Greetings and welcome to the Marinus Pharmaceuticals confernce call to discuss the top-line results of the phase II tuberous sclerosis complex open label, at this time all participants are on listen-only mode. After speakers’ presentation there will be a question-and-answer session. To ask a question during the session, you will need to press star one on your telephone. Please be advised that todays’ conference is being recorded. If you require any further assistance, please press star zero. It is now my pleasure to introduce to your host, Sasha Damouni Ellis, Vice President of Corporate Affairs and Investor Relations. You may begin, Miss Damouni. Thank you. Please note that the slide presentation for today's call can be accessed within the webcast or by clicking the presentation link, which can be found on the Investors Events and Presentations page of the Marinus website. Before we begin, I would like to remind everyone that some of the statements made today are forward-looking statements under the securities laws. These statements are subject to certain risks and uncertainties, our actual results may differ materially. These risks and uncertainties are further described in today's press release and in our SEC filings. I would like to introduce our speakers for the day. From Marinus, Dr. Scott Braunstein, Chief Executive Officer, will present an overview of the top-line results for our open label phase II trial evaluating the safety and efficacy of adjunctive oral ganaxolone treatment in patients with seizures associated with tuberous sclerosis complex. Dr. Alex Aimetti, Vice President and Head of Scientific Affairs, will present the study design and data in further detail. Dr. Joe Hulihan, Chief Medical Officer, will provide a summary and next steps for our plans for the phase III TSC trial. With that, it is now my pleasure to introduce Dr. Scott Braunstein, Chief Executive Officer. Good afternoon, everyone, thanks for joining our call. Before I turn the presentation over to Alex, I wanted to highlight an exciting update on the regulatory front and provide an overview on the current TSC development program. We are pleased to share that the FDA has granted orphan drug designation to ganaxolone for the treatment in tuberous sclerosis complex or TSC. Another great accomplishment from our clinical and regulatory team. Moving along to the current TSC development program, as a reminder, we started the phase II trial last year during the first wave of the COVID pandemic and initially saw a high level of interest from a number of leading centers and investigators. It confirmed our belief that physicians and patients suffering from TSC were eager to explore a novel therapy. Our internal research has suggested that roughly a third of the TSC patient population suffers from highly refractory seizures, and our phase II study targeted this specific subset of patients. Following a very strong efficacy and encouraging safety signal in the CDKL5 deficiency disorder trial, which also focused on a highly refractory patient population, we had even greater reason to believe that ganaxolone would be an appropriate novel therapy for the TSC community. With that insight, at the end of the first quarter, we performed an interim analysis of the phase II TSC study and prepared this evaluation for a Type B meeting with the FDA. That analysis was driven primarily on safety, but we also prepared an efficacy summary. After reviewing the data set and seeing several encouraging signals in the study, including robust early responses in highly refractory patients, we began the initial planning for our phase III trial. Given the small size of the phase II study, the final efficacy results in the data readout did not fully mimic the interim analysis. That said, we were highly encouraged by the data set in its totality and in particular, the 50% responder signal in patients who were currently on the two most recently approved products in TSC, specifically Afinitor and EPIDIOLEX. As a reminder, this endpoint is critical for potential European approval. We are equally encouraged seeing the drug's impact on focal seizures, the major seizure type in TSC patients. Finally, we also believe that there are several important lessons learned from our phase II experience, and I will let Joe elaborate on these later in his prepared remarks. We continue to analyze the data and believe that we have the potential to better refine the phase III trial design. Last month, we had the opportunity to meet with our Scientific Advisory Board, including some of the most well-respected experts in TSC, and much of their advice has been incorporated into our phase III design. Our team, which includes Dr. Joe Hulihan, who, as a reminder, completed his epileptology fellowship at the Mayo Clinic. Our newest board member, Dr. Santiago Arroyo, who has tremendous drug development experience in the field of epilepsy, and Dr. Ian Miller, a highly respected pediatric epileptologist who joined the organization last year, also have added their thoughts on driving a phase III trial with the greatest odds of success. I have tremendous confidence in our team's ability to design and enroll a successful phase III trial and navigate this complex landscape. Now, I will hand the call over to Dr. Alex Aimetti, our VP and Head of Scientific Affairs. Alex will walk you through the study design and the phase II slide deck. Alex? Thank you, Scott. I'm pleased to share with you top-line data from our phase II open-label study, the CALM study, which evaluated ganaxolone in seizures associated with TSC, a rare genetic disorder that affects multiple organs. First, I wanted to briefly review the study design, which is outlined on slide three. This was an open-label study where all patients received ganaxolone in addition to their other concomitant anti-seizure medicines. Screened patients tracked their seizure frequency during the four-week baseline period. Following baseline, patients were titrated up on ganaxolone over four weeks, followed by an additional Eight-week maintenance period. The primary endpoint was the% change in TSC-associated seizure frequency during the 12-week treatment period relative to the four-week baseline period. TSC-associated seizures were defined as countable focal seizures, which are the most common seizure types in TSC, along with generalized seizures with a countable motor component. The study also consists of an ongoing expansion portion; however, data reported here will be focused on part A of the study. Looking at slide four, 24 patients were screened, with one discontinuing during the baseline period. 