ACADIA Pharmaceuticals Inc (ACAD) 2021 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the ACADIA Pharmaceuticals Third Quarter 2021 Financial Results Conference Call. My name is Gigi, and I'll be your coordinator for today.

I would now like to turn the presentation over to Mark Johnson, Vice President of Investor Relations at ACADIA. Please proceed.

Thank you very much. Good afternoon and thank you for joining us on today's call to discuss ACADIA's Third Quarter 2021 Financial Results. Joining me on the call today from ACADIA are Steve Davis, our Chief Executive Officer, who will provide an overview of our Q3 2021 financial performance and a review of our business operations. Also joining us today is Brendan Teehan, our Chief Operating Officer and Head of Commercial, who will provide updates on our commercial performance. Dr. Serge Stankovic, our President, will discuss our pipeline progress; and our Interim Chief Financial Officer, Mark Schneyer, will then discuss our financial results in more detail before turning it back to Steve for final remarks and opening the call for your questions.

I would also like to point out that we are using supplemental slides, which are available on the Events and Presentations section of our website.

Before we proceed, I would first like to remind you that during our call today, we'll be making a number of forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. These forward-looking statements, including goals, expectations, plans, prospects, growth potential, timing of events or future results, are based on current information, assumptions and expectations that are inherently subject to change and involve a number of risks and uncertainties that may cause actual results to differ materially. These factors and other risks associated with our business can be found in our filings made 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 Steve.

Thank you, Mark. Good afternoon, everyone, and thank you for joining us today. I'd like to start with a quick recap of our commercial performance, followed by a regulatory update for pimavanserin and a preview of our upcoming catalysts. Please turn to Slide 4.

First, I'd like to take a moment to remind everyone of our announcement last week that Brendan Teehan has been promoted to the position of Chief Operating Officer and Head of Commercial. Brendan has been a tremendous leader for our team, and I'm very excited to be able to further leverage his deep experience and strong capabilities. While Brendan will take you through our commercial update in greater detail in a minute, let me just say that on a macro level, the Parkinson's disease market continues to be significantly impacted as a result of the pandemic. Despite these challenges, NUPLAZID has outperformed branded drugs in the space and for the third quarter of 2021, NUPLAZID achieved $131.6 million in net sales. This represents a 9% year-over-year increase with sequential and year-over-year volume growth.

As shown in the graph on the left, our team has continued to execute with year-over-year and sequential growth each quarter since the beginning of the year. To be more precise, NUPLAZID have outperformed a market basket of top prescribed brands in neurology and Parkinson's disease in the office-based setting. Similarly, NUPLAZID continued its strong outperformance in the long-term care channel.

Our strong relative performance underscores our team's ability to adapt and find ways to grow our brand. This includes our recently launched branded campaign targeting health care providers, which promotes NUPLAZID's unique safety profile, specifically as it relates to the Parkinson's community.

On the patient and caregiver front, we're in the process of launching a new DTC campaign aimed at solving one of our biggest COVID-related challenges, which is physicians not seeing as many PDP patients in person. As such, this DTC campaign is aimed at patients and caregivers who are at home and is designed to activate them to have a conversation with their doctor about their symptoms of psychosis and potential treatment with NUPLAZID. Our growth initiatives, new leadership and strong relative performance give us confidence that we will further accelerate the growth of NUPLAZID into 2022.

Now let's move to our regulatory update regarding our sNDA for pimavanserin on Slide 5. As you recall at our Type A end-of-review meeting, the FDA made it clear that today, they are looking at individual subgroups of dementia rather than DRP as a single group. Accordingly, they stated that they believe our best path forward is to conduct an additional study in each subgroup where we seek approval.

However, at that meeting, the FDA also stated based on additional analysis we've shared with them that they are open to having another meeting to discuss whether there is a potential path to resubmission without an additional clinical study in any of the subgroups. This meeting is now scheduled, and we expect to be able to report on the results of the meeting around year-end.

When we look at our dementia subgroup data, it is very clear that we have the most data for Alzheimer's disease, which represents somewhere between 60% and 80% of all dementia patients in the U.S. Our database includes 2 independent clinical studies providing evidence of pimavanserin's antipsychotic efficacy in Alzheimer's disease psychosis, Study 019 and the HARMONY study. Together, these studies demonstrate improvement in psychotic symptoms and reduction of risk of relapse of psychosis in ADP patients.

At our upcoming meeting with the FDA, we look forward to sharing additional analyses we've done in response to feedback from the last meeting. Serge will provide more details on this opportunity in his section.

As we look ahead, I'd like to highlight some important near-term catalysts as shown on Slide 6. First, as I've noted, we look forward to providing an update from our meeting with the FDA around year-end. Second, we expect to announce top line results from LAVENDER, our pivotal Phase III study for trofinetide in Rett syndrome later this quarter. And third, we expect to announce top line results from a Phase II proof-of-concept study evaluating ACP-044 in postoperative pain following bunionectomy surgery in the first quarter of 2022.

Finally, our company is well poised for further expansion. We have a healthy balance sheet, a growing revenue base and best-in-class in-house R&D and commercial teams. Business development continues to be a key priority of our strategy and we continue to be very active on that front.

I would now like to turn the call over to Brendan to discuss our commercial performance.

Thank you, Steve. Today, I'd like to review our third quarter performance for NUPLAZID in Parkinson's disease psychosis. Please turn to Slide 8.

In the quarter, we delivered net sales of $131.6 million. This quarterly performance represents sequential demand growth fueled by new patient starts and continuing patients with strong adherence and compliance rates. I'd like to start with a further dive into the current macro environment that we're operating in.

