an exciting hypothesis to hopefully we'll bore out this coming year in the data. I guess just like, were you surprised at all in the Ventyx data that it didn't show anything? Or I guess just what you know, just sort of on their penetration and exposures that it kind of all makes sense to you?
Yeah, absolutely. It was quite exciting to see some of the reductions they saw in cardiovascular biomarkers like CRP. I think that continues to validate that as a class effect. We'd also seen improvements in cardiovascular biomarkers in the non-THERA study where they also were not able to achieve clinically relevant weight loss. We think the data coming out of the Ventyx study clearly shows that improvement in peripheral cardiovascular biomarkers is likely a class effect. In terms of the weight loss, we frankly weren't surprised. If you look at their diet-induced obesity studies, they've shown about 2% weight loss based on what's in their corporate presentation on their website through a diet-induced obesity study. If you think that the model is highly translatable, you would not have expected to see any weight loss in the clinic, which is our belief, and ultimately played out.
We were not really expecting to see it. I think it all comes back to the pharmacology where we think it is absolutely paramount to hit IC90 concentrations in the brain to be able to achieve the weight loss, which is what NMRA-215 can do. We do not believe that other sponsors are effectively achieving that.
Got it. No, it makes sense. I guess I know these data are sort of fresh, but I guess thinking about your planned development path, what is sort of this next study going to look like first in human? When can we expect POC efficacy data on weight loss?
Absolutely. We'll come up with very specifics around the study design when we initiate the study in the first quarter. What I think you can expect is a fairly traditional early-phase obesity study design consistent with what other obesity sponsors have done recently, where we're going to look to do dose escalation through SAD and MAD studies. Ultimately, what we're going to look to deliver by the end of 2026 is 12-week human monotherapy in combination with a GLP-1 weight loss data as our full clinical human proof of concept.
Awesome. I guess just like in like a potential downside case, say the obesity results don't translate, are you still interested in sort of pursuing NLRP3 as like a cardioprotective agent? Where else could this potentially fit in the treatment paradigm?
Yeah. We fundamentally believe that the obesity and the weight loss data should translate from the DIO to the clinic, given what we've seen in some of the review papers and just the meta-analysis on prior DIO results translating. In the event that they didn't, Julian, I think we would still fully expect NMRA-215 to have all of the peripheral benefits we've seen from other NLRP3 inhibitors. We achieve well over IC90 concentrations peripherally as well as in the brain. Based on our DIO biomarker data where we've shown improvements in liver health, improvements in cardiovascular measures, as well as improvements in insulin sensitivity, we think that we absolutely should be able to retain those peripheral benefits in the clinic as their class effects. We can look at different paths forward for the molecule in that event.
I think in totality, really what we're looking at with our NLRP3 inhibitor is where do we point it first? I think in our view, that's obesity until proven otherwise. There are a number of other areas where it could go, whether broader cardiovascular disease, inflammatory conditions, neurodegeneration, which is why we've been working on a set of molecules here at Neumora. If there are other molecules that could be more appropriate for different indications, we're working on some follow-on compounds to bring forward as well.
Awesome. Great. Would love now to talk about the V1aR program. You have a readout expected by the end of the year, correct? Can you just recap sort of the biological thesis here and just sort of the design of your ongoing AD agitation study?
Yeah, absolutely. Julian, really the vasopressin system is thought to regulate stress response, anxiety, as well as agitation. There is a body of preclinical literature out there that really highlights how the vasopressin system could be implicated within agitation. We have also seen human studies run where if you administer intranasal vasopressin to a healthy individual, it will drive agitated and/or aggressive style behavior. It is clear that there is a correlation between the vasopressin system and agitation. In our preclinical work, we ran something known as the human threat test model where you expose a macaque monkey to a human they do not know. It drives anxious, agitated, aggressive response behavior in those monkeys. We showed that if you administer 511 ahead of that test, you clearly see a significant reduction in that behavior. In non-human primate models, we have shown the ability to improve this.
