ACADIA Pharmaceuticals Inc. (ACAD) Earnings Call Transcript

Good morning, ladies and gentlemen, and welcome to the ACADIA Pharmaceuticals conference call to discuss the RADIANT Phase II top line results evaluating remlifanserin for the treatment of Alzheimer's disease psychosis. My name is Krista, and I will be your coordinator for today. I would now like to turn the presentation over to Albert Kildani, Senior Vice President of Investor Relations and Corporate Communications. Please go ahead.

Thank you, Krista. Good morning, and thank you for joining us on today's call to discuss the top line results from the Phase II portion of our ongoing RADIANT clinical trial program evaluating remlfanserin for the treatment of hallucinations and delusions associated with Alzheimer's disease psychosis, or ADP. On today's call, Catherine Owen Adams, our Chief Executive Officer, will provide opening remarks. Following Catherine, Liz Thompson, PhD Executive Vice President and Head of Research and Development, will review the study design and discuss the Phase II top line efficacy, safety and tolerability results from RADIANT. Catherine will then provide closing remarks before we open the call for your questions. I would also like to point out that we are using supplementary slides, which are available in the Events and Presentations section of our website. These factors and other risks associated with our business can be found in our filings with the SEC. You are cautioned not to place undue reliance on these forward-looking statements, which are made only as of today's date. I'll now turn the call over to Catherine Owen Adams.

Thank you, Albert. Good morning, everybody. Thank you very much for joining us today. We are very pleased to share the top line results from our Phase II portion of the RADIANT clinical trial program evaluating remavanserin for the treatment of Alzheimer's disease psychosis, our Phase III enabling. Importantly, these Phase II results demonstrated meaningful and consistent efficacy trends across endpoints. The 60-milligram dose narrowly missed statistical significance on the primary endpoint of change from baseline in SAPS-HMD total score at week 6. and demonstrated nominal statistical significance on the key secondary endpoint of change from baseline in CGI-S-ADP. Remlfanserin across both doses also demonstrated a favorable safety and tolerability profile with no new safety signals identified. Taken together, we are excited that the efficacy and safety results support the further development of remlavanserin in the 2 Phase III ADP trials currently underway. I'll now turn the call over to Liz to review the Phase II study design and results in greater detail, after which I will return with some closing remarks. Liz?