23 patients were dosed and included in the intent -to -treat in safety populations. Of these, 17 patients completed part A, and six patients discontinued during the treatment. Four patients discontinued due to an adverse event, one patient due to lack of efficacy, and one patient withdrew consent. Moving to slide five. Of the 23 enrolled patients, the median age was 11 years old, and approximately 65% were pediatric patients. Despite trialing a median of 3 prior and 3 concomitant anti-seizure medicines, enrolled patients experienced a median of 36.6 TSC-associated seizures per 28 days in the baseline period, highlighting the refractory nature of the seizures and the unmet need. This was despite newer generation anti-seizure medicines, including cannabidiol and everolimus, being the most common concomitant AEDs. Greater than 80% of the patients were on cannabidiol or everolimus, or both, during the study. On slide six, we show the patient-level changes in TSC-associated seizure frequency in the waterfall plot, demonstrating a median 16.6% reduction. The figure on the right highlights the median 16.6% reduction, along with a 95% confidence interval of the median. We previously announced back in March that a preliminary analysis of interim data supported plans to move to a phase III study. We show the preliminary efficacy analysis in 12 patients with at least four weeks of treatment data, which demonstrated a 47.6% median reduction in TSC-associated seizure frequency. Despite these numerical differences in the medians, the 95% confidence intervals are similar, highlighting the limitations of interpreting median values alone in small sample sizes. On slide seven, you can see the proportion of patients who achieved a greater than 50% reduction in TSC-associated seizures. Approximately 1/3 of the patients achieved a robust reduction in seizure frequency, consistent with data from the phase III study in CDKL5 deficiency disorder. Additionally, similar 50% response rates were observed in the highly refractory patients on concomitant newer generation anti-seizure medicines, including cannabidiol and everolimus. As a reminder, focal seizures are the predominant seizure types in TSC due to the pathology of disease. On slide eight, you can see that in this study, focal seizure types accounted for more than 80% of the TSC-associated seizures reported. Based on a preliminary statistical analysis in the 19 patients that reported focal seizure types, they experienced a median 25.2% improvement in focal seizure frequency compared to baseline. This finding is important as it provides preliminary evidence that ganaxolone may be effective in reducing the most common seizure types in TSC, which are different compared to the CDD patient population. Moving to safety on slide nine, ganaxolone was generally well-tolerated, with somnolence reported as the most common adverse event in 43.5% of the patients. Of the 10 patients that reported somnolence, nine were mild and one was moderate in severity. Further, four patients, or 17.4%, discontinued due to adverse events. Although we believe the safety profile is consistent with previous studies, we have identified ways to further improve the tolerability in this patient population, which Joe will cover in his remarks. In addition, there were four serious adverse events, of which only one, noted as seizures, was assessed by the PI as treatment-related. The other three SAEs were assessed by the PI as not treatment-related. In summary, this study enrolled a highly refractory patient population that previously tried or were concomitantly on multiple AEDs, including newer generation anti-seizure medicines. This likely reflects a patient population that has not been studied in previous TSC epilepsy trials. Of these patients, approximately one-third experienced a robust reduction in TSC-associated seizures, including those on concomitant cannabidiol or everolimus. The most commonly reported adverse event of somnolence was consistent with previous ganaxolone clinical experience. We believe the totality of safety and efficacy data from this phase II study supports progressing to phase III, with the added learnings from this study related to seizure classification and education, titration schedule adjustments, and other study design attributes. I will now turn the call over to Dr. Joseph Hulihan to provide his clinical thoughts on the study and how we plan to move forward. Joe? Thanks, Alex. There are several points I'd like to make about the phase II trial and why we think the results are encouraging. First, the study enrolled a highly refractory population, with patients having failed a median of three anti-seizure medications, including many who continued to experience frequent seizures despite treatment with one or both of the two most recently approved anti-seizure medications, Afinitor or EPIDIOLEX. Despite this, 30% experienced at least a 50% reduction in seizure frequency. Second, one of the most encouraging findings was that the data support the potential for ganaxolone to be effective for focal seizure types in TSC, including