We're going to get started with our next session. I'm Andrew Tsai, Senior Biotech Analyst at Jefferies, and it's my pleasure to have Remy Lutheringer joining me today, CEO of Minerva. Welcome, Remy.

Thank you, Andrew. Thank you for the invitation.

Of course. And so maybe to help the audience, those who are less familiar with the Minerva story, would you mind spending a brief couple of minutes talking about Roll the Peridone, your lead asset, what you're trying to achieve, and what stage you are in the program, and milestones over the next six to 18 months could be helpful.

Yeah, thank you for this question. So clearly what is the overall objective of Minerva when we put together Minerva was to really address unmet medical needs in the space of psychiatry and neurology, but mostly psychiatry. And rolyperidone is a perfect example, yes, because the objective of rolyperidone is to treat negative symptoms in patients suffering from schizophrenia. And as you know, this is probably the highest medical need in patients with schizophrenia because we have a lot of treatment for positive symptoms as the antipsychotics. We have also no newer mechanism of actions. But I mean, all these molecules are definitely approved to treat positive symptoms. So what we are trying to address are negative symptoms, and what the literature is showing you, yes, is that why these patients are not functioning well, why these patients do not have a job, is because they have negative symptoms. So this is what we are trying to address. Now, the question is how do you address this, yes, because in the past a lot of people have tried because it's an unmet medical need. Most of the people have tried to put this, for example, on top of antipsychotics. And I think it's fair to say that this is somehow a graveyard. So what we have decided to do is to think more about how the disease is looking like, how these patients are looking like, and how heterogeneous the disease is, and what do we want to achieve. First of all, I think it's now very clear to KOLs. Also clear, I think, now, to the FDA, and this was one of the points of discussions with the FDA, is about the disease is heterogeneous, and not each patient needs the same treatment. So this is the first point. The second point is that negative symptoms, everybody is speaking about negative symptoms, but let me just give you a little bit more clarity. So you have two types of negative symptoms. We have the first type, which are the most important one, because these are the impairing negative symptoms, or the primary or disease-related negative symptoms. The second one are what we call the secondary negative symptoms, which can be, for example, induced by an antipsychotic, because it's very well known. If you're going to block dopamine in the brain of patients, you definitely are doing something positive on the positive symptoms, but you're worsening negative symptoms, and all the meta-analyses have shown this. So first of all, it's important between primary or disease-related negative symptoms and secondary negative symptoms. The second notion to keep in mind, which I think is key, is about a pseudo-improvement of negative symptoms. And it's true when you think about a patient who has an acute relapse of positive symptoms. He's not sleeping. You ask me if my sleep is adjusted. Definitely, I mean, when the sleep is adjusted, you're probably functioning better. But these patients are not sleeping, they're agitated, they have hallucinations, they have delusions. So because they have this acute episode of positive symptoms, indeed, I mean, they're worsening during these episodes of negative symptoms. But when you bring them down from this acute episode, you bring them just back to the baseline of negative symptoms. So this is a pseudo-improvement. And I think it's important to keep this in mind. So what we are trying to do here, and this is explaining why we designed the studies in the way we have designed them, is to first demonstrate that our drug is improving primary negative symptoms. And the only way to do this is treatment versus placebo in monotherapy without having the confounding factor of antipsychotics or whatever, any treatment you will put in addition to your drug and your experiment. So this is what we are doing. I think it is a very large population we are addressing here, and if needed, I can show a slide. But I know that this is a paradigm shift. The FDA even is calling this a paradigm shift, and here is what we have to do here, is to really educate, to explain. But this is what we are doing. It does not mean that when the drug is approved, you cannot expand. expand. Knowing about the stage, we are now running a confirmatory trial. We have approved or we have discussed in length with the FDA and we came to a consensus here. And we have started this study in March of this year after the financing, which took place in October, a pipe financing of $200 million, and the readout of the efficacy part, which is after 12 weeks, will be second half of next year, 2027, and the readout of phase B of the study, which is more focusing on relapses of positive symptoms, just to check if our drug compared to antipsychotics is having a very similar relapse rate in our targeted patient population. is similar, and this will be done in the second half of 2028.