Thank you, Catherine. I'd like to begin by thanking the patients, caregivers, investigators, study coordinators and of course, the dedicated internal ACADIA team whose dedication made this study possible. RADIANT is a global, randomized, double-blind, placebo-controlled, operationally seamless Phase II and Phase III program evaluating remlfanserin in Alzheimer's disease psychosis. The Phase II portion of the study was designed to evaluate the efficacy and safety of 2 doses, 30 milligrams and 60 milligrams given once daily and to inform the dose and patient population for the Phase III program. The results provide clear and important information to support the continued development of remlfanserin. -- including the selection of dose and considerations on how we could refine the patient population for Phase III. In RADIANT, patients were randomized in a 1:1:1 ratio to receive 60 milligrams of remlavanserin, 30 milligrams of remlfanserin or placebo once daily for 6 weeks. The primary endpoint was change from baseline in the SAPS-H&D total score at week 6. And the key secondary endpoint was the change from baseline in a clinician global assessment of severity called the CGI-S-ADP at week 6. Our prespecified population for our efficacy analysis was the modified full analysis set, which is referred to in these slides as the MFAS. This comprised 326 patients who received at least 1 dose of study drug, had a baseline SAPS H&D total score of at least 10 and had at least 1 post-baseline SAPH&D assessment. In the MAS population, 109 patients were assigned to the 60-milligram dose, 110 to the 30-milligram dose and 107 to placebo. Demographic and baseline characteristics were generally well balanced across treatment groups. The mean age was about 75 years old and approximately 2/3 of patients were female. Baseline SAPS-H&D and CGI-S-ADT scores were also generally similar across the treatment groups, supporting the interpretability of the results. Going to the next slide. Turning to the efficacy results. The 60-milligram dose of remlfanterin narrowly exceeded the prespecified threshold for statistical significance on the primary endpoint. At week 6, the 60-milligram dose demonstrated an effect size of 0.26 on the SAPS-H&D total score with p-value of 0.0603. For the key secondary endpoint, the effect size was 0.37 with a nominal p-value of 0.0077. This corresponded to reductions from baseline of 12.4 versus 10.4 for SAPS-H&D and 1.3 versus 0.9 for CGIS. A dose effect was observed as the 30-milligram dose showed minimal improvement relative to placebo with effect sizes of 0.05 and 0.11 across the primary and key secondary endpoints. Now as we have discussed this topic previously, I want to make a quick note about the NPIC. As you will recall, we added this as an exploratory endpoint and collected it only in patients who were enrolled later in the study. So we are early in our analysis of this endpoint. But to date, it appears that a small numerical improvement in NPIC was also seen in the 60-milligram arm. Now returning to SAP-HMDS. On this slide, we show the performance over time, where the data suggests increasing separation through week 6. In totality, the efficacy data, combined with safety data that support either dose, which I will review later in the presentation, gave us confidence in selecting the 60-milligram once-daily dose of remlfanserin for further development. The next question we turned our minds to was whether there were any modifications to the inclusion criteria that hold the potential to improve Phase III outcomes. We used prespecified subgroups to guide us regarding the potential impact on both efficacy and patient eligibility of several criteria modifications. One promising area with the potential to further improve efficacy in Phase III while maintaining the significant majority of the trial population is further refining the enrollment criteria for modestly higher baseline psychosis. On this slide, you can see the impact of one potential version of this refinement when applied to the Phase II trial. First, 80% of patients in the trial would still have qualified for participation. Second, the effect sizes in the resulting population increased for both the primary and key secondary endpoint. For the primary endpoint, the resulting effect size was 0.33 and for the key secondary, it was 0.43. Interestingly, on the later enrolled patients where we collected NPICHND data, there was a similar suggestion of improved effect in this population. And again, I must emphasize that the data are limited based on the timing of adding the endpoint to the study. We are considering whether this refinement to modestly higher baseline psychosis or similar refinements would be beneficial to the overall profile of remlfanserin. Turning now to safety and tolerability. The overall safety profile observed in this study was favorable and supported continued development of remlfanserin. Now recall that one of our main goals with remlfanserin was to avoid the QT prolongation signal that had prevented our evaluating higher doses of [ pimavanserin ]. Preclinical and Phase I data had supported our desired profile for remlfanserin, and we were pleased to see that in this data set, there was no signal of QT prolongation compared with the placebo group. Expanding to the safety profile more broadly, rates of adverse events, serious adverse events and discontinuations due to adverse events were all seen at similar rates to placebo. There were no deaths in either remlfanserin treatment group compared with 2 deaths in the placebo arm. Rates of individual adverse events were also low. In fact, no individual adverse event was reported in more than 5% of patients receiving 60 milligrams once daily remlfanserin. Somnolence at 3.2% and nausea at 2.4% were the only adverse events reported in more than 2% of patients receiving 60 milligrams that occurred at a numerically higher rate than placebo. Finally, I've talked previously about our target profile for remlfanserin and what aspects could yield an important treatment option for patients. I was pleased to see that this data set suggested no negative impact on motor function or motor symptoms or cognition. Taken together, the safety and tolerability findings were consistent with our hopes for the molecule and support continued evaluation of remlifanserin in the ongoing Phase III studies. Now looking ahead, we plan to present additional efficacy and safety results from the Phase II study at the Clinical Trials on Alzheimer's Disease or CTAD conference, which takes place November 16 through 19 in Boston. The Phase III program continues while we implement protocol amendments, including removing the 30-milligram dosage arm. And with that, I turn the call back to Catherine.