focal motor seizures, focal seizures with impaired awareness, and those that progress to tonic-clonic seizures. This finding is critical to the further development in this indication since, as Alex mentioned, over 80% of TSC-associated seizures are focal. Combined with data from the Marigold study, which supports that ganaxolone is also effective for the most disabling generalized motor seizures, we know that ganaxolone possesses a spectrum of activity to treat the multiple seizure types occurring in TSC. While we would've liked to see a more substantial median percent reduction in the seizures comprising the primary endpoint, the final result does not change the decision we made at the time of the interim analysis to proceed to a phase III trial in TSC. The relatively small N of the study contributed to the variability of results between the interim analysis and the final data set. However, the results fall within similar confidence intervals, both of which encompass the potential for clinically significant seizure reductions. Although the phase II study results only included 23 patients, we gained tremendous insights into the study population to inform our subsequent research. The phase II study also supported the safety of ganaxolone. We're pleased to see a safety profile consistent with previous studies. There's an expanding body of evidence on the drug's pharmacokinetic profile and its relationship to safety and efficacy that we gained from the database of the phase III trial in CDKL5 deficiency disorder, or CDD. These insights will help us design titration and dosing regimens, which will be critically important to the success of a phase III TSC study. In addition to our increased understanding of ganaxolone's pharmacokinetics, we've gained other valuable knowledge to guide study conduct in phase III. In particular, we continue to refine how best to educate and instruct parents and patients on the counting and classification of seizures. As with our CDD trial, we plan to engage the Epilepsy Study Consortium, a group of scientific investigators from academic medical research centers. They will validate the nature and type of seizure events, including review of clinical data such as EEG findings, video recordings of seizures, and other relevant information. I'd like to thank our Scientific Advisory Board, who provided invaluable guidance in developing these study procedures for phase III. We will also implement other features of trial design and conduct to ensure that the study is sensitive to detection of a treatment effect relevant to placebo. In summary, we're confident that a placebo-controlled trial will demonstrate the safety and efficacy of ganaxolone in treating TSC-associated seizures. We're continuing with our plan to initiate enrollment in a global phase III study in the fourth quarter. There remains a tremendous unmet need in treating the seizures of TSC, and we believe ganaxolone can be an important addition to the therapeutic options available to patients and their families. I'll now turn the call back to Sasha. Thanks, Joe. The presentation portion is now complete, and we will turn the call back to the operator for the Q&A session. As a reminder, to ask your question you will need to press star one on your telephone. To withdraw your question, press the dial key. Please standby while we compile the Q&A roster. First question comes from the line of Joon Lee. Your line is open. Hi, guys. Thanks for the questions and for the updates, and congrats on the data. Was there anything notable about the responders and non-responders, and do you have any strategies to enrich for responders in the phase III? I have a follow-on. Yeah, Joon. Hi, this is Joe. Thanks for the question. Yeah, these are the top-line results. It's a small sample. We don't want to make too much in terms of slicing and dicing, but we do intend to look at the characteristics of good responders or poor responders and keep looking at that. We haven't had much time to digest the data yet. We'll keep looking at it. One of the first things we looked at was whether patients were taking the newest meds, EPIDIOLEX or Afinitor. We were glad to see the responder rates were similar with those patients. Some of the variability is probably just due to what you see in epilepsy studies. There's a lot of variability. It's written about. Despite that, you're absolutely right. We're going to go back and look at the characteristics of those patients. Nothing really to highlight right now. Got it. Do TSC patients have seizures of other types that is not due to TSC? It seems like you wanted to make sure that these patients, when they report seizure, it is driven by TSC. Just curious if there are any other factors that can drive seizures that is not TSC, directly related to TSC. Your underlying assumptions of 150 patients that you're going to enroll in phase III, is that based on the final data in phase II or interim data? Thank you. No, it's actually based on external data and on our CDD data, a combination of both of those. We didn't have a placebo arm in this study, so it wasn't useful in terms of power calculations. Again, one thing it did tell us was about focal seizures and the potential efficacy in focal seizures, which we didn't have that many in the CDD study. As to whether they experience seizures of different etiologies, that's actually a really good question. I don't know. I think we assume if they have the genetic diagnosis of TSC, that the seizures are attributable to the TSC. That's far more likely. Got it. Thank you. Next question comes from the line of Alethia Young of Cantor Fitzgerald. Hey, guys. Thanks for taking my questions. Just a couple. One, it seems like with 23 people with seven sites, was there anything kind of due to just having a lot of sites and variability in measurement? I just was hoping maybe you could give us