As we stated previously, the pandemic has disproportionately affected the Parkinson's disease patient population, significantly impacting the market. As shown on the slide, PD office visits and occupancy rates at long-term care facilities are still down significantly from pre-pandemic levels.

Despite this, NUPLAZID has continued to grow meaningfully throughout the period with strong relative performance as compared to the overall Parkinson's disease and broader neurology markets. More specifically, when we compare recent quarterly performance with pre-pandemic performance of 2019, we see the following: In the office based setting, carbidopa and levodopa are down 4%, and the top 10 Parkinson's brands, on average, are down 7% over that time period. And while the top 15 marketed neurology products are up 15%, NUPLAZID is up fully 28% for that same time period.

Similarly, in the long-term care channel, carbidopa and levodopa are down 12%, and the top 15 LTC brands are down, on average, 22%, while NUPLAZID has grown 8%.

Perhaps the most important of all of these data points is the carbidopa and levodopa total prescription trend. These are the fundamental PD movement disorder medications and they have trended down during the pandemic. This decline clearly signifies just how challenged the PD market has been over the past 2 years.

However, against the backdrop of this environment, NUPLAZID has continued to grow in both channels, giving us confidence in our future growth. We've had to adapt and utilizing novel analytics and thoughtful implementation of our tactics have found new opportunities to grow new patient starts. Our strong relative performance and continued growth across all channels underscores our commercial execution and gives us confidence that our tactics are working and will continue resonating in months ahead.

Furthermore, our team is not simply waiting around for the macro environment to improve. We are buckling down and working even harder to ensure that we are interacting with and educating physicians, patients and their caregivers on PD psychosis and treatment with NUPLAZID.

Let's turn to Slide 9 to continue the discussion on what we are doing to further grow the brand. To continue to drive brand choice and growth in the second half of the year, we have introduced 2 new important campaigns, one focused principally on the health care provider or HCP audience and the other focused on engaging patients and caregivers.

On this slide, I'd like to discuss our new HCP-focused messaging platform that educates the treating community on the unique safety profile of NUPLAZID, specifically as it relates to the Parkinson's community. Our campaign is amplified across all promotional channels, including our field teams and is resonating well with HCPs.

This messaging incorporates market research insights focused on HCP's treatment goals for Parkinson's patients experiencing psychosis and specifically on NUPLAZID's safety and tolerability profile in light of other important medical consideration, the most notable being the desire to avoid impacting motor function. Our messaging demonstrates a deep understanding of our clinicians treatment goals by leveraging important clinical data to better educate physicians on NUPLAZID's safety and efficacy profile.

While early in the rollout of this campaign, this messaging is resonating with physicians, and we see the opportunity to further increase diagnosis and capture a higher percentage of new PDP patients for NUPLAZID moving forward.

Slide 10 highlights our new direct-to-consumer campaign, which is intended to reach patients and caregivers at home and drive them to their physicians to have conversations about their symptoms of psychosis and treatment with NUPLAZID. This campaign should help solve one of our biggest COVID-related challenges, patients staying at home and not seeing their physicians in person and therefore, not adequately discussing their symptoms or new treatment options.

Our campaign has been on air since mid-October. It tells an important patient caregiver centered story about the impact of PD psychosis. The ad introduces NUPLAZID and its potential treatment benefits to both patients and their families. The ad also includes a critical component of disease awareness to help patients understand and recognize the signs and symptoms and prevalence of psychosis in patients living with Parkinson's disease.

With the work our commercial team has been doing, executing on the new message platform on the HCP side, this is the perfect time to introduce a new patient-centric campaign, drive new patients to their newly educated doctors and grow the NUPLAZID brand. Of course, while we expect to see traction in the fourth quarter, the full benefit of these complementary campaigns will be mostly realized in next year's growth.

The bottom line, patients and caregivers are in need of a better treatment option, such as NUPLAZID and patients and caregivers need to recognize the symptoms of psychosis and its connection to Parkinson's disease. We are confident our campaigns will accomplish this. I will now turn it over to Serge.

Thank you, Brendan, and good afternoon, everyone. As Steve mentioned, we have had a constructive dialogue with the FDA and are looking forward to our upcoming meeting. While we continue to believe that evaluating pimavanserin as a treatment for the broad dementia-related psychosis indication is most appropriate, we understand the FDA's position and are preparing additional analysis to support pimavanserin as a potential treatment by dementia subgroup.

Let's start the discussion on Slide 12 with an overview of our HARMONY data, specifically looking at the Alzheimer's disease subgroup, which, as Steve mentioned, is the most prevalent form of dementia.

On the left side of the slide is an overview of pimavanserin antipsychotic efficacy observed overall and in the Alzheimer's disease subgroup in the open-label portion of the HARMONY study as well as the primary efficacy outcome in the double-blind portion.

As we previously noted, HARMONY was not powered to show statistical significance by dementia subgroup. However, we did prospectively plan exploratory efficacy analysis of dementia subgroups. In the Alzheimer's disease subgroups, we see that Alzheimer's disease subjects who remain on pimavanserin were about 40% less likely to experience a relapse of psychotic symptoms compared with those on placebo as shown with hazard ratio of 0.62. Post hoc analysis of the pimavanserin 34-milligram dose group, the dose currently approved for treating PDP patients showed about 50% risk reduction with a hazard ratio of 0.47.

For context, this magnitude of effect is in line with what has been observed in numerous relapse prevention studies of comparable design for approved drugs in various psychiatric indications such as schizophrenia or depression. In addition, this benefit is further supported by a number of complementary analysis, some examples of which are listed on the right-hand side of this slide. These analyses consistently show the antipsychotic efficacy across different endpoints within the Alzheimer's disease subgroup, which clearly support the overall conclusion of the clinically meaningful benefit that pimavanserin demonstrated in the treatment of psychosis in Alzheimer's patients.

