Acadia Pharmaceuticals at TD Cowen Summit: Strategic Pipeline Insights

On Thursday, 18 September 2025, Acadia Pharmaceuticals (NASDAQ:ACAD) participated in TD Cowen’s 5th Annual Novel Mechanisms in Neuropsychiatry & Epilepsy Summit. The conference call, led by analyst Ritu Baral, focused on Acadia’s strategic pipeline developments. The discussion highlighted both potential opportunities and challenges, emphasizing the company’s commitment to addressing unmet needs in neuropsychiatry and rare diseases.

Key Takeaways

• Acadia is preparing for the Phase 3 COMPASS data readout for ACP-101 in early Q4.
• The company’s pipeline includes promising candidates for Prader-Willi syndrome, Alzheimer’s disease psychosis, and major depressive disorder.
• Acadia is exploring business development opportunities, focusing on neuropsychiatry and rare diseases.
• Biomarker strategies and trial designs were key topics, with emphasis on data-driven decision-making.
• The company aims to expand its expertise in rare diseases more broadly.

Operational Updates

ACP-101 (Carbetocin) for Prader-Willi Syndrome:

• Phase 3 COMPASS trial data readout expected in early Q4.
• Target enrollment of 170 patients, randomized one-to-one (3.2 mg vs. placebo).
• Primary endpoint: Hyperphagia Questionnaire for Clinical Trials (HQ-CT).
• Secondary endpoints include clinician global impressions and HQ-CT responder analysis.

ACP-204 (5HT2A Inverse Agonist):

• Phase 2 and Phase 3 program ongoing for Alzheimer’s disease psychosis.
• Target enrollment for Phase 2: 318 patients across three arms.
• Lewy body study starting soon with a target enrollment of 180 patients.
• Biomarker confirmation is required in the ADP trial.

ACP-211 (Deuterated R-Norketamine):

• Phase 2 study starting in Q4 with approximately 150 patients.
• Focus on efficacy and safety, with minimal sedation and dissociation observed.

Future Outlook

• Acadia is interested in business development opportunities in neuropsychiatry and rare diseases.
• The company is focusing on Phase 2-ish assets but remains open to earlier or later-stage assets.
• Key pipeline catalysts include the ACP-101 data readout and the initiation of new trials.

Q&A Highlights

• Discussion on the potential for ACP-101 approval based on clinical benefit.
• Clarification on the powering of the Phase 2 ACP-204 study for Alzheimer’s disease psychosis.
• Details on the design of the Phase 2 ACP-211 study for major depressive disorder.

Readers are encouraged to refer to the full transcript for a detailed analysis of Acadia Pharmaceuticals’ strategic pipeline developments.

Full transcript - TD Cowen’s 5th Annual Novel Mechanisms in Neuropsychiatry & Epilepsy Summit:

Ritu Baral, Analyst, TD Cowen: Good afternoon, everyone. Thank you for joining us for the afternoon home stretch of the TD Cowen New Mechanisms in Neuropsychiatry and Epilepsy Virtual Conference and the Acadia Fireside Chat. I am covering analyst Ritu Baral. I am joined by my colleague Athena Chint, who works with me on Acadia. From the team, we have Dr. Elizabeth H.Z. Thompson, Executive Vice President and Head of R&D. Liz, thank you for joining us today.

Elizabeth H.Z. Thompson, Executive Vice President and Head of R&D, Acadia Pharmaceuticals: Thank you. I’m delighted to be here with you in the home stretch.

Lots going on. Definitely one of the bigger, more active names during this conference. Let’s jump into 101, your Prader-Willi program, with the pretty much imminent readout of your top line phase 3 COMPASS data that’s due early Q4. We’ve got a preview out finally, and we have suggested it might be October data. Let’s first talk about the 101 mechanism in addressing Prader-Willi. Can you talk to the hypothesized mechanism of action on carbetocin and oxytocin on Prader-Willi behaviors overall? Then we’ll get into hyperphagia.