Thanks so much, Liz. So with these results providing us valuable information needed to continue the development of remlfanserin, I'd like to put the opportunity ahead in context, beginning with the significant unmet need it may address if successfully developed and approved. Alzheimer's disease psychosis is characterized by hallucinations and delusions occurring in patients with Alzheimer's disease. According to the Alzheimer's Association, more than 7 million people in the U.S. are currently living with Alzheimer's disease, and approximately 30% of these patients are estimated to experience psychosis, commonly consisting of hallucinations and delusions. These symptoms can be frequent, severe and highly distressing for both patients and caregivers. They're also associated with worse clinical outcomes, including increased likelihood of institutionalization, greater disease severity and increased morbidity and mortality. Despite the substantial burden of disease, there are no currently FDA-approved therapies specifically indicated for hallucinations and delusions associated with Alzheimer's disease psychosis. We believe there remains a significant and important need for new treatment options, and today's findings reinforce our confidence in remlfanserin's potential to address this underserved patient population. Additionally, and importantly, we also continue evaluating remlifanserin in an ongoing Phase II study in Lewy body dementia psychosis. Lewy body dementia is a progressive brain disorder affecting thinking, movement, mood and behavior. More than 1 million people in the U.S. may be living with LBD and an estimated 50% to 75% experience psychosis. There are no therapies approved for Lewy body dementia psychosis, underscoring the significant need for new treatment options. Approximately 200,000 patients with Lewy body dementia psychosis are currently treated with antipsychotics, further illustrating the need for an effective and well-tolerated therapy, specifically developed for this population. Looking ahead, we expect our existing products to generate meaningful near-term growth with projected global net sales of approximately $1.7 billion through 2028. Over the long term, we have several attractive pipeline assets with strong commercial potential in areas of significant unmet need. Collectively, we estimate these programs could represent approximately $11 billion in unadjusted global peak sales potential, including $4 billion for remnifanserin across both Alzheimer's disease psychosis and Lewy body dementia psychosis. While these estimates are not financial guidance and drug development carries inherent risk, they illustrate the substantial opportunity we see across both our existing portfolio and pipeline. More broadly, remnifanserin is part of a robust pipeline with multiple ongoing and planned programs across neurological and rare diseases. We expect to announce a further 3 Phase II or Phase III study readouts through 2027 with additional development and regulatory milestones across the pipeline. Today's RADIANT Phase II results reflect Acadia's continued commitment to developing innovative therapies for patients with serious neurological conditions where effective treatment options remain limited. I would like once again to thank the patients, caregivers, investigators and study personnel who participated in the study as well as our employees and collaborators whose hard work and dedication has made today's milestone possible. We look forward to presenting additional Phase II findings at our CTAD meeting in November and providing updates on our continued progress. And with that, operator, we're now ready to open the call for questions.

[Operator Instructions] And your first question comes from Ritu Baral with TD Cowen.

So many questions. We'll start with the NPI. I think, Liz, you mentioned that you're seeing something. Are you seeing an effect size similar to the CGI and the SAPS HD? And have you sort of done the correlation analysis that we've spoken about that would really validate the SAPS-HD. I guess I'm wondering, is the SAPS-HD in your mind, still the appropriate endpoint? And do you have to talk to FDA before making any changes to the F3 -- I'm sorry, the Phase III around powering?

Excellent questions, Ritu. A couple of thoughts. First, I do want to say these are top line results. We've had these data for a very short period of time. So we've done initial analyses. There is more that we have yet to do, but there are some comments that I can make for now. The first of these, as we have discussed before, One of the things with the NPIC was we did move it late. And so there is a relatively limited proportion of the population that actually do have these data available. If you look at the overall population there, again, going back to my remarks earlier, it is a relatively small numerical improvement that you see with the 60-milligram arm. When you get into the enriched population that I showed an example of what it could look like, you do see an effect size that is more similar to the effect size that you see with the SAS-HMV. But again, this is very preliminary at this point. We absolutely have not, at this point, done the kind of validation analyses that we would need to do to support primary endpoint dossier. That is we will do now that we have the Phase II data in our hot little hands, but that is part of our intent. In terms of talking with FDA, we don't believe that we need to, for example, discuss with them to get their conclusion on dropping the 30-milligram dose arm. I think that that's very clear and the data supported, and we are planning to go forward with that as quickly as possible. Were we to make a change to the primary endpoint in the Phase III eventually, I would want to discuss that with FDA. At this point, we're not proposing that.

Your next question comes from the line of Yigal Nochomovitz with Citigroup.