Okay, very good, and thank you for the introduction. So, in terms of the market opportunity, I think that's one thing you're about to reference. I think there's maybe 3 million schizophrenia patients in the U.S. How many of them do have negative schizophrenia?

Yeah, so, you know, it's obviously variable from one study to the other, but I think a real consensus is that at minimum 60% of the patients with a diagnostic of schizophrenia have negative symptoms. Now you have to be a little bit careful because schizophrenia is a very dynamic disease, and so when you think about patients having negative symptoms over the course of the disease, you can come to the conclusion that quasi 100% of the patients have negative symptoms at a certain stage. But if you want to go with very good publication, recent publications, you can speak about 60%.

Right. And so let's just say, big picture, if Roloperidone were to succeed in phase three, approved, as we think about the eventual use case, you would find patients who have elevated negative symptoms, take them off an antipsychotic, dose this as a monotherapy. would it be acutely and then or do they dose chronically before they go back to antithesis can you would just walk through yeah so I know that this is a he's an important topic is and and and and I would say we we have made a lot of progress I mean this is a scientific medical community has made a lot of

progress between this notion of continuous treatment with a non deep psychotic versus alternative treatment or a treatment where you're using the antipsychotics mostly to treat an acute episode of psychosis and I think the space is really moving, asking the right questions, doing the right research and more and more I think the idea is here to come up with an approach where you're involving the patient and the prescriber, even the caregiver as a family, and not just to say, okay, antipsychotics are good for you, keep it forever, and we hope for the best. I think the things are really evolving, because first of all, patients have a lot of insight. So let's be clear, everybody who has worked with these patients, they see that these patients are really having a lot of insight. So what will happen is that that if, I mean, when the drug is approved, you know, the prescriber and the patients will probably ask a lot for going on Rodiperidone for a very simple reason. The first reason is very clear, is that let's speak about side effects. So the existing therapies have quite a lot of side effects. I'm not saying that they are not useful and they are not important to treat an acute episode of psychosis. And this is the reason why antipsychotics have been developed and why these treatments are approved. The second reason why these treatments are approved is, and this is really the main reason, is because they are reducing the number of relapses and the number of hospitalization. But again, they have a lot of side effects. and even the newer antipsychotics with different mechanisms of actions as, you know, the existing dopamine-blocking molecules. So really, I think the patients will really ask for this because they want to have less side effects, and they are well aware that it's not taking an antipsychotic continuously, which will help them to improve on a functional level because you have to improve negative symptoms to function better. So I think patients will definitely ask for it. And again, because there is no more discussion between prescriber and patient, this will help a lot. I think for the prescriber it will be also important because I have been among these guys. And when the guidance is telling you do all what you can to keep someone in an antipsychotic to reduce hospitalization and relapses, the day where we have the data in hand about the fact that our drug is protecting as much as antipsychotics against relapses, this is a paradigm shift. Here, the prescriber and the patient can definitely think about droly-peridone immunotherapy. Again, I'm not saying that antipsychotics are not useful. They will be useful in case a patient has a relapse of positive symptoms. So if you allow me to take an image, and I take this image from someone who is a well-known KOL, you know, schizophrenia is like having an infection, yes? So you have an infection, you have to treat fever. Treating fever is like treating positive symptoms in schizophrenia, but this does not mean that you are treating the disease and the functional impairment of the patient.

And so to be clear, if a patient is controlled for positive symptoms but the negative symptoms are there, take them off, use the Roliparidone, control for that. If the positive symptoms return back, then swap. Is that the strategy?

Yeah, swap until the patient is back to his baseline. Again, speaking about this notion about bringing them to the baseline. And if, I mean, they have still impairing negative symptoms, you can think again about Roliparidone in monotherapy.

