Wonderful. Thank you very much. It's my pleasure to be here with Shawn Singh, President and CEO of VistaGen. You know, maybe, Shawn, you could just kinda give a quick update on where things are with the company, you know, what you learned from this recent phase III readout that didn't go your way, and then we can kinda get into it and be more forward-looking on the next data set for fasedienol.
Great. Well, thanks again, Paul, for inviting us to the event. Always appreciated. Just a quick note, of course, I'll be making forward-looking statements, so I encourage everyone to take a look at our SEC filings for information about the risks relating to our business. Just a quick background for people who might not know Paul. Fasedienol is our lead rapid onset, non-systemic, intranasal product candidate that is under development under Fast Track designation for the acute treatment of social anxiety disorder, which manifests in daily lives of people over many, many years, starting in adolescence, fear of embarrassment, humiliation, judgment in what most would consider kinda ordinary daily social and performance situations.
Fasedienol is designed to reduce the wave of anxiety that people usually experience right before they engage, or during a feared, anxiety-provoking social or performance situation in their daily life. We have a PALISADE program, it's a phase III registrational program, and it consists of several studies. You referred to PALISADE-3 . That followed PALISADE-2 , which was a positive study. PALISADE-3 did not separate. PALISADE-4 is what is ongoing now, and we expect that to read out during the first half of this year. PALISADE-3 gave us a real valuable opportunity to take a step back and look carefully across the development program. When we analyzed PALISADE-1 , 2, and 3 together, I think what stood out for us was the treatment effect with fasedienol across the studies.
It's been pretty constant, and what's been variable, of course, and we see this often, is the placebo effect. That insight's been pretty important to us in how we approach PALISADE-4. We've obviously focused heavily on operational execution and placebo mitigation strategies, including additional site training, refinement of recruitment channels, optimization of some of the sites, careful monitoring, really, on how people execute the Public Speaking Challenge, and making sure, to the extent possible, that's consistent across sites. Another interesting area is what we've learned with some advanced analytics. One advantage of PALISADE program is sort of the richness of the dataset at this point. You know, for example, each Public Speaking Challenge generates a lot of data.
Now leveraging AI and machine learning tools across the historical dataset through PALISADE-1 through PALISADE-3 to look at patterns that might explain some of this variability. We're working with collaborators and using their proprietary AI advanced neural network approaches to really take a look at the vocal biomarkers. For example, from PALISADE-3, we recorded all the speeches in that, we're looking for behavioral indicators of anxiety and other potential covariates that might illuminate or influence the SUDS responses that we see. The goal is to better understand what are the drivers of both the treatment response and the placebo response, and determine whether any of those variables could somehow inform the statistical analysis plan we've got for PALISADE-4 prior to the database lock.
Yeah.
Again, trying to understand the totality of data across the studies, which is always important. In a sense, what we're doing is amplifying what we learned from the earlier studies by layering onto those studies sort of modern AI and ML approaches and analytics to really try to get a better sense, and refine how PALISADE-4 is being executed, and as we get to top-line results, how it would be analyzed.
Makes sense. Can you give 'cause a lot of this stuff sometimes I think when investors talk to CNS companies, like we hear a lot of companies saying the same things about like auditing data, AI, like, you know, monitoring. Are there any examples you can give about something you've learned from PALISADE-3 that informed something you tweaked with PALISADE-4 or something you did at like a site level?
Well, again, we know experience matters, for example, right? The more often that the
At the site level, you mean.
At the site level. The more often that the site is conducting the study, the more experience they have with studies, and particularly the nuances with this protocol really require rigorous adherence to the recipe. In our case here, you know, sometimes the number of subjects randomized at a site might make a difference. Again, that's the experience factor. Not just general industry experience, but actual experience conducting the study. That helps. I mean, also, we're not there yet, so we'll see whether the AI and ML does actually inform some adjustment to the SAP for PALISADE-4, but it could be something related to the nature of the subject themselves, their level of anxiety up front of the study.
There's different things that we're looking at and the partners are looking at, our collaborators are looking at.
Makes sense.
You're trying to really tease out subtle nuances associated-
Yeah, it's hard.
with the subjects, and the execution, and who's executing.
Right. Right, okay. At this point, like, how far along are we with PALISADE-4? Like, have most of the patients completed the study at this point? Like, you know, if you wanted to make a bigger change, like, is the study already kinda two-thirds baked? Do you know what I mean?
I think all I can tell you is our guidance is we're sticking with our guidance. By the end of the first half, we'll have the PAL-4 data set at the-
Okay
top line level.
Okay. Okay, makes sense. Do you wanna go back maybe into just, like, a little bit of the history of fasedienol? Like, you know, we've talked about this before, right? It's a unique product, right?