There have also been other V1a sponsors out there, such as Azovan, who have run studies in patient populations like Huntington's disease irritability, where they've been able to demonstrate a clinical benefit. We think there is a breadth of preclinical, clinical, and then experimental evidence from our molecules supporting the move forward into Alzheimer's disease agitation. In terms of the study that we're running, it's a two-part study. Part A was a multiple ascending dose study in healthy elderly patients to confirm the PK and safety tolerability in elderly is consistent with healthy normal adults. We presented the data at R&D day, confirmed it's safe in elderly, and the PK was appropriate to move forward. We're currently running the phase B of this study.
That is a randomized double-blind placebo-controlled eight-week study where we're going to look to assess NMRA-511 relative to placebo on the Cohen-Mansfield Agitation Inventory, which is the registrational endpoint for AD agitation. We're very excited to be ready to put those data out around year-end.
Got it.
I think just a couple of reminders on the study design. That study is not powered for statistical significance. To Josh's point, this is really a signal-seeking study where we're looking to understand what 511 could do for these patients. In addition to the CMAI, we've included a number of additional endpoints that take a look at not only the patient severity measures, but also the physician's perceptions of the patient's agitation and the caregiver experience at home. It will really give us a comprehensive view of how all the stakeholders around a patient are experiencing the agitation.
Got it. No, great. That all makes sense. I guess just before coming back to sort of expectations for the readout, is there anything that you've learned from some of the other V1aRs that have been tested in the clinic? What are the implications for your program?
Yeah, absolutely. Julian, as we're thinking and really analyzing the class, there are two products that we view that have come ahead of us or before us that have shown benefit in terms of reducing agitation in patients with Alzheimer's disease dementia. One is Rexulti, which is approved to treat Alzheimer's disease agitation. In their studies, Rexulti showed about a 20-point change on the CMAI with the high teens response in placebo. They looked at a range of severities, including milder to moderate patients in the outpatient setting, but also more severe patients in the inpatient setting. Axsome, which is developing Auvelity for the treatment of AD agitation, in their studies, showed a different response in the CMAI in the kind of low double digits on active for the CMAI and similar in the low double digits on placebo.
They've been able to produce a data package that they're going to put in front of the agency. As we look at it, really the value proposition for that program is a product that seems to have a safety and tolerability profile that is improved relative to what we've seen from Rexulti, with also being able to demonstrate an improvement in symptoms of AD agitation. We think based on early evidence from 215, it could even be more well tolerated than Auvelity. We think if we see data coming out of 215 on the active arm in a zone similar to where we've seen the Auvelity data, it should give us confidence that we can think about a path forward for development of the molecule.
Got it. No, it makes sense. I guess just the idea of this as being like a two-part study, signal-seeking study, I guess just thinking about the translatability of results from a smaller cohort to a larger randomized study and what you could say for that.
Yeah, absolutely. It's a consideration that needs to be taken into any study as you scale up from early phase to later, in particular in neuropsychiatry, where you're really going from small studies where you can manage placebo response to larger studies where it becomes more challenging. I think what we will look for here is just the magnitude of the signal and then thinking about how that scales to larger studies in the future. We will put appropriate controls around larger studies in terms of the number of sites we go to, the quality of the sites, and other things that we've implemented in the KOASTAL program to improve placebo response, such as placebo-controlled reminder scripts or other elements.
Got it. And potentially better tolerated than Auvelity, can you just remind folks of what are some of the common adverse events of that profile and which side effects may elderly patients be more sensitive to?
Yeah. I think really as you look at NMRA-511, so far our data suggested that the program and the product is very, very safe. I think as you consider an elderly population, it is not only the tolerability in the AEs, but it is also the dosing frequency and pill burden. As we know, elderly patients tend to be on a range of medication to control a number of disorders. As you think about 511, it looks like a BID product based on how it is formulated today. We might look at formulation ways to make it QD moving forward. As you think about it relative to Auvelity, not only does it have the improvements potentially in the side effect profile, but in the elderly population, Auvelity could be dosed BID or TID.