Okay, very clear. And so let's talk about your phase three that you started, data, 12-week data, second half, 2027. And so maybe describe the kind of efficacy benefit you saw in the prior phase two and phase three studies. What kind of absolute, what were like the baseline? You're using an endpoint called MARDR. And so what kind of reduction, score reduction did you see? And what is clinically meaningful? in terms of the score reduction.

So clinically meaningfulness is a complete debate, and I will come to this in a second. But in order to include a patient in our study, and this comes to the 60% of the patients we were discussing before, they need to have stable positive symptoms over the last six months. Obviously in clinical practice, it will be stability of positive symptoms over a longer period of time. It's not six months, but for the study purposes, it's six months. This is what you have to check, and you have to check if these patients have impairing negative symptoms. In the study, it is 20 points or more on the PANS negative score, but in clinical practice it will be this patient is not really bothered anymore with his positive symptoms, they are stable, and is not functioning well because he has negative symptoms impairing him. So this is the patient population. So I think I missed your second point.

Like what kind of score reduction is meaningful?

So when you're looking to the baseline value of this patient's head in terms of negative symptoms, they were in the range of 25, 26 points. It's well described in the literature that when you're above 18 points of negative symptoms, you have an impairment. You have real impairment which are not allowing you to function well. So here, I mean, these patients have quite a high level of negative symptoms. So what we have seen, and again, keep in mind that, I mean, here, we are not bringing down someone from an acute episode. So we have not the pseudo-improvement of negative symptoms. But so what we have seen during the first 12 weeks versus placebo, it's around the four points improvement. And what is also important, and we discussed this recently, we had obviously an open label extension in the two studies. When you're looking to where the patient is after one year of treatment, you have improvements which are going to eight, nine points of improvement. So this is what you see with our drug. So again, purely improvement of primary negative symptoms, not a pseudo effect. But what matters is really functioning, and as you know, we are using a scale which is called PSP, which is really looking to everyday life activities of these patients. Are they able to take a shower? Are they able to take off cleaning their rooms? These kind of things is what you're measuring here. And in the two studies, we have seen improvements of more than seven points, six, seven points of this scale. And the literature is very clear. When you have improvements which are going beyond six points, this is clinically meaningful. Now, your point about clinically meaningfulness is difficult to answer for a very simple reason. If you have no reference, you cannot really decide what is clinically meaningful. But what I can tell you is that we have done a responder analysis. We have done anchor analysis. We have anchored the primary endpoint to the CGIS. and it is very well described that one-point improvement of CJIS is clinically meaningful. When you're doing all this analysis, all is in favor. In the two studies, by the way, yes, in the CESAR-3, in those phase 2b study, and in the phase 3, all this is showing a highly, significantly improvement compared to placebo for 64 milligram when you're doing this kind of analysis. So it was one of the questions of the FDA, and I think we really addressed this now and what ZFJ is asking is to see a p-value and not anything else.

I see. As long as you hit stat-sig, basically. And so that goes to my next question, the powering assumption to this phase re-study. You're 90 percent powered to detect what kind of placebo-adjusted change on the market.

So the study is even overpowered. It's more than 90% power, but basically the assumptions we are using here is a 1.8 difference between placebo and treatment. We are using the dropout rate because you have to integrate this in your power calculation. We use around 30% dropout, which is in line with what we have seen in the two previous and in terms of standard deviation we went above what we have seen in the two previous studies so we have really taken a standard deviation which is a more or less one point more than than what we had in the previous studies in order to come up with with a number of patients that we have in our study so again it's more than 90% powered with 380 patients but so this is all the assumptions we have here.

I see. And so hitting STAT-SIG, what is that threshold? How low can you go on the placebo adjusted change to still be STAT-SIG?

So we have obviously analyzed the worst case scenario. So I think with this, how to say, set up powering of the study with something like one point, we still have a p-value. And this is a worst-case scenario in terms of statistics, yes, but definitely I don't think that this will be the result because, as you know, we have, and as we discussed, we have implemented even some additional layers in order to maximize the probability of success of the study compared to the two previous studies.