a little bit. Can you hear me? Yeah. Yeah. Okay, perfect. Yeah. All right, sir. I was also hoping that you guys could give me a little bit more clarity on why you might think that focal seizures, you saw better activity versus maybe the others, even though they're the majority. My last question is just, in light of kind of where we are with EPIDIOLEX and that rate they see in their label, what gives you the confidence to move into phase III versus doing maybe another bigger phase II to make sure that you can kind of repeat the signal? Thank you. I'll take the last one first, Alethia. We could do another phase II, but essentially, we think the efficacy signal, it doesn't dissuade us from moving straight to phase III. A phase II study would need to be pretty large. I think, besides looking at the efficacy signal, there's a lot of other stuff we got out of the study to inform the phase III study design. I think, on balance, I think we believe the best thing is to move straight to phase III, and we're confident that we'll see a result there, especially with the modifications that we see. In terms of the different seizure types, I think that was mostly just variability. Those seizure types, we actually saw good efficacy in the CDD study, and so I'd anticipate with a larger sample, we're going to see good efficacy in those seizure types in TSC. The focal seizures were much more common. The effect on the generalized seizures overall didn't influence the results that much. It did have some effect, obviously, but we think we'll see a better result in phase III. Thank you. In terms of sites, early on, I think we did have some getting sites started with COVID and so on. I think that may have been a factor. It's hard to know. There was some site variability. You expect that in any study, not so much in terms of results, but in terms of numbers. We're going to look into that more closely. It's hard to say. I think we'd need a larger number of sites to determine any patterns. Alethia, this is Scott. I'm going to just add on to Joe just a little bit. We saw a very strong signal early in the first half of the study. We talked about internally, should we stop the study, go right into phase III? We made the decision to continue to phase II because we wanted a bigger N. I think what we've come away with is really understanding a lot better this population, their concomitant meds, what to expect in terms of seizure types. Between the study, our interactions with our experts, our advisory board, and looking at this data, I feel confident that there are some additions to the protocol that we can make that's going to increase our likelihood of success, both on the efficacy and in the tolerability side. My standpoint, we would've loved to see that efficacy signal that we saw midway through the study, at the end of the study. At the same time, without question, I feel there's a real ability for us to add some lessons learned from the phase II into the phase III, and I am as confident today, particularly working with the team and our experts, about moving forward to phase III. Did you have a follow-on, Alethia? Yeah, just one quick follow-up. Yeah. In the first 12, were there any super responders that maybe shifted it, or was it just kind of well distributed across the whole thing? I think the patients who did well did very well. There were three groups, the patients that didn't do so well, the patients that did well, and then in the middle. Those ones in the middle are the ones I think that shifted the median. A small sample, a tiny shift in one way or the other can shift the median, and I don't think the median really reflects what we expect to see in phase III. I think that was maybe a bit of bad luck that the certain patients right around the middle gave us a low median, a lower median than we would've liked to see. I don't think it changes. There's still wide variability in the large confidence intervals that encompass a good result. The patients who did well, did well. That was reflected in the responder rate. I would add, Joe, we had good responders in the latter part of the study as well, just in smaller proportion relative to the first half. It wasn't like we had all the responders in the first half. We just happened to have 2/3 or three quarters respond, and then in the second half, a smaller percentage of responders. Mm-hmm. Great. Thank you. Next question comes from the line of Jay Olson. Your line is open. Hey. Thank you for taking our question. This is Matt on for Jay. We were just wondering, even prior to the interim, how does the top-line result of 16.6% compare to what you were originally at the outset of the study? Also, how does that compare to what you saw, which was around that 30% placebo-adjusted number in the phase III Marigold study for CDD? If you could just talk about the magnitude of 16.6% in terms of clinical meaningfulness stat, that would be great. Well, let me kick it off. Thanks for the question. Let me kick it off and then I'll turn it over to Joe. I think all our market research in both CDD and TSC have been very consistent. We understand that the clinical community is looking for a 20%-25% absolute delta when compared to placebo. We also understand that the clinical community wants a patient population that has a meaningful responder rate, and durability is particularly important to that. Certainly, by those terms, this number as our median, is shy of those expectations. Again, we do believe this has a lot to do with the sample size, and we unequivocally feel that we can make course corrections in the phase III to achieve that number, and that's exactly what we will be doing in the phase III study. Joe, do you want to add to that? Yeah. I think I caught your question about the comparison to CDD. We had a relatively low placebo rate there. I think part of that's the study population, part