Absolutely. A lot of the data that comes from that is about research with oxytocin in particular. ACP-101 (carbetocin) is a modified version of oxytocin. It was modified both to impact the PK, to make it a longer-lived drug, as well as to modify its selectivity for the oxytocin receptor. I anticipate that that facet is going to be relevant later in our discussion. For now, in terms of what we know about oxytocin in general, we know that oxytocin is a potent anorexic hormone that’s involved in the normal satiety response. We also know that oxytocin has a role in emotional and social behaviors. We have some animal data that suggest, for example, low doses of oxytocin can reduce food intake and the propensity to initiate feeding. We also know that animals without either oxytocin or oxytocin receptor can develop obesity over time. There’s an animal model of Prader-Willi.

In that animal model, we see that replacement of oxytocin can actually rescue the hyperphagic phenotype, this unrelenting need to eat. In individuals with Prader-Willi syndrome (PWS), we know that they have significantly decreased numbers of oxytocin-secreting neurons. Taken together, there are a number of lines of evidence that suggest the relevance of oxytocin and therefore ACP-101 (carbetocin) in Prader-Willi and in particular in hyperphagia.

Can you describe the prior oxytocin research? This is something that was brought up to us in our KOL conversations and folks who have been involved in prior studies, noting that they on a very high level indicated that they definitely have seen benefit in patients in those studies.

Absolutely. I talked about some of the preclinical work with oxytocin, but there’s also been, as you note, a number of clinical studies done with oxytocin. What I can say that’s fairly consistent across the board is very supportive safety and tolerability data. There’s something like five clinical trials out there. Across the board, I don’t think there were any serious adverse events that occurred in any of these studies during controlled periods. That said, efficacy results have been mixed. There have been some studies that have been supportive. There have been some that have been less so. What we have seen is that higher doses of oxytocin have been associated with temper tantrums, and there is at least the possibility that some of these off-target effects of oxytocin may limit the therapeutic window, and that may be part of what has contributed to an overall complex literature.

How do you view the direct line between what you’ve described and hyperphagia, direct or sort of indirect, between just those networks?

Yeah, it’s a really interesting question. You know, speculatively speaking, I would say that.

Oh, Elizabeth, as you go through that, Liz, if you could tie that into the potential inverse dose response that was seen with Neil and Reddy, I think.

I’ll try to check off on both of those. I mean, I think that it is plausible to say that we could be talking about a direct impact on hyperphagia here based on the types of mechanistic data that I kind of went through already about impact on satiety, impact in animals, in particular about being able to induce or sort of suppress these hyperphagic behaviors. I think there’s some good evidence there that suggests that it could be direct. I do feel obligated to note that the endpoint that we use in clinical trials here, it’s something called the HQ-CT, and it measures behaviors that are associated with hyperphagia. It measures things like, you know, the number of times that a patient is rummaging through the garbage to find something to eat or the degree to which they get upset when they’re denied food.

What we are measuring is in itself a little bit indirect, but I do think that the potential impact of ACP-101 on hyperphagia has the potential to be a direct one. To your point about the inverse dose response, I’m pretty sure there’s no one in your audience who isn’t aware of this, but just to level set really quickly.

Yep.

There was a prior phase 3 trial that was run with ACP-101 in Prader-Willi syndrome, and that study was about 140 patients. It was across three arms, two active and one placebo. The way this study was set up was that the primary comparison was between the 9.6 mg dose, which was the higher dose that was used here, and placebo. That didn’t meet statistical significance. Any comparison after that’s just a nominal one. I should note that the p-value for the comparison between the 3.2 mg dose, the lower dose, and placebo on the HQ-CT was 0.016. Had that been the primary endpoint, that would have been a supportive positive, statistical significance. Overall, this looks like an apparent inverse dose response. I mentioned earlier some of the challenges in the oxytocin literature.

What we know is that oxytocin has roughly equivalent affinity for the oxytocin receptor and also for vasopressin receptors.

Yeah.