We think we'll even have an advantage there just on the pill burden.
That makes sense. Yeah, especially CID for elderly patients. That can be really challenging. Great. I guess just from a strategic point of view, is sort of everything on the table potentially take forward yourselves into randomized placebo-controlled study, collaboration, out license, just like how are you thinking about it?
Yeah. At this point, I think what we've set up is we've gotten all four of the programs past critical inflection points and on their way to their next one. As we've outlined, we have six clinical data readouts and catalysts across these four programs over the next 12 months. They're all fully funded. We are working diligently to execute as well as we can on the operational plans to deliver high-quality data. Julian, we're really just going to let the data speak for itself in terms of what we do next with the programs. I think we've got a team that is absolutely capable of taking any one of these all the way. We've shown our ability to raise capital in a pretty significant way throughout our funding history.
I think what we're going to do is just be prudent as we move forward and look at what is the optimal path not only for the molecules, but for patients as well as shareholders.
Excellent. Great. Would love to now switch gears and talk about the M4PAM franchise. What's the latest there? Maybe if you could just give a brief sort of background on what makes the M4PAM different from some of the other muscarinic compounds that are out there.
Yep, absolutely. In terms of our take on the muscarinic class, we have a high degree of conviction that M4 is really the receptor that's driving the antipsychotic benefit of this class. We think the optimal way to target M4 is through a PAM approach, not necessarily through a selective agonist approach, because we think that you can more selectively target M4 that way and potentially reduce the AEs that exist. Based on all the preclinical and clinical evidence we've seen across the class and some of our team members having spent 20-30 years in this field, we fundamentally believe M4 is the right way to go.
In terms of recent updates from Neumora, we announced at our R&D day in late October that our second M4PAM, NMRA-898, had entered the clinic in addition to NMRA-861, which was our first M4PAM from this next generation that entered the clinic in July of this year. What we're really looking for out of this class is to bring forward what we think is the next wave of M4PAMs. The first wave that we believe is optimized for CNS penetration and has the ability to be highly brain penetrant to fully unlock the potential of this compound. As we put out in some of our slides, we do believe that some of the first-generation M4PAMs, such as emraclidine, do have fundamental pharmacological liabilities in terms of their opportunity to truly drive robust CNS penetration.
Yeah. Obviously, there's a lot of theories out there with what happened with emraclidine failed in the late-stage studies, KarXT and AbbVie. What do you think sort of led to that? What are some learnings we can take away? Is it all sort of drug property related? Or thoughts on potential execution? It would be great to hear your thoughts on that.
Yeah. I'm not going to speculate really on what caused the failure. What I will talk about is the data we saw out of that program as it was being developed that caused us to shift and optimize 898 and 861 for high CNS penetration. We had seen a number of years ago that a publication with emraclidine had been put out showing their human PET receptor occupancy data. In that publication, if you double the dose from 15 to 30 milligrams, you double the peripheral plasma concentration, which is what you'd want to see from a drug that has linear kind of dose proportionality. You do not see any increase in CNS receptor occupancy, which led us to believe there is clearly some form of brain saturation effect going on here with the compound.
As we were developing and designing 898 and 861, we always wanted to make sure with our compounds we were fully characterizing the CNS exposure potential of those compounds to ensure that we were overcoming what we thought were limitations for emraclidine. To do that, we also ran emraclidine head-to-head in those same assays. What we've demonstrated is that NMRA-861 and 898 are highly permeable into the brain, whereas emraclidine seems to be right on the borderline in preclinical cellular assays if it does have penetration potential into the brain. If we look at liabilities for being effluxed out of the brain, which is based on some PGP enzyme transporters, what we know with 898 and 861 is there is very low liability that they will be effluxed from the brain, whereas with emraclidine, there's moderate to high efflux potential.
Even if the compound's getting into the brain, it's being effluxed out, which in our mind could be explaining some of this saturation challenge that they saw in their human receptor occupancy study.