may be some of the ways we built some things into the study in terms of study design, which we're going to try to replicate in the TSC study. The placebo rate, I think we'll be able to get a favorable placebo rate and then also, pretty much what Scott said, we anticipate seeing a more clinically meaningful response in phase III with the larger sample. I think we suffered from a smaller sample in this study, in terms of the absolute median and the variability there. I hope that answers your question. If not, let me know. Oh, sure. Yeah. That makes perfect sense. I appreciate that. That's helpful. The one other thing we were wondering about is, Scott, we spoke a bit about this last week, is if you could just talk about the additional formulation work that you're doing. I don't know if the timing will work out to be implemented in phase III, but if you could talk about how that could potentially increase efficacy and be something available down the road, that would be really helpful. Thank you. Yeah. Happy to. One of the big things we saw in the Marigold study was that patients unequivocally did better in terms of seizure reduction based on plasma levels of ganaxolone. We certainly feel as though the current formulation in the CDD population has real limitations in terms of achieving proper blood levels. We're working on, right now, four internal programs, all of which we would look and hope would deliver higher blood concentrations across all patients, minimize variability, really be cognizant of Cmax and delivering too much drug up front. Equally important, work on a trough level that will be sustainable in patients. We think all those attributes plus the ability to dose titrate is going to be very important. That said, we have PK data in this study, and I think we really need to review that and understand how PK fully affected the TSC population, their underlying concomitant meds, and want to make sure that that signal is consistent in this population as well. My assumption is there's certainly going to be the same bioavailability issues in a TSC population. We're certainly going to review the PK data for this study as well. I think where we stand today for TSC, we are confident this formulation can drive a good outcome. I think in particular, we're really focusing the new formulation on our next indication, LGS, where we want to be able to drive even greater response rates and be able to dose titrate incrementally in that population. Thank you. Thanks for the question. Got it. Thank you. Next question comes from the line of Jason Butler of JMP Securities. It's Roy on for Jason. Thanks for taking our questions. Just had a couple follow-ups from the last three questions or so, actually. I guess in Marigold, you guys saw a pretty similar rate of responders, 50% responders to this study. The median response was higher. Was there a subpopulation of high responders in the CDD trial or are there other factors at play? For this phase II in TSC, how many patients titrated up to 600 mg in the trial? Thanks. Thanks for the question. That's interesting in the 50% response versus the median. I think overall, the fact that we had good responses among that subset, that contributes to the variability, but also gives us confidence. If things had been clustered around that 16%, I would've felt much differently, but they were spread out with some very good responses. I think the difference you see between the 50% responder rate was similar and the median was different, just has to do with the small sample and the variability, and a couple of patients can shift that median either way. I think that's why we really need a good, adequately powered study to see what the real kind of median result is. I'm sorry, the second part of your question, I forgot the second part. Joe, it was around reaching full doses in this study. Either 600 mg or 63 mg per kg per day. I'll take it, Joe. Yeah. We haven't looked at all that data as of yet, but there's no question that we saw some greater dosing variability in this study, and it's one of the issues that we're really focused on. We do believe that a new dosing titration paradigm in this phase III study would be very helpful. Joe, you want to maybe give a little bit more color on our thinking around a new dosing titration schedule for the phase III? Sure. Yeah. Our thoughts about dosing mainly come from the population PK work we did from the CDD study. It was clear from that patients are getting to therapeutic levels early in the titration, and that's supported in the CDD study. We saw efficacy early on. The efficacy in the first four weeks during titration was the same as during maintenance. Maybe we're going a little too fast, and we have an opportunity to slow the initial part of the titration a bit. With very small modifications in the titration schedule and adding one additional step to the titration, we think we can get patients at a therapeutic dose, with incremental improvement in tolerability, especially in terms of things like sedation. I think it's particularly important in TSC that the tolerability could be different from population to population. The somnolence was a bit higher in the TSC population, and we're looking into why that might have been. We think that adjusting the titration schedule is going to be a sound move for the phase III study. We'll implement it there and modify the protocol accordingly. Okay, great. That's helpful. Was there anything in the phase II that maybe made you think differently about the patient population to target for phase III, or is the phase III population going to be pretty much like the phase II? Yeah. No, actually, I think we're going to go into the phase III with pretty much the same patient population. I think it did give us confidence that we're getting a representative part of the population in terms of concomitant medications. A