Furthermore, these additional analysis also demonstrate and confirm pimavanserin's observed antipsychotic treatment effect across all dementia subgroups and thus provide further supportive evidence for the benefit of pimavanserin in Alzheimer's disease psychosis.

Please turn to Slide 13. In the upcoming meeting, we will also discuss with the FDA our positive 019 study, a randomized placebo-controlled study of pimavanserin for the treatment of symptoms of psychosis in Alzheimer's disease patients. The primary efficacy outcome was positive with p-value equal to 0.045.

In a prespecified analysis of Alzheimer's disease subjects with more severe psychotic symptoms, the magnitude of efficacy more than doubled with a p-value equal to 0.011. In response to issues raised in the complete response letter, ACADIA has conducted multiple sensitivity and responder analysis that support the primary efficacy conclusion. Note that the responder analysis, shown on the right of the slide, also demonstrate that pimavanserin had even greater efficacy in those patients with a higher severity of psychosis.

To recap, pimavanserin has been evaluated in ADP patients across 2 independent clinical studies and in addition to positive and clinically meaningful efficacy has been well tolerated and, importantly, has not shown negative impact on cognition or motor function. In our next meeting with the FDA, we will share several analysis and a substantial data set demonstrating the potential utility of pimavanserin in the treatment of ADP in order to further examine the potential for resubmission without conducting additional clinical studies.

Now I would like to discuss our pivotal Phase III study evaluating trofinetide as a potential treatment for Rett syndrome on Slide 14. LAVENDER is a randomized double-blind placebo-controlled study in approximately 180 young females age 5 to 20 with Rett syndrome. Patients are evaluated for 12 weeks. The co-primary endpoints are Rett Syndrome Behavioural Questionnaire, or RSBQ, a caregiver assessment tool and the Clinical Global Impression of Improvement, or CGI-I, which is a physician assessment.

We are currently on track to announce top line results by year-end. And importantly, we have agreement with the FDA that positive results from LAVENDER in addition to supportive efficacy data from the previous Phase II study and the safety database we are collecting could be sufficient to support a new drug application.

Finally, let me remind you that there is nothing approved for the treatment of Rett syndrome. Rett is a devastating and burdensome disease for both the patients and their caregivers, often requiring a round-the-clock support. We have had several discussions with the key opinion leaders and experts in the field. They inform us that even a modest improvement of symptoms could make a significant and very meaningful difference in the ability to care for the patients and for their overall functioning. We look forward to announcing the results of this study in the near future.

Slide 15 shows a summary of our Phase III program evaluating pimavanserin for the treatment of the negative symptoms of schizophrenia, which includes 2 pivotal studies, our positive ADVANCE-1 study and ADVANCE-2, which we initiated in the third quarter of last year.

Please turn to Slide 16 for an update on our ACP-044 program. Our ongoing Phase II study evaluating ACP-044 for the treatment of postoperative pain following bunionectomy surgery is nearing enrollment completion. However, we now expect the results in the first quarter of 2022. This slight delay is due to slower-than-expected enrollment with the postponement of many elective surgeries during the summer COVID-19 Delta variant surge.

In addition, as a reminder, earlier this year, we initiated a Phase II study for patients suffering from pain associated with osteoarthritis and plan to provide an update on this study next year.

Slide 17 highlights a brief summary of our ACP-319 M1 PAM program for the potential treatment of schizophrenia and cognitive impairment in Alzheimer's disease. We recently initiated a multiple ascending dose study as our Phase I work continues for this program.

Turning to Slide 18. At ACADIA, we are committed to investing in therapies to address high unmet needs in CNS. Our clinical development pipeline has 2 late-stage Phase III programs as well as multiple early-stage program. This year, we have initiated multiple clinical studies, including ACP-044 in acute pain in Q1, ACP-044 in chronic pain in Q2 and most recently, the ACP-319 multiple ascending dose study.

With that, I'll turn the call over to Mark.

Thank you, Serge. Today, I'll discuss our third quarter 2021 results. Please turn to Slide 20. In the third quarter of 2021, we recorded $131.6 million in net sales, an increase of approximately 9% compared to $120.6 million of net sales in Q3 of 2020. Our net sales increase in Q3 2021 represents 3% volume growth year-over-year.

The gross-to-net adjustment for Q3 2021 was 15.2%. Weeks of inventory in the channel at the end of the third quarter was slightly down. As a result, sequential demand growth of approximately 3% was slightly higher than sequential selling growth of approximately 1%.

Moving down the P&L. GAAP R&D expenses decreased to $58.6 million in the quarter compared to $120.1 million in Q3 2020. Recall last year, GAAP R&D expense included the $52.8 million upfront consideration and transaction costs related to our acquisition of CerSci Therapeutics.

GAAP SG&A expenses were relatively flat at $81.7 million in the third quarter compared to $81.6 million in the third quarter of last year. Noncash stock-based compensation expense during the quarter was $15.5 million compared to $21.4 million for the same period in 2020. Our cash balance at the end of the quarter was $540.3 million.

I'll now provide some additional color on our financial expectations and guidance ranges for the remainder of the year. Please turn to Slide 21. As we mentioned previously, our commercial team is continuing to execute well and deliver quarter-over-quarter growth with strong relative performance in both channels. As a reminder, in the fourth quarter, we expect a much higher impact from gross to net as a result of accruing for the donut hole obligation associated with year-end inventory in the channel. In addition, recall that last quarter, we guided to gross to net being somewhere around 20% for the full year.

As we get closer to the end of the year, we have narrowed the top end of our full year 2021 net sales guidance range to $480 million to $500 million from the previous range of $480 million to $550 million. Our net sales guidance assumes that inventory levels remain relatively flat in the channel as we approach the end of the year.