I see. And so when I looked at your phase two and phase three prior studies, the absolute drug effect was consistent about minus four point reduction placebo one of the studies was 1.6 reduction the other one 3.5 yeah so you know placebo is a variable here so um where do you think placebo falls and why in this so i think it falls around 1.8 or something like this yes i mean in the middle between the first study and the second study it's always always uh clear that the first study where there is no expectation because there was no approved drug because nobody knew about this.

Everybody was staying very objective. I think this is the best case scenario if you have a drug which is working. And this is exactly what we had here. When you're moving up towards to a second study, first of all, a larger study, more clinical sites, more patients, more variability, expectation, which is high, a drug which is extremely well-tolerated, so you cannot discriminate the drug from placebo. You know, I started my career with Risperdal as a PI, and Risperdal, I could tell you after two days who is treated and who is not treated just based on side effects. In our case, you cannot do this. So it's not unusual. It's very well described that indeed, I mean, placebo is picking after. There is another aspect which was important in our study why placebo picked up is that correctly so, yes, the FDA asked us at the end of phase two meeting before running the phase three that we are interacting much more with the patients than in the previous study because, yes, it's a paradigm shift because we are taking them off antipsychotics. It's monotherapy. And so we interacted much more with the patients. And clearly, this is what I call the nursing effect. You know, think one second. Here you have patients who have negative symptoms, which means that they are not having a lot of social interactions, they are not stimulated, and suddenly you put them in a study and they have a lot of social interaction. So this is an additional layer who really was doing not the best effect on placebo, was improving placebo. In this ongoing study, in this confirmatory study, we are minimizing the interaction with the patient. We also keep in mind that, I mean, here we have only one dose versus placebo, and so because the expectation to get active drug goes down, placebo will go down. It's completely described in the literature. And in addition, I think we took a lot of care to select the right sites, to train the sites. During the study, we are monitoring the performance of the scoring, not only at the level of the clinical site, but at the level of each person doing scoring, because you have always two, three, four people scoring in each site. So we have a lot of ingredients here to keep normally the four points improvement, because if you do this in two different studies and you have the same result, I think this will be reproduced. and, I mean, the placebo will be more under control. So I think the 1.8, you know, hypothesis here is fair. And last but not least, in the two studies, we had a functional improvement, yes, which was significant, yes, which was even, I put it in brackets, clinically meaningful. So this is what we have, yeah.

And is there any reason to believe why the drug could be even stronger than four points? or maybe the root of my question is is this population more enriched in any way to help you? Are the baseline scores higher and if they're higher does that help reduce more?

I think it's premature to do this hypothesis. The only thing I can tell you is that what we have seen in the two previous studies is when you analyze your data per quartiles and you go with a younger population elderly population You know, we have seen some bigger improvements in the younger patient population because here we're speaking about chronicity of the disease, which is pointing to the fact that if you want to really change the course of negative symptoms or even to bring back patients to normal, you'd rather start sooner than later. So this is what I think of the data. The level of negative symptoms, we did the analysis, but there are not really big, big differences depending on the level. What is important, but this has nothing to see with the response to rolyperidin, we have just published a paper showing that if at baseline your level of negative symptoms is high, you are less keen to relapse in terms of positive symptoms. If your level is lower and if you have more positive symptoms, obviously you're more keen to relapse during the study. So our patient population is definitely less keen to relapse than the overall patient population suffering from schizophrenia, which probably explains a lot why we had a very low relapse rate in our two studies.

And how low of a relapse rate percentage-wise?

So in one study, it was 9%. In the other one, it was 13 or 14%. So compared to what you have in these randomized withdrawal studies with antipsychotics, where you are at 20 to 35%, it's quite low. But again, I'm clear about it. I mean, we are targeting this patient population is less keen to relapse because they have negative symptoms. Right.

And the baseline score, I think you said the prior studies were around 25 to 30, and you expect around the same?