This is actually hypothesized to be the reason for some of the challenges seen in the prior clinical studies. As I mentioned, or rather, carbetocin was engineered, not by us, by prior owners, for more selectivity for the oxytocin receptor. That said, it does maintain some vasopressin receptor activity at higher doses. That does have the potential to mute the efficacy based on overlapping nature of behaviors. That’s our hypothesis as to why you might see an inverse dose response and what might have explained the behavior of the 9.6 milligram dose in the prior trial.

Got it. The historical safety and tolerability, I mean, we talked about some of that. It almost sounds like irritability at potential higher doses. One of the things that has come up in our KOL conversations is the fact that it’s TID intranasal. I guess it’s from a more practical aspect, just sort of nasal tolerability and compliance, that sort of thing.

Yeah. A few things. I talked about the oxytocin historical profile, which has been pretty supportive from a safety efficacy perspective. In terms of 101 in particular, again, I’d say that that profile has looked pretty supportive to date. Obviously, I’m talking about previously run studies, not commenting on the currently blinded ongoing study that we have. As a few data points to support this, in that prior phase 3 trial, there were no study drug discontinuations during the placebo-controlled period for the lower dose, at a relatively low rate even for the higher dose. It was about seven patients, I think, overall. There were no serious adverse events during that placebo-controlled period. No AEs of edema, only one transient event of hyperglycemia. Overall, a fairly supportive profile.

To your point about the specifics of the intranasal dosing, with the 9.6 mg dose, we did see, and I forget the exact numbers, but it was less than 10% of patients who did have some nosebleeds. It was a much lower rate in the 3.2 mg dosing arm. Overall, we did see good compliance with dosing, suggesting that people are, you know, families are willing to and supportive of this administration and this administration profile, presumably particularly if they’re seeing results.

Got it. Can you walk us through the COMPASS design? You already, we already spoke about the HQ-CT, but can you spend a little time on these secondary endpoints? You know, this is, I think I haven’t spent this much time on exploratory endpoints of a phase 3 trial in a long time. I’d love for you to touch on some of those.

Absolutely.

You know which one of which I speak.

I believe I do. In overall study design, 12-week primary endpoint, we did take one of the big learnings of the last study that we needed a more robust sample size to really give the drug the best possible chance to show what it could do. Our target enrollment was 170 patients, randomized one-to-one, and we’re focusing in, of course, on that 3.2 milligram dose from the last study that appeared more supportive. So 3.2 milligram versus placebo. As noted, HQ-CT is the primary. From a secondary endpoint perspective, we’re looking at some clinician global impressions, both of severity and change for Prader-Willi, as well as for hyperphagia severity. There are a number of ways that you can look at the HQ-CT. We’ve picked one responder bar, which is at least an eight-point improvement from baseline.

In the prior trial, about a third of patients saw at least an eight-point improvement from baseline. That’s one of our other secondaries. We are looking at exploratories, assessing a number of other aspects of disease or things that we think could be important, including things like caregiver burden and healthcare resource utilization, and I imagine anxiousness, which is what you’re referring to. The PAD-Q is one of the things that we are looking from an exploratory perspective.

Can you talk a little more about the history of the PAD-Q and where, you know, how the FDA may, how the community looks at it? It seems to be very novel, but very useful.

Yeah, so the PAD-Q is looking specifically at sort of anxiety and some of these associated behaviors in PWS in Prader-Willi patients. There were some interesting data on that in the prior phase 3 trial suggesting that, 101 might have some impact there, which, you know, certainly would be great if that were to be something that was replicated. In terms of importance of that endpoint and other endpoints to FDA, to stakeholders, to patients and caregivers, to KOLs, I guess I’ll start with FDA and say, you know, as far as FDA is concerned, I think probably our most objective measure of what matters to them primarily is their recent behavior.

If we look at the VICAT approval, the only efficacy endpoint that they’re noting in the label is the HQ-CT, which it’s fair to interpret as an indication that FDA considers this to be sufficient information for prescribers overall. I think across some of those other stakeholder groups that you talked about, I think there’s interest in various other endpoints, and probably in particular in the HQ-CT responder analysis and the anxiousness in PWS. Our overall conversations from KOLs and caregivers is that if the primary endpoint results, just the HQ-CT change from baseline, are similar to what we saw with the 3.2 milligram dose in the prior study, that in and of itself would represent an important treatment option from their perspective.