large percent of the patients were on Afinitor, a lot of patients on EPIDIOLEX, and so, if that's the population, I think that's what we need to reflect in phase III. I think it's pretty much the same as the phase II population. Perfect. Thank you. Your next question comes from the line of Andrew Tsai from Jefferies. Okay, great. Thanks, good afternoon. As we think about the phase III, I think you mentioned a meaningful placebo-adjusted rate would be maybe 20%, 25%. As we think about handicapping phase III, what drug could show, what placebo could show, I guess, how low of a placebo rate do you think you could get in phase III? That's my first question, I have a follow-up, please. Yeah, sure. I think our assumption is somewhere around 10%-15% placebo rate, although that could be variable. Again, it was 7% in the CDD study. Really the thing we're interested in is obviously the net over placebo. We've powered it conservatively for phase III on the lower end of that desired range. I think regardless of the placebo rate. One of the things is our patient expectations about response, those need to be managed. And the patients coming into the trial in terms of their seizure counts and the stability of their seizure counts, pretty much straightforward clinical trial measures. I think we'd shoot for something around 10%-15% placebo rate. Yeah. Andrew, let me just jump in and add, I think we don't really see the phase III EPIDIOLEX study as the right comparator. We've learned from many of our KOLs, and experts in the field, the interest in taking CDD was extremely high, and we certainly believe that had a major element in driving placebo in their phase III study. We also believe that working with the Epilepsy Consortium and really targeting seizure types will be critically important for controlling the placebo rate. At least one of our SAB members, you may all be able to guess, really believes that that interaction with the Epilepsy Consortium in the phase III CDD study was critical in driving a low placebo rate. I think we feel as though unequivocally, we can drive a low placebo rate, as Joe mentioned, low double digits. Clearly equally important will be driving the efficacy from a median perspective, into the mid-30s or high 30s. We think with those adjustments, we have a real shot of doing that and with some additional learning from the phase II trial in terms of patient enrollment, seizure criteria, et cetera, all going to be critical for us to achieve that threshold. Right. Okay. Thanks. As I think about different angles of your data set in terms of potential differentiation, you did mention 25% focal seizure reductions. I'm curious what EPIDIOLEX, Afinitor, what they show in percent seizure reduction so we can kind of compare more apples to apples. Secondly, safety discontinuation rates, I believe, were about 17% or maybe four patients. I guess, what are the discontinuation AE rates for the existing drugs? Maybe talk a little bit about how maybe ganaxolone safety profile could be differentiated, because when I've looked at the label. Right. they do have some safety issues. Maybe talk about that. Yeah. Sure. I don't think we've seen, I'll ask Alex to correct me if I'm wrong. I don't think we've seen a breakdown of seizure types responses in the EPIDIOLEX study. That has not been part of the publication. Afinitor, actually, their study was interesting in that it was all intended to be focal seizures. Now they had a very broad definition of focal seizures. I think that it was all potentially focal seizures. There are some investigators that postulate that all seizures in TSC are focal regardless of what they look like. There's some debate about that, but so that was the approach. In terms of the discontinuation rates, I know that the absolute AE rates in EPIDIOLEX in terms of somnolence were just a little bit higher than we saw in the CDD study. In terms of the phase II TSC study that we're reporting, I think that, again, it's a small sample. I think adjusting the titration rate could potentially influence that for the sedation and the kind of maybe more dose-related side effects. I don't know. Alex. Joe, maybe I'll jump in. Yeah. Let me jump in and add one or two comments. I think, Andrew, importantly, just a reminder, the Afinitor study that was completed several years ago, EPIDIOLEX was not approved in the market, that was add on to older anti-epileptic therapies. I think it's very important to remind everyone on the call that EPIDIOLEX had a specific contraindication around elevating Afinitor levels. The entire phase III program for EPIDIOLEX was on addition to older therapies and not Afinitor. We really felt in the real-world scenario, we'll have lots of patients on Afinitor. This is the first study that I'm aware of that was conducted on top of Afinitor in TSC patients, and obviously one of the first studies that's now been conducted on top of EPIDIOLEX. We understood that we were taking the most refractory patients in this study. Certainly, we were expecting an incrementally lower efficacy rate. We are very happy to see great activity or very good activity on focal seizures. Really a very good responder rate. As I mentioned, although the median seizure reduction number was on the lower side, we certainly think it's within the normal confidence intervals. More importantly, we have several tools at our disposal that we think will really bump that number. I think we're dealing with unchartered waters here, and maybe going back to Alethia's comment, we've learned a tremendous amount in the phase II, and I think we're very well prepared for the phase III. Okay. Very good. Maybe a very last kind of nuanced question. I could be wrong with this, but I think this trial is like 12 weeks long, including a four-week, whereas phase III might be 16 weeks long. Yes. If so, if that's true, in the maintenance