I had a few on basically related to Slide 11. So first of all, you mentioned, Liz, that this is one of the exploratory analysis you did. Is this the most likely course of action with regard to a change to the Phase II population? Or are there other slices that you might enumerate that would be of interest with regard to enriching for a higher responding group? Secondly, could you comment at all with respect to whether the efficacy was more enriched on the H hallucination versus the delusion with regard to the primary endpoint. Just wondering how that shook out? And then my last question is, regarding the 30 milligram, I know it didn't hit, but I'm just curious if you did the analysis on Slide 11 for the 30 to potentially show that there may have been a little bit of a separation there if you enriched for the high baseline psychosis even at the 30, which would give you more confidence in that subgroup.

Okay. Quite a few things there. Hopefully, I got them all jotted down. So we are evaluating various possible options of what we might do in terms of minor modifications to the patient population. It's always a trade-off in terms of the proportion of your population that remains relevant and the potential bump in efficacy that you could see. I will say that I think that focusing in on a little bit more higher baseline psychosis severity, I think, is a promising area. And that is where I am most enthused, but we are thinking about what the appropriate way of doing that would be. But I do think Slide 11, we gave that as an example because we think that this is a strong contender that really does balance the -- keeping the broadest possible eligible patient population for the trials while also somewhat increasing potentially the efficacy that we might expect or at least would have seen in Phase II with that application. We are keeping high level at this point in terms of data. So I'm not really going to comment on the breakdown between hallucinations and delusions. But I will say that across endpoints, we did see impact on both. And in terms of we did also look at the 30-milligram dose Slide 11 type analysis as well as other versions of that. And we did see some increase, though it is consistently not in the realm of the 60-milligram arm. So again, we feel very confident that 60-milligram is the right dosage arm to take forward.

Your next question comes from the line of Tessa Romero with JPMorgan.

Catherine, Liz, as you saw these results and thought about the pushes and pulls of moving remlfanserin forward, what ultimately made you decide that moving this program forward was worth the investment for ACADIA relative to all of the other initiatives at the company? And then as a follow-up, what additional analyses are you working through that we can expect to see at CPAD here in a couple of months that may be supporting your view?

Thanks, Tess. I'll start and give Liz a quick break. In terms of balancing investment choices at ACADIA right now, we're in a fortunate position that we don't have to make those trade-offs specifically for any of our programs. We have, as you know, a strong balance sheet and a strong investment focus on our R&D portfolio, we were very encouraged by the results we've seen. We are continuing to look at how to enrich the population for further enhancement through Phase III. And we believe with the high unmet medical need in this population, the efficacy that we've seen so far that Liz has described, and we will continue to describe as we move forward in the next few months as well as the safety and tolerability profile that this molecule offers the potential for patients with Alzheimer's disease psychosis, and it is absolutely worth our investment to move this forward into the Phase III trial.

Absolutely echo all of that. It is good to be in a position where what we -- the choices we are making is, is this worth taking forward versus which thing can't we take forward. So we're pleased to be in that position, and we feel good about the fact that this is a program that has the potential to meet a real, real meaningful unmet need for patients. In terms of CTAD, we're having only gotten this data very recently, we're very early in putting together what our actual presentation for CTAD would look like. And obviously, we're going to have to work with investigators on what the most appropriate things to put forward are. But certainly, you should anticipate that you're going to see some additional endpoints beyond this, some additional description of safety. And if there are things that we think are especially meaningful in terms of ways that we could refine the patient population, we likely would address that as well.

Your next question comes from the line of Sumant Kulkarni with Canaccord Genuity.

Given the safety you've seen so far, could a dose higher than 60 mg be added? And is there any merit in considering later time points beyond week 6? And if you look at dementia with Lewy bodies, that tends to have higher rates of psychosis versus Alzheimer's disease psychosis. So do you expect any different types of behavior of this product in that patient population as you consider the Phase II trial there?