I'd like to provide a little bit more commentary on our net sales guidance. As we saw in Slide 8 of Brendan's presentation, our leading indicators of PD office visits and LTC occupancy rates have started to improve in the first half of the year. However, in the third quarter, they seem to have leveled off or declined. This is a reflection of the COVID-19 Delta variant surge we experienced in the summer. Our net sales guidance range factors in scenarios that depend on how long our leading indicators remain relatively flat versus how quickly they return to growth.

Moving on to the expense side for 2021. We have lowered our GAAP R&D guidance to be between $230 million and $245 million for the full year from a previous range of $250 million to $270 million. And finally, we have slightly narrowed the top end of our GAAP SG&A full year guidance range to be between $385 million to $405 million from $385 million to $415 million.

And with that, I'll turn the call back over to Steve.

Thank you, Mark. Please turn to Slide 23. In closing, I'd like to remark on how proud I am of our teams for their execution this year. Our commercial team has worked tirelessly all year on delivering pimavanserin to more Parkinson's patients and their families who are suffering from the symptoms of psychosis. They've achieved more with less and continue to push through the pandemic conditions.

Our R&D team has executed during a time when patient enrollment is difficult. We are now poised to report out on 2 key clinical studies: Phase III results from LAVENDER, our trofinetide study in Rett syndrome; and Phase II results from our ACP-044 study in postoperative pain. And of course, we look forward to providing additional clarity regarding ADP around year-end.

Finally, I would like to thank all of our employees for their commitment to our mission to elevate life. I'll now open up the call for questions. Operator?

(Operator Instructions) Our first question comes from the line of Neena Bitritto-Garg from Citi.

Congrats on the quarter. So just a question about the commercial performance of NUPLAZID. I guess can you comment at all on the pace of new starts during the quarter versus Q2? And then also just on Q4, just from some of Mark's commentary just now, it sounds like we should expect a higher gross to net and kind of similar patient dynamics quarter-over-quarter. I guess should -- is it kind of safe for us to assume that Q4 could actually be kind of a down quarter?

Yes. Thanks for the questions, Neena. I'm going to ask Brendan to answer the first question and Mark the second.

Neena, thanks for the question. The question regarding new patient starts in the third quarter, we are seeing new patient starts that are approaching the pre-pandemic level, so largely comparable between the quarters. As we head into the fourth quarter with DTC and our increased -- and our efforts with our HCP campaign, we would expect to continue growth moving into 2022.

Thanks, Brendan. Mark?

Yes. Neena, thanks for the question. I think with the range of guidance that we put out for net sales, there's -- I think what I want to point out is there's volume growth across that full range. But yes, we are with an expected higher gross to net adjustment for the fourth quarter. At the lower end of that range, the implication for fourth quarter net sales is that reported net sales at the bottom end of the range would be lower than the third quarter reported net sales.

Our next question comes from the line of Cory Kasimov from JPMorgan.

This is Gavin on for Cory. We had a question actually on the DRP regulatory review. Karuna is expected to invest their drug candidate for DRP but are now planning to launch a Phase III next year in ADP specifically. Have you had any feedback from the agency about how they're broadly thinking about the DRP setting relative to the independent components, as this development would seem to increase it to help your case?

Gavin, I didn't hear the second part of that question. Could you repeat that for us?

I just said that this development would seem to help your case.

Got it. Okay. Thanks. Well, we can't obviously comment on what other people's interactions are with the FDA. I would just simply say that the FDA has made it clear to us that they think the correct way to assess this population based upon reviewing our data is on a subgroup by subgroup basis. So that's what we're focused on. That's what we'll be focused on at the next meeting.

Our next question comes from the line of Ritu Baral from Cowen.

I did want to follow up on that last question. I guess I was going to ask it a slightly different way. As you approach this meeting and come out on the far side, if the FDA does prove immovable, is an ADP specific trial sort of the best next option? Insofar as you clearly have treatment effect, it would just take a well-powered study to address, as you mentioned, 60% to 80% of the overall DRP population.

And then further, as you plan to discuss 019, can you remind us -- I believe the FDA sort of changed its mind on whether they consider 019 a well-controlled study. And I think that was around sort of like the whole "administrative single-center" aspect of it. Can you remind us of what their interpretation of well-controlled was versus ACADIA's conduct of the study?

Yes. Thanks for the question, Ritu. I'm going to take the first part. I'm going to ask Serge to answer the 019 portion of the question. Serge, if you have any other comments regarding the HARMONY please chime in as well.

So just to recap very briefly, we received a CRL in April. And through that process and our Type A meeting, the FDA made it clear that they are looking at dementia-related psychosis on a subgroup by subgroup basis, as I mentioned a second ago. At our Type A meeting, we presented data to them that was intended to demonstrate both the consistency of response on drug-treated patients versus placebo-controlled patients or placebo-treated patients and also cluster analysis that broke down the individual components of the SAPS-H+ D scale, which was used throughout the study. It was used for enrollment entry criteria. It was used during the 12-week response period and then during the 6-month randomized portion of the study.

And when we broke that down on individual components, what we saw, again, was a very consistency of response irrespective of underlying subgroups. So across each of those elements, patients looked pretty much the same when they came into the study. They looked -- they responded in a very similar way during the 12-week open-label period. And then they responded -- those on drug, that is, responded in a very similar way during the randomized portion of the study.

One area where we did see a difference was in the placebo group on patients that have both comorbid Parkinson's disease and frank dementia. In that group, we saw a more rapid relapse in that group relative to patients -- placebo patients in other subgroups. And we attribute that to the fact that these patients are taking dopaminergic therapies for their motor function, and that is known to -- can exacerbate psychotic symptoms.