Got it. Yeah, really interesting. Thinking about your phase I program, going to be entering first in human studies, right? What gives you the confidence that you'll be able to, I guess, or I guess have you decided if the data is supportive, would you advance both? Are you going to try to choose between one or the other? How are you sort of thinking about that ahead of the data?
Yeah, I think right now what we're thinking is we'll likely pick a lead asset, either 861 or 898 for acute schizophrenia, which will be the first indication since that is the indication that is currently clinically validated. There are a range of other indications that you could go into, other neuropsych indications, neurodegenerative conditions, and others. Really the benefit of having two compounds is we can run a best horse race early in development to see, is there anything that separates the molecules as they move into the humans for the first time that would cause us to select one over the other? If they look fairly similar, what we can think about is potentially a franchise of molecules that we can develop in different indications that would ultimately require different commercial products and potential pricing.
I guess just generally speaking, I mean, there's been like a couple, we already mentioned one, but there's been like a couple setbacks in the muscarinic space, I guess, and sort of you guys are in phase I. I guess what gives you confidence just on the class in general and being there for it and as well as thinking about potential path of differentiation later down the line as you think about getting to market?
Yep. I think what gives us confidence in this target and the class is there's decades of evidence that supports the muscarinic's role in a range of disorders, both central and peripheral. As we look at the class, you've seen with KarXT that they have shown positive data and replicated it. We've seen with emraclidine positive early stage data that was not replicated in late stage studies, and some positive signals that have Neumora. As we know in neuropsychiatry, some of the world's leading blockbuster drugs today have failed multiple phase III studies before hitting positive. In neuropsychiatry, we do not really view the first phase III study as being a write-off on the class or the mechanism. A lot of the time, it can come down to execution challenges as we've seen.
We just think there's such a paucity of evidence here supporting the role for M4s and muscarinics in the role to treat schizophrenia that that's truly, Julian, what gives us kind of a lot of confidence.
No, I think that makes sense. A lot of the failed studies don't make it in the FDA label, right? So I think people tend to forget that. Great. That was a great discussion on the muscarinics. I guess just to wrap up, would be great to hear the latest on the core program and anything else that you want to highlight.
Yeah, absolutely. As we've commented earlier in the year on the back of the KOASTAL-1 data, we made a number of changes to improve ultimately ensuring that we are getting the right patients into the study that have chronic major depressive disorder. Some of those measures that we introduced were we partnered with Mass General Hospital, CTNI, the SAFER group to really have them help do some of the medical monitoring and ensure that the patients go in. What they do is they look at all the screening information for the patients, their prior medical and prescription history, and conduct a proprietary interview with the patient to confirm that a Mass General clinical psychiatrist believes that that patient truly does have MDD and is appropriate for the study. That is working as we would want it.
We've seen substantially higher screen fail rates, and we're not seeing the abnormalities in baseline patient characteristics that we had seen out of KOASTAL-1 . We are building confidence internally that that is doing what we would want it to do. We also employed the VCT database to help identify patients that could be professional in nature, whether enrolling in other MDD studies or other non-MDD clinical trials that might not be appropriate for our patient population. What I can say is that tool has excluded a number of patients that otherwise would have gotten in. We feel like both of those measures are doing what we would have wanted them to do. Both KOASTAL -2 and 3 are on track for their respective readouts in the first half of 2026.
Got it. Very exciting. I guess just thinking about just Neumora, where do you see Neumora being a year from now or even a few years from now?
Yeah, I see ultimately we're at a critical phase now where we've spent a lot of time this year reorienting the pipeline, setting up the catalyst that we have through 2026. Ultimately, Julian, where I see Neumora getting to at the end of 2026 is where the company really started. I see Neumora exiting 2026 as a really big neuroscience company.
Amazing. Great. I guess we can wrap up. Thank you so much, Josh and Helen. This was awesome.
Great. Thanks, Julian.