Mm-hmm
very different mechanism, intranasal, no detectable blood levels, phase II study at a couple different sites. Like, what was sort of all the biological and mechanistic work behind this drug, and, you know, what other sort of beyond the subjective clinical endpoints, like, what other data could you point us to that shows that this drug is actually active?
Yeah, It's an interesting class, as you know. It's a fascinating class. Fasedienol belongs to a class of compounds we call pherines, which as you know, they work very differently from traditional CNS drugs. Instead of circulating systemically, pherines are delivered in microgram level doses directly onto what we call chemosensory neurons in the nasal cavity. These neurons project to olfactory bulb and limbic regions in the brain that are involved in emotional regulation, including the amygdala. Because the compound acts locally through neural circuitry rather than through systemic absorption, we typically don't detect measurable blood levels, and so we've had to take a different look, kinda innovative approaches to understanding dosing from the very beginning, different than the typical PK/PD studies, you know, you and others, I think, are used to. We've conducted about 15.
Actually 15 phase I studies over the years to better understand dosing and biological activity, as well as a lot of in vitro work, calcium imaging work, and other work on nasal human and animal nasal mucosal cells. All trying to really understand receptor activity. Across the development program, including the phase two and the phase III studies, we've consistently seen evidence that supports biological activity. Many of these studies also collected autonomic data, showing physiologic effects of the administration. The dose selection was informed by all these studies, but in particular through studies using electrogram of nasal receptors or EGNR studies. These studies they measure the local depolarization of the nasal chemosensory receptors, which when the drug is administered again directly onto them, as a result of that, we get to dose at microgram levels.
We also used dosing response curves that resulted in selection of the current dose. At higher doses, we saw very minimal impact to the response of these receptors. More recently, we've done some studies using electrograms of the olfactory bulbs. The nasal chemosensory neurons project to the olfactory bulbs at the base of the brain. That neural circuitry relay then goes to different regions of the brain, depending upon the compound. For fasedienol, it's the amygdala. What we show there is that further down the cascade, when fasedienol is administered, in milliseconds you see these activated receptors, and then they send signals to the olfactory bulbs, and then further to the amygdala. Again, it's not the normal way because you're right, you cannot detect the drug in plasma.
The early clinical studies, the pharmacodynamic observations, the controlled trials, with the behavioral effects that we've seen all lean into that anxiolytic activity that you're trying to achieve at the end of the day.
Yeah. Okay. I guess going back to PALISADE-4, if the study works, what's next?
Well, we're on a conventional pathway for two adequate and well-controlled phase III studies. We round that out with a package that includes other ways to demonstrate the clinical meaningfulness and duration of effect of what we do in the clinical studies. There's a whole range of materials that the broad base of knowledge right now around this program over many years, it's just robust. We bring that whole package to the agency and, you know, we get some understanding of what it is that we think needs to comprise that submission. There may be some additional real world work we wanna do. There's a human factor study that would be completed. There's the various standard. The safety database should be pretty well rounded out.
Right now we're up approaching near the full 1,500 exposures, so that's been the benefit of the open label studies added onto what we already have done. Really, it's a combination of everything. Everything you typically would wanna put into your best advocacy for a registrational outcome that would be the very first ever acute treatment for social anxiety disorder in a population that's over 30 million fully in need, and nothing.
Mm-hmm
has been approved ever.
Yeah
for the acute treatment.
Yeah. Okay. I guess what's your level of confidence if PALISADE-4 is successful that PALISADE-2, which was stopped early, would count as another adequate and well-controlled positive trial from a regulatory perspective?
We think we're strong in it. I mean, bottom line, it was adequate and well controlled. It had a meaningful outcome. A lot of times it's even harder to achieve that outcome with fewer subjects than what was fully contemplated. We're confident that PALISADE-2 is a registration worthy study, and, you know, everything's always subject to FDA review. You know, the standard jargon. The reality is, yeah, we think that PALISADE-2 anchors an NDA along with a potentially successful PALISADE-4 and the rest of the work. The totality of data, the weight of evidence, the risk-benefit calculus, all those things go into an assessment, especially in this particular division. You know, we know you could have identical protocols often and not have the same outcomes.
Yeah
in the clinical side of things. I think that's important to realize about this particular division in the history of drugs in CNS, especially in neuropsych.
Yeah
You see different types of studies. You see studies that. You know, it's not a deal breaker if you don't necessarily hit your primary endpoint. We'll also have some look, Paul, over all the studies. Taking a look again with the power of AI and ML, and especially as the agency continues to emerge with focus on, you know, analytics from different ways of demonstrating efficacy. As long as you have, you know, not a red-faced argument about why you think you're seeing the drug effect or why it might have been obscured in the study design or in the demographics or other ways that you can tease out and articulate.
Yep. Yep. Okay. I guess maybe it's a hard question to answer, but, do you think you're gonna be able to leverage a learning from PALISADE-3 onto the analysis for PALISADE-4?