Yeah, interesting enough, you know, we have recruited quasi 900 patients, and between the two studies, it was very consistent. We used different clinical sites. So I think two conclusions from these patients exist, because, I mean, the PIs are able to pick them up. There is another company who did a very similar study. It was Lundbeck. You can speak about it because they have published the data. They used exactly our study design, and they picked also up the patients that were able to recruit the patients. So this population is definitely out there and with very similar negative symptoms level. But what is also important, the positive symptom level was exactly the same between the two studies, and it was at around 14, 15 points. So very low, yes, because as you know, if you have no symptoms on the PAN scale, you have seven points because zero symptoms is one, and there are seven items. So 14 points is very low levels of positive symptoms. And the good news here is part of relapses, this patient stayed very stable in terms of this very low level of positive symptoms. even in the second study we had a p-value of improvement of positive symptoms it's not a lot because they're already very low but just to say that it's important these patients stay completely stable over one year Great, and so as you're enrolling again data, second half, 2027 is there a possibility of an interim analysis for you to just double check you've got everything correct upsize or continue as planned for instance? So it's not an interim analysis. We have no interim analysis plan for fertility or for, but what you're obviously doing, you're always assessing your data moving forward or the parameters of, you know, redapse rate. And I'm sorry, the percentage of dropouts, excuse me, not redapse rate. Variability of the data. All this, you're checking this continuously because this is just good practice.

Yes.

And so, again, data, 12-week data, second half 2027 is there a possibility you file actually as you wait for patients for you to accrue data in the relapse obviously this would be the perfect situation and and and and I think the the agency is completely open to this and we we are engaged on a regular basis with the agency and and definitely we will interact with the agency when we have the 12 week primary efficacy end point. Something which is important and which is probably speaking in favor of having the FDA engaged at this stage is that the FDA has confirmed several times that phase B, yes, I mean this relapse part, it has nothing to see with efficacy, it has to see with safety. So I think there are a lot of arguments to re-engage at the level of top-line results with a phase A and keep in mind that when the phase A is completed we already will have a lot of patients who have completed phase B and even if it will be blinded if you know the relapse rate is at whatever 15 percent just to take a number you are reassured that I mean whatever rodiparidon or antipsychotics you are you are below the magic 20 percent or 25 percent of what you have in the randomized withdrawal studies currently with antipsychotics yes okay very good and so these patients are re-randomized to roll a period on or three antipsychotics for another 52 weeks yeah and you're looking at the relapse rate and so you just mentioned the the goal I think you mentioned earlier the end goal is to have a relapse rate that's similar at worst compared to the antipsychotics which you just said comparably comparable okay comparably it needs it's not to be similar at the at the percentages it needs to be very comparable and and it needs to stay in in a range which is acceptable yes which which does not tell you that you need to to have antipsychotics to to to keep it low yeah so again if you're below 20 percent uh you you are you are you are you are in a in a in a range which is which is uh very low in terms of relapse rates overall yes i mean when you're considering schizophrenia patients.

OK, very good. And so you've been working on Roloperidone for a while. I would love to just get the latest and greatest on your IP position. Are there ways to extend it further, especially if you get this approved?

So definitely the composition of matter is no longer available. We have seven families of patents, and I think very strong patents here. But there is one patent. You know, when you think about Orange Book, there is one patent which is very important, which is called a formulation patent, which is much more sophisticated than a formulation patent, which is going to 2038 without extension. So this one is already giving you quite a lot of runway to commercialize the drug. But yes, you're right, because now we have a little bit more room to do some additional work on strengthening IP. As we speak, we are working on strengthening IP, and I think we will find elegant ways to extend IP. Because if this drug gets approved, as you know very well, negative symptoms is a transdiagnostic problem, yes not not specific to schizophrenia so so more runway we have more indications we can go after when the drug is approved right okay well i think that's all the time we have but yeah thank you for sharing the sorry i was speaking too much maybe oh no you answered all my questions so um so but thank you remy and i thank you and uh we'll look forward to the data next year thank you so much thank you very good thank you