I'm not sure that I grabbed all of those. So I'll comment on the ones that definitely stuck with me and then somebody around the table will prompt me for what I forgot. So I'll say, certainly, it's an interesting question. We really did feel like we had -- we were targeting an exposure that was going to be associated with near maximal efficacy. I will say that one of the things that we'll be continuing to think about is exposure response analysis to see whether there's a reason to believe that going even higher might be helpful. All that said, we think 60 milligrams represents a based on the profile seen to date, recognizing we would -- we have to do the Phase III trials and are, we think that 60 milligrams represents a profile that is potentially an approvable medication if the Phase III were to look like we are thinking that it could based on this information that we have about how to refine the Phase III program. So I think we don't believe we have to go to a higher dose, and we're looking for the exposure response to see if there could be any benefit to this. But for now, we really do think 60 milligrams is an important treatment option. As far as a longer time point, patients do roll into an open-label into open label here. Obviously, there are always that you need to think about in terms of comparing open-label to double-blinded data. We think that week 6 gives us a good demonstration of the efficacy. The shape of the curve, particularly if you look at SAPSHD, does suggest that maybe you could get a little bit more if you look a little bit further out, but we want to focus in on week 6 as a place that can demonstrate the efficacy and where we have an understanding of both how our drug works as well as the placebo arm. So we do anticipate maintaining the week 6 time point as our primary endpoint for the Phase III program. As far as DLB, we continue to be -- or Lewy body, we continue to be very enthused about the potential profile there. I have always thought that the efficacy, while in limited numbers of patients with pimavanserin was striking. So this -- suffice it to say nothing about these data have decreased our enthusiasm about that. So we really are looking forward to those results. We do have to run the study in Lewy body because exactly as you note, it is a distinct patient population that can have distinct reactions to medication and sort of a distinct profile in terms of how sensitive they are to safety. So we look forward to understanding the application of remlfanserin in Lewy body, but we think this is certainly supportive of that and feels good. So I think that was all the questions.

Well done.

Your next question comes from the line of Ash Verma with UBS.

I wanted to understand the enrichment for more severe patients that you've talked about. So I know you mentioned that this Phase II RADIANT is enriched already with 80% of these patients have modestly higher baseline psychosis. But you're also saying that you will try for modestly higher baseline in the Phase III. So what would that look like? And then just on the same topic, maybe like when you compare the psychosis level in this Phase II RADIANT, how did that compare to the prior study 19?

Thank you so much for giving me an opportunity to clarify something that I may not have said very clearly. So that example that we gave of an enriched population, when I was saying 80%, that was actually indicating that 80% of our currently existing population would fit under this new definition. So that's just to be clear on what that meant. So we do think that, obviously, sort of -- obviously, that does mean that there is a potential to nudge our population such that everybody fits within the future study. We are considering it. We have not landed on this as the absolute way that we're going to go forward, but that is what we are thinking of because it does appear that there's a possibility of increasing the efficacy that we could see out of it. In terms of comparisons with the prior study, what I will say, and again, you'll see more granular data on this, I think, at CTAD. But what I will say, generally speaking, is that we have indeed enriched our Phase II program compared with the 019, the Phase II study with pimavanserin to try to move patients -- try to move the general population to a somewhat more severe psychosis phenotype. And we were successful there. We had increased the severity here compared with the 019 study. That said, we do think that there could be some benefit potentially to increasing that just a little more, a modest increase there, and that is what we're focused on as one very promising potential area for refinement of the population.

Your next question comes from the line of Brian Abrahams with RBC Capital Markets.

This is Nevin on for Brian. Just wondering if you could talk a little bit more about what you think might be clinically meaningful delta for the SAP HD, just given that the 60 mg arm showed around a 2.2 point delta. How do you kind of interpret this in the context of the current regulatory landscape, particularly given kind of the high unmet need that we've seen?

Fantastic question. I think a lot of this, we do need to ground in the significant amount of unmet need that I think Catherine eloquently covered in the presentation. And given that, I think that we do believe that these results colloquially, we believe that these results are clinically meaningful and would represent -- could represent a clinically meaningful treatment option for patients. From a technical perspective, some of the analyses that we are going to be doing is demonstrating that for regulatory purposes of regulatory consideration in terms of what degree of change is clinically meaningful to meet a regulatory bar. But certainly, based on conversations running into this readout and very, very limited conversations since, we think that this represents a potentially clinically meaningful profile one, in the context of what thus far is a very supportive safety profile; and two, in a patient population with truly significant unmet need and at least at this moment, no approved treatment options.

Your next question comes from the line of Malcolm Hoffman with BMO Capital Markets.

I want to ask why you guys think that there wasn't as much activity in the 30-milligram arm. Do you think this is a receptor occupancy issue? Or what else could be potentially driving this? And then thinking about the Phase III, are you thinking about increasing the actual enrollment size for the 60-milligram arm, given that you're going to be dropping out the 30-milligram arm? I appreciate your thoughts here.