Well, what I can tell you is, there's a lot of power going into trying to generate that. I don't have the answer yet, but there's a lot to look at, and there's a lot of ways to look at it that are fundamentally different from even before the initiation of PALISADE-3, certainly before the initiation of the whole program. Definitely before the initiation of PALISADE-3, we come to a moment in time where not only are the technologies multiple different types of AI and ML approaches. For example, someone has a particular expertise around vocal biomarkers. Very important. We can learn from that from recording the speeches and the intake visits from PALISADE-3. I mean, I'm optimistic about the power and whether or not it has any impact on the SAP. We'll know before database lock.
We'd have to modify the current SAP with any adjustments that result from new learnings about covariates. What you're looking for is there any covariate that would have a fixed effect on the ANCOVA or the analysis of the covariates? It's certainly technically possible to find something. Whether we do or not, we'll see.
Yeah. Okay. Maybe, you know, I'm not trying to beat a dead horse, but always just trying to think about the different ways we can get at this. I mean, you have a good window from your seat onto who's going into these studies, right? Like, the baselines and the things like that. Like, at a high level, do the patients that have been enrolling in these trials look similar? Are they using similar sites? Like, anything you can kind of say there?
Well, again, you're looking for consistency. You're looking for ways to control variability. The way that we've enhanced or modified the I/E criteria up front of PALISADE-3, we think has made a difference the way that we get back and consistently are on top of the sites and refresh best practices. The benefit, while they're replicate in design, right? Three and four and as well as one and two, same study design, same Public Speaking Challenge, same SUDS endpoint is the primary. PALISADE-4 really isn't a full replicate in that it's got the cumulative learnings from the prior studies, in particular PALISADE-3, and that works operationally, and I think it may have an impact statistically in the methodology that we use to analyze it.
Again, you're trying to take not only best practices that come out of things that might have gone the wrong way, as well as things that might have gone the right way when you take a look back.
Yeah
All I can tell you is that to the extent there is anything observable, and there's anything that we have been able to implement as a result of getting feedback from sites, trying to understand, you know, sort of the operational nuances related to the study, we put them into play.
Yeah. Okay.
Can't guarantee any outcome. You know how that goes. The same part.
Yeah
what you gotta do though is make sure that you're learning. I do think it's different though, because at a minimum, as we moved into the statistical analysis of PALISADE-4, you're just approaching that differently with a different body of potential knowledge to impact it.
Yeah. Okay. For the FDA, do you need to hit on CGI as well as SUDS?
CGI is a secondary, so is PGIC.
Okay.
We like to see concordance. We like to see the really significant concordance that we saw in PALISADE-2 for both PGIC and CGI-I, 'cause the PGIC is the patients recording it, second speech versus first. CGI-I is the clinicians reporting how they thought the subject did second speech versus first. Then that generates into your group to group. It's a really important endpoint as a secondary and, you know, it's not a key secondary, so it's not necessarily gonna affect labeling if we get approved. We do see one of the things we like from the open label, and we hope to have some more more data related to that open label by the end of this half, this first half of the year, the PALISADE-3 open label.
It helps us take a look at utilization. It gives us a chance to take a look in the real-world setting, how that's impacting against our target product profile. How is it actually being used, and is it impacting people in their daily lives? That's a look under the Liebowitz scale, 'cause again, we're looking over time in that context, as opposed
Yeah
To the studs, which is minute by minute in the Public Speaking Challenge. Like we reported back in 2023 after PALISADE-1, that was very helpful to us in how we've been able to figure out how we think this drug's gonna make an impact if it gets over the line. In real-world, daily-life situations where hopefully what you achieve is increased confidence as people use the drug more often and avoid their stressors less, engage in their stressors more. The combination of the real-world experience with the controlled in-laboratory experience or in-clinic experience, that's all part of the totality of evidence that continues to drive our confidence in this drug.
Sorry, I muted myself. Okay. All make sense, Shawn. Anything else you wanna add before we wrap up here, either on PALISADE-4 or the rest of your pipeline?
We're excited about the rest of the pipeline. As you know, we've got five assets. All of them have achieved clinical success, in at least in a Phase II-A study. The lead, the depression program and the hot flashes program, what we hope to see is the IND cleared for the hot flashes program, by the end of the first half. That's a really exciting program, non-hormonal, non-systemic, rapid onset, you know, tailored to fit how women experience menopausal hot flashes in a market that obviously needs some help. We're excited about that program, and they're staged by the end of the fourth quarter, both those programs I hope are staged for further clinical development, either by us or with collaborators.
Okay. Great. Cash runway heading into this?
Runway into 2027. First half flip, PALISADE-4 top line, and right now we're sitting at runway into 2027. Good shape there.
Okay. Great. All right. Well, thank you. Best of luck, Shawn. Appreciate it.
All right, buddy. Thanks for having me again.