period, do you think having a four-week longer, I suppose, maintenance period could drive efficacy better as patients stay on the drug longer? Do you see a benefit early on? Yeah. Thanks. Yeah. Sure. No, I think we did see a benefit early on, and it was stable, at least in the CDD study, through the maintenance period. We haven't really looked at that yet in terms of this study, but it was brief. I think what we anticipate is continued stability of efficacy in the maintenance period. In CDD, it was really when the patients rolled over into open label that we started to see, for the patients who stayed in the open label, some incremental efficacy. We're going to be presenting that at the AES meeting this year. Within the double blind, I think stability of the response is really what we anticipate seeing. Right. Okay, thanks. Joe, maybe just worth commenting, as you've reminded me from a regulatory perspective, the agency would like to see 12 weeks of maintenance therapy. That's the reason from the change from phase II to phase III. Next question comes from the line of Brian Skorney. The line is open. Hi, thank you so much. This is Jack Allen dialing in for Brian. Our question again is on the phase III design. I know we've touched on a number of the topics here, but I was hoping you could provide a little bit more detail on the titration and the changes you're looking to make to the titration method and how that will play a role in, I guess, the difference in the maintenance length as well. I guess there's going to be a four-week longer maintenance period, but if the titration is also longer, is the maintenance still net the same, or is there going to be a longer maintenance period here? Thank you so much. Sure. Right now, we haven't really completely finalized. I can say conceptually that if we add just one additional step, really, the drug's been titrated in equal amounts, just divided in four and giving the patients, starting them at a quarter of the dose and going up by another quarter in three increments. There's nothing behind that, really, except giving an even amount each week or increasing by an even amount each week. What we plan to do in concept is take those early doses and cut them back, maybe by a third or even by a half for the first two or three weeks, and then go up. Even if we do that, we're able to titrate it and never exceed an actual jump in dose that's greater than what's in the current titration schedule. The population PK suggests that patients achieve the levels early and that those later increments that are a bit bigger aren't going to bump the blood level as much. It's really the beginning part of the titration where the blood level starts to increase. If we get them at that level more slowly, and then at the back end, go back to the rates of titration that we're using now, we think that'll improve tolerability quite a bit without compromising efficacy. The other piece of that is, you asked about the length of the titration. That's something we're still working out. Right now in the current titration schedule for four weeks, the patients are at the top dose for one week before they enter maintenance. Potentially, they could enter maintenance immediately after getting to the target dose, we wouldn't need to increase the length of it. That's stuff we're still working out. Not final yet. Hey, Jack, let me just add, I think what's important here, just a reminder and we own this, but we did inherit the titration schedule from the prior management team and their decisions about how to titrate. Certainly, one of the first questions we asked the investigators is how was titration schedule in the CDD trial going? Generally, as we saw in the phase III results, the tolerability was not an issue in CDD. There's always been a question in other disease states whether tolerability could be a bigger issue, and as we were preparing for the NDA and we saw this population PK data, I think it really made us think that there could be a smarter and more scientifically sound titration schedule that really aligned with what we believe to be most patients' absorption. We understand that most patients are absorbing less than the maximal dose, and we want to make sure the titration schedule aligns much more with the patient's ability to absorb the drug, get plasma concentrations that will allow good titration, and we ultimately think that will drive best tolerability and hopefully, incremental efficacy with a good titration schedule. We certainly saw more variability in dosing in this study, and we think unequivocally we can fix that one problem. Thanks for the question. We have just a few more minutes. Operator, can we go to the next question? Next question comes from the line of Peyton Bohnsack. The line is open. Hi, this is Peyton on for Joe. We were wondering if you could talk a little bit briefly about what the compliance rate was like between maybe the interim analysis and the final analysis, and then also if you could go into any more details about the improved electronic seizure diary. Thank you. Sure. In terms of the first part, in terms of diary compliance, we don't have the data for that in the top line. We haven't yet gotten a full set of final data tables, and we'll know more about that when we look at those. In terms of the diaries, we're using a new platform for the diaries, different than the one we used for the CDD study or even for the phase II TSC study. The main advantages of the platform we're using will be that it can be on the patient's own phone or personal device. They don't have to carry an additional device, and that actually is a pretty big deal in terms of the seizure diaries. They can use it on their own device. The other thing is, in terms of diary compliance, it was good in CDD. We want to do everything we can to maintain that in the TSC study. If the patient doesn't complete