That echoes what our conversations with the stakeholders have been—KOLs, parents, patient groups, et cetera. We sort of spoke to the gamut. The question is, you know, if it doesn’t work on hyperphagia by statistics alone, do you think at this point that there is a path to approval on clinical benefit on other aspects of the disease, understanding that based on our conversations and our preview, you would have landscape support depending on the magnitude.

I was just going to say, you know, this is going to feel like the most mealy mouth answer I’ve ever given you, but it’s going to be so dependent on the nature of the data. I do anticipate, you know, I think I have heard pretty consistently, good support from the community. It’s wonderful that they have an option that’s available to them, but I think they’re also very interested in additional options. I think that were we to have, you know, a totality of data that was supportive of benefit, even missing a missing statistical significance, I think we’d be interested in figuring out whether there was a viable path forward.

Got it. For folks listening in, if you guys have any other questions, please send them through to email ritu.baral@cowen.com or ritu.baral@tdsecurities.com. We’ll get them worked into the conversation. We’re going to move on to 204, but we will loop back if we have other questions over email for Liz. Let’s move to 204. I would say at least a third to maybe half of my conversations have been around flagging this upcoming data. Can you walk us through the, first of all, it’s your next gen 5HT2A inverse agonist. It’s next gen in place of, can you walk us through the ongoing phase 2 ADP study design, and how that’s going, enrollment, conduct, et cetera?

Absolutely. As you note, ACP-204 is our new 5HT2A inverse agonist. We really designed this to build upon learnings from NUPLAZID, both from a molecule perspective as well as a program design perspective. Really briefly, we were looking to minimize or eliminate a QT prolongation, give ourselves the ability to dose range and maybe get to greater levels of efficacy based on higher exposures, and also get to faster onset of action, which hopefully one would anticipate would translate into faster onset of clinical benefit. We also took our pretty substantial experience with NUPLAZID across a number of different indications to help us select where we thought it might be most beneficial to focus in on development. We’re looking primarily into Alzheimer’s disease psychosis and Lewy body, which I know we’ll come to. For Alzheimer’s disease psychosis, we have a currently ongoing phase 2 and phase 3 program.

This was designed to be, what I’ll call, operationally seamless but statistically significant or statistically separate. Essentially designed such that once enrollment is complete in the phase 2 portion, sites can start enrolling in the phase 3, gives us the opportunity to save some time, but statistically separate. We have the opportunity to analyze, evaluate, and report out information from the phase 2 component by itself, and if need be, make some modifications to the phase 3. High-level study design for the phase 2 portion is our target enrollment is about 318 patients spread across three arms, placebo and two active arms. We’ve got a week six primary endpoint of the SAPS-H and D. We are evaluating other measures to fully understand the totality of impact in the Alzheimer’s disease psychosis patient population. Broadly speaking, in terms of how the trials continue, it is going well.

At this point, we’re not providing additional granularity beyond noting that we continue to anticipate data for the phase 2 portion in the middle of 2026.

Any additional clarity on that timing that you can offer at this time?

That’s it for right now. I look forward to being able to be a little bit more specific as we get a little bit closer.

Understood. Can you speak more to the powering of the study and what, in the phase for the phase 2 portion, what will you disclose with top line data beyond the primary endpoint effect size?

Yeah, so in terms of powering, broadly speaking, how we went at it for this particular trial is that we powered for an effect size, rather than a delta. The effect size that we’re powering for in this study is 0.4, which is roughly a medium effect size. In terms of the specifics of top line data, I’ll be honest, we’ll be making those decisions a little bit closer to the time of readout. At this point, I would anticipate it would focus on primary, but we’ll see what makes sense at the time. Broadly speaking, there’s what we were looking for in that particular study. I guess what I just want to briefly comment on is sort of what we’re hoping for in the Alzheimer’s disease psychosis program as a whole.