So at the conclusion of that meeting, the FDA said, we get your points. We think you made some valid points. We want to reiterate, we think the best way to study this population is on a subgroup by subgroup basis. And we think your best path forward is to run an additional study in each of the subgroups that you are seeking approval.

However, we didn't get to complete the conversation. It's a 1-hour meeting, of course, there is dialogue back and forth before the meeting. And they said, we would be very open to having another meeting with you, if you'd like to further discuss the potential for resubmitting without an additional clinical study.

And so what we're telling you today is we think the subgroup where that is -- where there's the greatest potential for doing it is in Alzheimer's disease psychosis. That's the group that was about 70% of the population in our HARMONY study. Also epidemiologically, it represents about 70% of dementia patients. And we also have our study 019 that Serge will speak in a second, where we had a very positive result in the Alzheimer's dementia psychosis population.

And so for all those reasons, as we said, we're very focused now on this next meeting in taking new analyses that we've done since the last meeting that focuses on the Alzheimer's disease psychosis population. So we'll look forward to having that meeting. Obviously, that's the next step we need to have a meeting. And as soon as we have, we'll be able to give further guidance about the next steps. Serge, you want to address the 019 study?

Yes. Thank you. In respect to 019 study, just a reminder, our Alzheimer's disease psychosis study that was conducted in the care homes -- in a number of care homes in the U.K., FDA raised certain issues that in the complete response letter, they expressed concern that makes it difficult for them to consider this study adequate and well-controlled study.

These issues by the FDA can be roughly categorized into 2 buckets. One is related to the design of the study and the others are related to conduct of the study or the protocol deviations that were recorded and reported in the study. In respect to design, 2 specific issues were emphasized: one, that this is a single center study; and the second is that the secondary outcome measures, there was no Type 1 error control for the secondary outcomes.

We believe that we can and intend to address these concerns, particularly from the perspective that in our understanding, none of these are requirements for the adequate and well-controlled study. Just as a reminder, as I said, this study is done under coordination of a single investigator in the U.K. but it was done in a number of care homes or nursing homes throughout the broader U.K. London area.

From the perspective of the conduct, we also intend to address not only in detail, specifically the protocol deviations that are reported -- and by the way, they are reported in the study report from very beginning and we discussed that. But we will further elaborate on the specifics and timing and their ability to impact the study per se and the interpretation of the results. But also, we will provide a number of additional sensitivity analysis that are intended to demonstrate that this protocol deviations did not impact the overall conclusions of the study in any way.

So in short, we are -- from the data perspective, we have a level of comfort that we can address these concerns raised by the FDA. And that is exactly what we intend to do in the -- in our meeting as well as in -- eventually in the resubmission. And we believe that following that, this study can and should be considered adequate and well-controlled study and supported in the overall as an independent evidence of efficacy of pimavanserin in the Alzheimer's disease psychosis.

Our next question comes from the line of Charles Duncan from Cantor Fitzgerald.

Steve and team, congrats on a solid quarter. I wanted to ask you one commercial question and then one pipeline question. Regarding the commercial question, I guess I'm thinking about the 2 new campaigns for HCPs and direct-to-consumer. And I think Brendan suggested that higher percentage of PDP diagnosis would be the goal. I guess I'm wondering if you could provide us some color on where you're at today and where you'd like to be in, say, 12 months.

Sure. Brendan, go ahead.

Charles, thank you very much for the question. First, I'd say we're excited about both campaigns. The HCP campaign is focused on drawing attention to the ability to improve psychosis while not impacting motor function. We know that our patients are going to see their HCPs principally because of their Parkinson's disease. And we need to weave in the story about Parkinson's disease psychosis and the need to treat.

We've done market research on the campaign. HCPs are responding very favorably to it. They demonstrate an interest -- a higher interest in prescribing NUPLAZID as a function of telling the story that way.

The second campaign, more focused on patients and caregivers. You can see in the DTC story, which is to, first of all, highlight NUPLAZID as the solution to Parkinson's disease psychosis but every bit as importantly, to make sure that those patients and caregivers understand that this is a normal part of their Parkinson's disease experience. And thus, it's important to bring that situation to their HCP's attention. So we are using those in a complementary fashion. We've been very proud of the DTC campaigns up to now and how that has engaged patients and caregivers with their treaters in that conversation. And we're very confident that, that will happen again here.

Okay. And then second question is for Serge, on the pipeline. Perhaps you just answered this in addressing Ritu's question. But I'm wondering if you could provide us a little bit more color on the new analyses that have been conducted. Are they primarily around efficacy? Are they around safety? I mean Brendan just mentioned a lack of impact on motor function. And it would seem to me that would be important as well as cognition in the ADP population. So what kind of analysis have you conducted?

And then secondarily, when would you be able to update this? Would the meeting happen by the end of the year? Or would you be in a position to update by the end of this year or so?

Serge, please go ahead.

Yes. Thanks, Charles. Let me just get one by one. In regard to new analysis, I'll start, first of all, by stating something that we previously also stated, that questions raised in the complete response letter were related to the way of how the efficacy evidence in the patient population should be derived and that the -- as Steve says, what is the appropriate way of evaluating efficacy in dementia patients with psychosis, whether as a group or as by dementia subtype.

So as you can imagine, most of the discussions that we have is around that evidence of efficacy in patient population. And in this specific case, we are preparing a number of analysis that are specifically designed to where it's a natural to start with a subgroup that is the largest subgroup of Alzheimer's disease psychosis, where we are providing a broader additional analysis. And I'll comment on that just in a second.