the diary entry for a given day, there's an immediate notification from the central vendor for the seizure diaries. That increases the completion rate. We want to do everything we can to ensure the accuracy and the completeness of the diaries. Those are the main distinctions. Yeah. If I was to summarize, the easier the diaries, the higher accuracy, and the higher accuracy, the better off we are. I think our diaries from CDD were highly accurate. There were over 10,000 seizure collections. We're asking patients to do a lot and family members to do a lot, and making their lives incrementally easier with a one device and a little bit more friendly device, our hope is that will only increase accuracy on the margin. All right. Thank you. That's very helpful. Next question comes from the line of Marc Goodman of SVB Leerink. Hi, this is Rudy on the line for Marc. Thanks for taking my question. The reduction in focal seizures of 25% was stronger versus the overall data set of 17%. It seems to be in line with your prior studies in refractory focal onset seizures in the 20%-25% range. I was just curious, how should we think about ganaxolone's opportunity in TSC versus the broader indication of focal onset seizures? That's a good question. I think, really looking at the data from the focal onset study, there was a phase II and a phase III. The phase II, we saw a separation from placebo and a quote-unquote positive study. In phase III, there was not a statistically significant result. It's an interesting comparison. I think those studies in total support efficacy in focal epilepsy, particularly three times a day dosing. The phase III used two times a day dosing, while the phase II dosed three times a day. That seems to have been the key. Our current dosing with the suspension is three times a day. We feel with this formulation, that's necessary. I think, if anything, the dose is slightly higher than in the phase II focal onset study. If anything, I think we can anticipate an improvement in terms of the seizure rates. The reductions in the TSC study. We're optimistic about that. Does that answer your question? Yeah, sure. That's very helpful. Thank you. Okay, sure. Next question comes from the line of Douglas Tsao of H.C. Wainwright & Company. Hi, good afternoon. Thanks for taking the questions. Just given the fact it was an open label study, I'm just curious if you had a chance to look at patient characteristics and did the mix of patients change over time? The reason I ask is because oftentimes with an open label study, when investigators have positive experience early going, they tend to start to sort of change a little bit in the patient population or the types of patients they're targeting in the study. Yeah, that's right. Yeah, it's a good question. Again, with the top-line results, we haven't really dug into the patient characteristics. We've only had a few days to digest the data. That's something we'll definitely do, including what you mentioned, looking at the patients who came in early versus the patients who came in late. Was there some difference in the severity of the epilepsy or the concomitant medications? We'll look at all of that, but we haven't yet. Okay. Just as a follow-up, I think I heard you say, but I am just curious, have you had a chance to look, for patients who were responding early on, and I guess for patients early in the study, were they maintaining their responses? It was just simply that you had a higher proportion of non-responders in the sort of second group of patients that were enrolled or the later enrolled patients? Thank you. Yeah. Again, I'm sorry. We haven't really looked at those detailed characteristics yet. I think it's overall that the patients coming in later. When we did that early interim cut, yeah, the patients who came in early did better than the patients who came in later. It's just, again, with a small sample, it's the luck of the draw. We did have some patients, as Scott mentioned, who came in later who did well. There were more patients who came in later with the iffier responses than the patients who came in early. Okay, great. Douglas, I think you may be asking too, did we see a lot of early high reductions in seizure counts that disappeared over time? We did not, which actually, it's a great question because we really think it was the tolerability issue and the titration issue early on that had the majority of the impact on a lower response rate than we've seen in other studies. Okay. That gives us a lot of confidence moving forward that if we make that adjustment, we can have a meaningful impact. For the patients who were responding, they showed good durability, I guess, is sort of where I was getting at, and it sounds like that was the case. Correct. I think that's fair. It's unusual that we see someone who has a 10% or 20% response go up to 40% at the end of the study. That's relatively unusual. They tend to respond within the first three to four weeks and maintain that response or not have a response early and really stay unchanged. Okay, great. Thank you so much. Look forward to some of the data. Thank you. There are no further questions at this time. Presenters, you may continue. Thank you, operator. Thanks everyone for dialing in. Thanks for the questions. Just as a reminder, we got this data pretty late on Friday. Joe, Alex, Dr. Miller, who's joined the team, just done a tremendous job to not only put together the top-line results for us internally and for you, but I think we have some really important data points that we've recognized here and some exciting updates that we can make over the coming weeks. We plan on sharing those in greater detail with you at our Analyst Day in October. Thanks everyone for dialing in. Appreciate it and have a good day. This concludes today's conference call. Thank you for participating. You may now disconnect.