With each step along the way, we’re looking for evidence that suggests that ACP-204 is consistent with that profile. Of course, we’re looking for efficacy. This is at this point, both Alzheimer’s disease psychosis and Lewy body dementia psychosis are areas where there are not specific treatment options, and this is a huge unmet need. These are complex patients with high needs. We’re looking for things that are easy and convenient to dose for them in terms of once a day dosing, which seems consistent with our profile, being able to dose with and without food, that kind of thing. We’re looking on the safety side to make sure that we are avoiding sedation, negative impact on cognition and motor function. We think these are all going to be important things in this patient population.

You’re starting your Lewy body study soon with ACP-204, and that’s going to enroll both LBD and PDP. Can you review for us that design and the potential top line timing effective as well?

Absolutely. I’ll start by saying that this is probably one of the first programs that I actually initiated once I got to Acadia Pharmaceuticals. While the data are, well, we don’t have a lot of data in terms of numbers of patients, I was really heartened by the strength of the data that we appeared to be seeing in the Lewy body dementia, the LBDP subpopulation. In one of our prior trials, the Harmony trial, which is a randomized withdrawal, we saw only about 5% of patients on NUPLAZID relapsed, whereas 55% of those who went on to placebo did. It was only in about 20 patients per arm, but I thought this was sort of intriguing and worth following up on. The study we’re doing now is not a randomized withdrawal. It’s a prospective parallel group study, target enrollment of 180. It’s across three arms.

The primary endpoint here is going to be the SAPS-LBDP, and that’s going to be at a week six time point. We will be enrolling both dementia with Lewy bodies and Parkinson’s disease dementia psychosis patients. These fall under the broad umbrella of LBDP. We are looking to make sure that we’ve got informative numbers of both of those subpopulations, and we put a cap on the PDP population of 50%, with the intent, again, of getting informative numbers in both dementia with Lewy bodies and PDDP.

Got it. How are you using the biomarkers and genetic assays to confirm the diagnosis of those, of LBD, PDP, as well as the earlier ADP, the trial that we discussed earlier, ADP?

Yeah, thank you. I realized I glossed right over it with ADP. Thank you for the opportunity to kind of go back to that. You know, the science and understanding of ADP has sort of evolved over the years. At this point, confirming by biomarker is an important part of the clinical diagnosis. We are requiring in our ADP trial that everybody is confirmed by biomarker, either historical or a plasma biomarker in the work.

That wasn’t done in the prior Harmony study, and prior PDP historical Acadia studies?

We did not require the same level of biomarker confirmation in historical trials. I mean, this has been an evolving space. The biomarker requirement kind of went from a research use to really expectation within the clinical diagnosis. Knock on wood, I am really optimistic that by going this extra step to confirm this patient population in ADP, should we be so fortunate as to get all the way through with a positive program, we’ll be able to very convincingly demonstrate that we have shown the efficacy of ACP-204 in an Alzheimer’s disease population that we’ve really clearly identified. I think that’s going to be an important component of regulatory success eventually, assuming that we have technical success to go along with it. In the Lewy body trial, the science hasn’t evolved quite as far as ADP. Biomarkers are not really part of the definitive diagnosis at this point.

We’re including them from a learning perspective, but we’re not requiring them at entry. We’re looking at alpha-synuclein, and that is the primary component of Lewy bodies. We do anticipate that we are going to see alpha-synuclein positivity in a reasonable number of these patients. We’re looking at a number of genetic assays as well, where there’s some information in the literature suggesting that they are associated with the presentation of the disease. We think this is going to be interesting from a learning perspective, but at this point, we’re not requiring any of those to get into the study. We’re just going to look at the impact at the end.

Understood. Athena, if you want to take ACP-211.

Yes, moving on to 211, you are a noradrenaline compound in development for MDD with an adequate response. You have a phase 2 starting in Q4 of this year. Can you briefly talk about how 211 differs from IV ketamine, Spravato, and what the safety kind of looks like as well?