But also, I do want to address your question about safety and that is there weren't issues related to safety because the pimavanserin safety profile is unchanged and all the additional data that we have provided in the supplemental NDA did not open any new questions in regard to the safety of pimavanserin. That holds as well for the certain safety benefit that we see with pimavanserin and that is related to the motor function and specifically in this patient population with dementia to lack of negative impact in cognition and cognitive functioning. And of course, that is a part, it was a part of our supplemental NDA submission. It's also a part of this discussion where we certainly are providing the information in that respect.

In regard to specifically Alzheimer's disease psychosis and the additional evidence that can be looked at in a 3 -- grouped in 3 different groups. One is related to the primary and secondary outcome, and that is of the reduction in relapse as well as reduction in the -- all-cause discontinuation in this specific group, where we are providing a variety of ways of looking into that analysis and demonstrating the benefit that is clinically meaningful and substantive. As I mentioned in the prepared remarks, between 40% and 50% reduction, where we are also providing not only the overall in the Alzheimer's group but also specifically in the larger group of patients that received 34 milligram, which is a dose we -- recommended dose and the approved dose in PDP and recommended dose in DRP or Alzheimer's disease psychosis.

Second group analysis is following the different ways of measuring severity of psychotic symptoms in all of the patients following randomization either to continue active treatment with pimavanserin or to discontinue pimavanserin and switch to placebo or to placebo. We are following over time the severity of symptoms in these 2 groups of patients, both on the scale for psychotic symptoms for hallucinations and delusions as well as in the Clinical Global Impression of Change and Improvement, where we are looking at and demonstrating that patients that remaining on treatment continue -- overall continue with the benefit of pimavanserin in terms of control of psychotic symptoms.

We are also looking on the relationship of the dose and drug exposure, pharmacokinetic exposure and the efficacy of the drug as another way of looking at the effects that we are seeing and benefit that we are seeing in patients that are receiving pimavanserin is a real true effect and not a spurious finding.

And the third group of evidence is really related to the broader overview of the response across different dementia subtypes and characterization of psychotic symptoms prior to treatment following the treatment with one purpose, and that is to demonstrate that the benefits that we are seeing across the subgroup is another supportive evidence that what we are seeing in Alzheimer's disease psychosis subgroup is a real effect.

And finally, I will just mention that we are also looking at a categorical presentation of response across different categories of patients, specifically in the Alzheimer's disease psychosis and confirming again both overall and in the 34-milligram group the substantive evidence of benefit in treating patients with pimavanserin. So I hope this helps on a broader scale.

Definitely. Look forward to hearing the response from the agency. Hopefully, they allocate more than an hour to you. And I'll forget about the timing of the response. We'll just hear it when we hear it.

Yes. Well, I'm sorry, I will address that. It just slipped my mind. We are saying that we -- meeting is scheduled, we will have the meeting and anticipate that we will receive minutes from that meeting around year-end just because we are talking about year-end holidays, it's a little bit difficult for us to confidently say that whether we will receive the minutes before the end of the year or maybe just after the end year. And therefore, we are characterizing that at around year-end.

Our next question comes from the line of Yatin Suneja from Guggenheim Partners.

Just a few questions on trofinetide. So you guys are looking at 2 -- or a co-primary end point. Could you comment on what sort of agreements you have with the FDA, if you hit on one and what exactly you are aiming to show both on RSBQ and CGI?

And the second question I have is that could you also confirm to us or let us know if all patients have entered the OL portion of the study? Or are there any patients who decided to enter -- to not enter the OL after the 12-week treatment period?

Yes. Thanks much for the question. Serge, do you want to take both of those?

Yes. Thank you, Yatin. First, in terms of the co-primary measure, as we are this close to actually having the actual data from the study, I think I would hesitate to speculate on different scenarios that may occur. Obviously, what we are focused on is obtaining a positive result and statistical separation of both the Rett Syndrome Behavioural Questionnaire scale and on the clinician scale. Rest assured, we are certainly preparing for all different scenarios in the background. But I don't think it would serve any good purpose for us to speculate what would be -- what would happen if it happens.

From the -- on the second point, vast majority of patients -- we, of course, don't provide, again, the specific numbers. But I can say that vast majority of patients had rolled over into the open-label extension of the study. And therefore, I think that's -- that hasn't been an issue from the perspective of the further safety exposures on the -- for the patients in this program.

Got it. Just one more clarifying question on the time line. Could you just maybe let us know how long it might take for you to analyze the data once the trial is completing? Just trying to get a sense that if these data could sort of get pushed to January.

I can just say that we are on track to report top line results from the study before the end year. There hasn't been a change in that respect.

Our next question comes from the line of Marc Goodman from SVB Leerink.

Serge, you explained nicely what new data you're going to be giving to the FDA when you do the meeting. But I'm just curious, there were a couple of slides that you went over with us today. Has the FDA even seen that data, that subgroup data in Alzheimer's?

Yes. Marc, thanks for the question. We have, in the previous meeting, discussed some of this. So they have seen some of the data discussion in the general terms of what we should be providing have been discussed. But some of the output on some of these analyses hasn't. So it's sort of a mixture in that regard, but they have -- we have provided them with a good general idea what direction we are moving in that respect. And we'll be providing much more color in the next meeting in terms of the output and specific data.

Right. Okay. And then just separately a commercial question. Can you just give us a sense of how the reps in the office has just changed relative to second quarter and how -- in the third quarter and then how fourth quarter has changed? Just give us a sense of the past 4 months, how that's changed.

Yes. Thanks for the question, Marc. Brendan, do you want to take that?

Sure. Marc, thanks for the question. We have seen improved access rates for our representatives, both in the community and LTC setting. So I think we previously reported very much in line with what we've seen in the industry. We're up around 70% of visits being face-to-face. It varies a little bit between community and LTC. But we are encouraged by what we're seeing so far in the early fourth quarter.

And where were we 3 months ago?