Thank you. So about the 30-milligram dose, I guess I'm going to take a step back and say a couple of things. We designed our Phase II program based on learnings from pimavanserin. And the good part is now we have the opportunity to design our Phase III based on learnings from remlavanserin. That said, we did always anticipate that increasing exposure was going to be helpful in this. And I think that the behavior -- the relative behaviors of the 30 and 60 milligram really just bear that out of the exposure -- increasing exposure here does seem to be important. I think we've talked before about the -- it is hard to draw a direct line to what we know about receptor occupancy because our studies, by and large, have been in healthy young male volunteers. And so that receptor occupancy information is maybe not as directly translatable. But certainly, data with PIM suggested that getting to higher exposures would be helpful. And while we have not yet been able to do the exposure response analysis on remlfanserin and certainly looking at the behavior of the 2 dosage arms does seem to suggest that exposure is important here as well. In terms of enrollment size, what I can say is that originally, we were anticipating we were going to need 3 arms. Now we think we're only going to need 2 arms. So that should be helpful from that perspective. We are going to need to -- we do need to consider what other modifications we might be making, including around the target population, and that could have impacts. But at this point, I'm not in a position to really say that.

Your next question comes from the line of Paul Matteis with Stifel.

This is Julian on for Paul. Would you guys mind stating like what is the actual threshold for this modestly higher baseline psychosis population that you're citing on Slide 11. Is that a specific SA H+D score? If you could share it, that would be really helpful. And then can you also confirm that this was prespecified? And were there any other prespecified analyses that you're planning on sharing today? And then just one other question is just like given effect sizes tend to regress in Phase II to Phase III in neuropsych, as you guys know, how do you balance in the Phase III powering this for success while still trying to achieve a result that's clinically meaningful?

Sorry, I was madly writing, and I'm 100% sure I missed one. So again, somebody will prompt. So we are not going to share right now what the specific threshold is that we're -- this example on Slide 11 is. And there are 2 reasons for that. Reason number one is this is an example. We did look at a number, not an infinite number. We're not trying to data dredge here, but we did look at a number of different potential subgroups to inform us. So that's reason number one. But reason number two is that we also do want to be careful about how -- about certain specifics that we put out publicly because it does have the potential to impact placebo response rates in future studies based on if people are trying to target a certain level that they are trying to achieve to get into a study. So we likely will continue to take the approach we've taken historically, which is that we are not overly specific around what our inclusion criteria is from a point of view of those measures that could impact our overall placebo response rate. We did have a number of -- a relatively limited number of prespecified subgroups that we've looked at. We are not planning to share any additional ones today other than -- we're not planning to share any additional subgroups today other than the one that we have already shared. And the -- there often is a change in effect size going from Phase II to Phase III. And that is one of the considerations that we are looking at here as we think about ways to increase the effect -- to potentially increase the effect size we might be able to see in Phase III so as to balance exactly what you say, the practicalities of making sure that we have something clinically meaningful and enrollable while also reflecting what we think the likely profile of the drug could be.

Your next question comes from the line of Jack Allen with Baird.

Sorry, apologies. I was just looking for the button for the present. Just a few from our end. On the receptor response analysis, I'm just curious, it sounds like that work is still ongoing. Could we get that data at CTAD? Or do you have any time line as it relates to when we could get that receptor response analysis? And then the other one we had briefly was on the Phase III and this modestly higher baseline psychosis potential modification. I know you're still considering whether to implement that, but any high-level thoughts as it relates to how that would impact trial time lines would be helpful as well.

Thanks, Jack. I'm not going to give specifics on when we expect the dose response analysis at this point. If we have something meaningful out of it that is by the time of CTAD, we will include it, but we may or may not have that by that time. In terms of the Phase III potential modification and whether that could have an impact on overall trial time lines, there's obviously kind of a complicated interplay here of the -- what that does to your potential effect size and what it does to the size of your potentially enrollable patient population. So right now, I'm not prepared to say what that could be. Again, I would say, if nothing else, it's probably useful to keep in mind the fact that certainly compared with a 3-arm study, we would be looking at a smaller study here comparatively speaking. But the exact impact is kind of going to depend on the specifics of which flavor of refinement we pursue.