Sure. At the most basic, the molecule level, ketamine is a racemic mixture of the R and S isomers of ketamine. Spravato is the S isomer of ketamine, and ACP-211 is a selectively deuterated form of R-norketamine, which is, norketamine is a metabolite of ketamine, so molecularly distinct. At an administrative level, IV ketamine is IV. Spravato is inhaled, and ACP-211 is going to be oral. Based on data to date, we see the potential that 211 may be able to provide antidepressant efficacy with no sedation and minimal dissociation. We think this would be distinct from the profiles clinically demonstrated of both ketamine and Spravato. I do want to be clear, ACP-211 is currently early stage, and our understanding of that safety profile is informed by preclinical experiments and phase 1 data in healthy volunteers, so not nearly as robust of a safety understanding as Spravato has, for example.

That said, to date, we do have animal models suggesting that there’s similar efficacy with 211 as is seen with ketamine, that there is, again, in animal models, less motor impairment than is seen with ketamine. In healthy volunteers, we currently have seen no sedation and only low level and transient dissociation in healthy volunteers, and even that low level and transient was seen only at the highest doses. Thus far, a profile that seems supportive of our goal for this molecule, which is ketamine-like efficacy with a patient experience that doesn’t, that has no sedation, has minimal dissociation, and hopefully therefore would require less significant monitoring.

The dissociation that was seen with the highest dose, how long was it?

It was not that long in duration, and it also was of the level of, you know, people seeing their surroundings more clearly. It was really very minimal in terms of its impact.

What is your phase 2 study design?

The phase 2 study design is, we are looking at starting this up towards the end part of this year. We’ll have about 150 patients in it who are randomized one to one to one, a high dose and a low dose. We haven’t disclosed what those doses are at this point versus placebo. The primary endpoint would be early. It would be week 4, and we’re focusing in on the moderates here. We are looking in MDD patients who have had an inadequate response to prior antidepressant treatment.

What do you hope to see in that moderate response in the top line?

We haven’t given exact details of what that delta we would want to see, but in general, what we’re looking here for is robust efficacy with a rapid onset, sort of similar to what you would expect out of a ketamine-like or a Spravato-like profile. I think the thing that’s important to note here is that we certainly designed our phase 2 to look at efficacy, but we also designed it specifically to rule out sort of unacceptable levels of either sedation or dissociation. This is intended to help us inform on both aspects of that target product profile.

Got it. Moving on to the rest of your pipeline, during your R&D day in June, you mentioned your business development strategy will focus on neuropsychiatry for the larger indications and also rare diseases. Can you provide additional color on your strategy? Are you pursuing earlier or later stage clinical assets? How are you balancing that development risk when evaluating the different agents?

Yeah, so, you know, I think we are broadly interested, particularly in rare diseases, where we really have opened up our aperture significantly in terms of what we’re willing to consider. Historically, we’ve really looked in at neuro rare, which has been kind of our sweet spot. It makes a lot of sense. We’ve built substantial internal expertise, both commercially and from an R&D perspective in rare disease, and we want the opportunity to deploy that a little bit more broadly. That said, we are interested in probably phase 2-ish assets, which is really our sweet spot right now. Obviously, we’d be willing to go one direction or the other depending on the profile of the individual agent. That’s probably what’s going to be the main target for us at this point.

Understood. Before we wrap, can you walk us through any additional pipeline catalysts that we should expect for the remainder of this year?

In terms of actual data coming out, the main thing to keep your eye out for, of course, is that ACP-101 data that’s going to be coming out in early Q4. We are starting a number of trials that I think are going to be interesting over time. We’ve talked about ACP-211. We have an early-stage program called ACP-271, and targeted dyskinesia and Huntington’s disease are the target indications. We’ll be taking that into phase one starting by the end of the year, getting that Lewy body study up and running as we talked about. We’re also looking at expanding into Japan. We’re going to be starting our triphenotide trial in Japan in the coming weeks as well. A number of good starts are going to be setting up our catalysts for future years.

Awesome. With that, we are at time. Thank you so much, Elizabeth, for joining us today, and thank you everybody for tuning in.

Thank you. Thanks, everyone.

This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.