So we would -- we were reporting in the 60%, 65% range. So it's continuing to arc up.

Maybe just I can say a little bit more, it's very low at the beginning of the year for us and everyone else. And what we've seen in the industry is month-by-month from the beginning of this year, those numbers were really going up on a very continual and meaningful basis (inaudible). With the Delta variant, you saw that slow down quite a bit. So where we stand today is, as Brendan mentioned, about 70% of our details are in person and about 30% are virtual.

And I think as we see the impact of the Delta variant kind of pull all the way through, and you guys are all seeing the same press that we are, I think we're probably approaching that point now. We'll see the impact of the Delta variant pass as well.

Our next question comes from the line of Gregory Renza from RBC Capital Markets.

Steve and team, congratulations on the quarter. Steve, just a follow-up on the regulatory update that's expected and as you proceed with engaging the FDA. I know you -- in the past year, you have characterized the stance as more of an agree to disagree position with respect to the totality of DRP versus the subgroups. And I'm just curious if you can comment quickly on how that's evolved, if that's changed as you're going into the subgroup meeting or the meeting on the subgroup with respect to starting with Alzheimer's disease. And just your level of enthusiasm that that's the right way to proceed with the DRP indication going forward.

Let me take -- thanks for the question. Let me take a little bit of a running start at it. There is no drug approved to treat dementia-related psychosis. There's no drug approved to treat Alzheimer's disease psychosis. So we're breaking ground. We today have undoubtedly the largest safety database in dementia-related psychosis patients or Alzheimer's disease psychosis patients.

In addition, we've done really groundbreaking work in terms of trying to determine how to treat these patients. Our view has not changed. We still believe that the best way to study this population is looking at dementia-related psychosis as a whole. Reasons for that, described very thoroughly before, the underlying etiology is relevant as it relates to the dementia itself. But as it relates to psychosis, the symptoms are very similar. They respond in a very similar way. And we think that is the best way to study population.

However, in this groundbreaking area, where there's very little data to go on, we understand that reasonable minds could differ and the FDA does not share that view. Their view is that we should be looking at things, and they should be looking at things on a subgroup by subgroup basis.

So recognizing that, we have pivoted to focus on that for the next meeting. I've described the history of the CRL dialogue we've had. And I would simply say at this point in time, the FDA, this includes the Division of Psychiatry as well as the Office of Neuroscience. They have been very engaged. It's a very important topic to them. Of course, it's a very important topic to us as well.

And together, we expect at this next meeting to come to a conclusion. We are eager to make a decision and move forward. And if they remain engaged in the way that they have been up to this point, I think the prospects are very good. But I'll simply say, again, this is a new area where there's a lot of virgin territory that has not been encountered before.

Having said that, when we look at our data in ADP, that is Alzheimer's disease psychosis, we think that data represents a strong case for the utility of pimavanserin from both an efficacy as well as a safety and tolerability perspective for utility in those -- in that patient population. We need to have a meeting with FDA to see if we can find alignment around that.

Our next question comes from the line of Ami Fadia from Needham.

Great. I had a follow-up on DRP. And I want to ask the question in a slightly different way. Is the point of contention with the FDA around whether or not there is adequate data in the Alzheimer's disease subpopulation or whether or not the Phase II study can be viewed as acceptable as part of a registrational package? If you could give us your best sort of view on that. And then I have a different question.

Well, I'll try to give you a little bit of additional color. And then Serge, feel free to jump in, if anything else you want to add. I think when we designed our Phase III program and we aligned with the FDA about how to pursue that program, we and the FDA aligned around a design that was not to study Alzheimer's disease psychosis or dementia with Lewy body psychosis or Parkinson's dementia psychosis. It was to study all of the subgroups as one single group. And so that's the Phase III program we ran.

Today, as I've just described, the FDA has a perspective that we should be looking at things or they are looking at things on a subgroup by subgroup basis. And so from that perspective, we're in the situation it's a little bit atypical because we're talking about now a group that was not -- where we did not design the study to show statistically significant data on a subgroup basis. But we have very clinically meaningful results in that subgroup.

As Serge described, I think, very thoroughly, we have -- we've looked at things from a multitude of cross-sectional perspectives to see if we have consistency in that data and we do. We also have our 019 study that was focused on just Alzheimer's disease psychosis patients. We have a positive study there. As Serge just described, we need to have further dialogue with FDA around the perspective of that being an adequate and well-controlled study. So that's where the pieces on the chessboard are. We're very eager to have the next meeting with the FDA to see if we can align around the data that we have.

Yes. If I just may add, I think it's a very important to remember that with everything that Steve said that we did not set out to study Alzheimer's disease subgroup or any other subgroup specifically but more as a group, that now that we are lining out the evidence for benefit in that specific subgroup, this has to be taken in the context of actually us having 2 positive studies: One positive study is 019 study, and the second positive study is DRP study and looking at the subgroup of Alzheimer's disease psychosis is in the context of the robustly positive overall evidence of efficacy.

This is a very different situation that if you would look on the study that failed on its primary analysis and then you're looking at a subgroup and seeing whether there is a benefit in the subgroup. Completely different and from that perspective, much more persuasive and convincing in terms of evidence. So we believe that there is independent evidence to 2 separate studies demonstrating benefit in this particular subgroup of dementia.

Great. That was very helpful. My second question is on trofinetide. And you answered a question earlier about the core primary endpoints and the clinical meaningfulness of those. I understand that any type of improvement would be welcome in this patient population. However, as we think about the chronic nature of treatment and the possible requirements of other supportive care for these patients, how would you think about pricing the drug should it get approved?

Yes. Thanks much for the question. I think as much as I'd like to answer the question, I think it would just be premature and imprudent for us to comment on pricing at this juncture. The next step of course in that program is to open the envelope, determine the results. And if there -- if we have results that support an application, we'll file it and look forward to moving forward.

I would just simply say that as is the case with NUPLAZID, we would seek to price the drug based on the value that we're delivering. In the Rett syndrome population, it's a very, very, very unfortunate situation where it's extraordinarily high disease burden. And if we can help improve the lives of those patients and their caregivers, we think that will be very important from a medical perspective.

Our next question comes from the line of Paul Matteis from Stifel.

I had a couple of questions on the refiling strategy and just the DRP data in general. I guess one is just kind of related to reconciling one thing we've had trouble wrapping our heads around. And that is you speak to patients with DRP presenting similarly and responding similarly. But just going back to the forest plot, the hazard ratio in Parkinson's is around 0.1, and it's a lot higher in other subgroups except for Lewy body. So I was just kind of curious how you explain that and if that is not driven by biological differences in disease in your view?

And then second, on the ADP subgroup, can you just speak to statistics or anything around that, that can kind of quantitatively bolster your case? I know you talked about a 40% relapse reduction, but the hazard ratio upper bound is almost 1.5. So how do you think FDA will view that quantitatively speaking or if there's any sort of p-value that's tied to it?

Yes. Thanks much for the question. Serge, do you want to take those?

Yes. Let me address first question. Paul, you characterized the -- well, the principal question about the difference that we see in the observed hazard ratio in Parkinson's disease psychosis versus the other subgroups of dementia like Alzheimer's. And what I think the question is, is that really a reason for the biological difference in the response or there is something else going on here that we could point out to and think of.

I will, first of all, say when you just look to the relapse rate in patients on pimavanserin versus relapse rates in patients on placebo after randomization, what we see actually that relapse rates among dementia subtypes are very similar for the groups on pimavanserin, where the difference really is striking is in placebo, where patients are randomized to discontinuation of treatment. It turns out that the Parkinson's disease patients relapse much faster, at a faster rate than all other patients from the group.

So we thought about that, why is that happening? Because it's on the active treatment, the relapse rates are fairly similar among different subtypes, and thought about what is unique about Parkinson's disease patients that is not present in other subtypes. One thing that is unique is actually dopaminergic therapy. It is dopamine stimulating therapy or replacement therapy that these patients are on. All of them are on that. There is virtually just in the whole sample we have just one that wasn't.

And when you look at that and what is happening is that there is additional sort of stimulus that while the psychotic symptoms are controlled, once you discontinue antipsychotic treatment in these patients, they relapse faster because they have this additional kind of promoting element in their concomitant therapy. So we have presented -- and we intend to present some additional data to FDA in that regard because we look -- there are small samples. Obviously, there are very few other patients that are not Parkinson that are on dopamine stimulating therapy.

But we can look at that and look at the patterns. And it seems that there is some confirmation of this hypothesis that we are. So what we see in our data, it's not about biological difference. It is about confounding related to the concomitant therapy effects that we are seeing in the subgroup.

So I hope that helps a little bit with your conundrum and reconciling the different hazard ratios. And now I'll ask you in regard to the second part of your question, it escaped my mind while I was thinking and talking about this. So if you can remind me, please?

Yes. No worries, Serge. And thank you for the thoughts there. That was super interesting. As it relates to the ADP subgroup, right, I mean, we see the 40% relapse reduction, but we also see the hazard ratio for the 95% confidence interval. The upper bound is almost 1.5, which would suggest this is really far away from statistical significance or at least there's -- it's telling you there's a ton of, I guess, just sample size issue in the data. And I'd be kind of curious how you think about that.

Well, the -- as Steve said, the study is not designed to test statistically separation in the subgroups because we needed to have more patients in each of the subgroups in order to see that as a reasonable power. So confidence intervals that you are seeing are actually simply a consequence of the sample size that we have. Smaller the sample size, larger the confidence intervals that you are talking about. But what we are presenting as a body of evidence related to Alzheimer's disease psychosis is a number of different analysis. Some of them reach this nominal statistics significant, some doesn't. But what is important point is consistency of that effect.

The question is, is the 40% or 50% risk reduction that we are seeing, nobody argues, it's clinically meaningful and robust. The question is, is that a spurious finding or it is a real true effect. And one way to demonstrate that if you look through consistency of different analytical approaches, different outcome measures, different time points and point out to that consistency.

And as I said, sometimes it's not significant. Sometimes it's significant nominally because these are -- some of these are post hoc analysis or not controlled for Type 1 error. But the point is if you see something consistently, you start to be convinced that it is actually a true effect and true benefit rather than some spurious chance finding.

We have time for one more question. Our next question comes from the line of Jeff Hung from Morgan Stanley.

For ACP-044, can you provide more granularity on when you expect to complete enrollment and when in the first quarter the data might read out? I guess when would you need to have the enrollment completed for the data in January versus data in March? Just appreciate any color you can provide.

Yes. Thanks for the question, Jeff. Serge, do you want to take that?

Yes, happy to. Thanks, Jeff. What I will say is that we -- as I said, we are experiencing some headwinds but it's a slight delay. We anticipate that we should be completing enrollment before the end of the year -- by the end of the year. And therefore, by sometimes within the first quarter or towards the end of first quarter, we should have top line results. That's, again, as I'm saying, is in the context of dependency, particularly sensitivity of this type of elective surgeries and enrollment, there is always a certain level of uncertainty, but we are -- that's generally our expectation.

At this time, I'm showing no further questions. I would like to turn the call back over to Steve Davis for closing remarks.

Great. Thanks so much, operator. Thanks again, everyone, for joining us today. We look forward to updating you on our progress next quarter.

This concludes today's conference call. Thank you for participating. You may now disconnect.