Great. Thanks very much. It's my pleasure to be introducing Minerva Neurosciences team. Remy, it's always good to see you. Thanks for joining. I appreciate it. Maybe I'll have you give a ten-minute overview of the company, and then we'll do Q&A. How does that sound?
Yeah, this sounds perfect, Paul, and really thank you for having Jeff and myself here just to give you an update about the progress we are making with roluperidone.
Great
which is potentially the first drug to help patients suffering from schizophrenia and having impairing negative symptoms. Yes. Again, thank you for the opportunity. What I'm trying to do here, just before jumping into the data, is just in three slides to give a little bit an update about what is schizophrenia and what these patients are suffering, basically in terms of symptoms. As you can see on this schematic here, yes, which is coming from Professor Correll and Professor Schooler, you see very nicely that when you're looking to the negative symptoms here in brown, you see that the negative symptoms are part of three different types of symptoms constituting, you know, the symptoms of patients with schizophrenia.
Positive symptoms here in green, again negative symptoms here, and cognitive deficit in blue, yeah? What is important to understand is that when you treat patients with antipsychotics, you definitely are improving positive symptoms, but you're definitely not improving negative symptoms. As you can see, negative symptoms are present prodromal before you're making the diagnosis of schizophrenia, before you have the first episode of positive symptoms. The things are building up over time and are present all lifelong for these patients.
As you can see also, what is very important to keep in mind is that, in terms of severity of illness and disability, you see that, what is really impairing these patients and keeping them out of any decent life, are negative symptoms, clearly as demonstrated here.
Now, what are negative symptoms? Because we are all speaking about negative symptoms, and I think it's maybe good to have just few words about what we mean by negative symptoms and what these negative symptoms represent for the patient.
So first of all, it is clear that, negative symptoms, are not treated because there is no approved treatment, in the U.S. for negative symptoms as of today. I think they are also underdiagnosed. When you are...
When you're going, you know, with all the consensus papers or discussing with Correll, it is clear that everybody agrees that negative symptoms is what is impacting most quality of life in these patients. Again, it presents a real challenge for treatment because again, we have no approved treatment as of today. What is also something we have to keep in mind is that antipsychotics treatment might even worsen negative symptoms, so this is what we call secondary negative symptoms. Because of all this, I think, clinical practice and the drug development efforts have obviously tried for negative symptoms. They failed.
I mean, people are mostly focusing on, treating positive symptoms, coming up with a new, mechanism of actions and probably also with drugs with a better side effect profile. This is what we have to achieve here, when we try to improve negative symptoms. Negative symptoms is a construct of different symptoms, as you can see on this slide. You have two different dimensions, the experiential dimension and the expressive dimension. Without going into the details, I think, one type of symptoms, which is extremely important to keep in mind is avolition. Avolition is this lack of motivation, this lack of drive.
Patients are no longer engaged in things they were engaged before they had the disease, and obviously as a consequence, the self-care is also very limited. There are a lot of data, and I, because we have only 10 minutes here, showing that negative avolition is probably driving not only the improvement of negative symptoms overall, but it is also heavily improving functioning and probably cognitive impairment. Yes, because I really think that there is now more and more consensus saying that if you're able to improve negative symptoms/avolition, you might also see an improvement not only in functioning, but also in terms of cognitive aspects.
Now, what is the best way to assess negative symptoms?
There have been a lot of debates, obviously, over the last decades, and I have been involved because I was running a research institute and had quite a lot of companies. I think what is very clear is that, you know, using the classical trial design to assess positive symptoms is probably not the way to go.
Yes. I mean, this is my first bullet point here. Yes. I'm
I will not go to the upper part of the slide, but it is very clear that this is also very clear in the mind of the regulator is that when you bring down someone from an acute episode of positive symptoms, you bring them down to their baseline negative symptoms, to the disease-related negative symptoms, and this is what we are usually calling a pseudo effect. Now, the other way to do this is to put your drug under investigation on top of an antipsychotic, yes, because you know, the belief, which is true, yes, is that when you keep someone treated with an antipsychotic, the risk of relapse and the hospitalization rate is decreased.
What is also something we should take into account when we are thinking about designing trials for negative symptoms or assessing negative symptoms is that when you treat someone with an antipsychotic slash a dopamine-blocking molecule, you induce sedation, you increase avolition, you increase the movement disorders, EPS, akathisia. Definitely this does not help, and this might induce secondary negative symptoms. When you go a little bit more into the how the brain is functioning, and particularly in terms of emotions, in terms of things which are probably involved in negative symptoms, like the limbic structures, going to block dopamine can definitely not help us in order to see an improvement.
Last but not least, because we have primary negative symptoms, disease-related negative symptoms, which are really the impairing part of the negative symptoms, and because you have secondary negative symptoms, if you're doing this on top of antipsychotics, you cannot discriminate between primary and secondary negative symptoms.
Hey, Remy, can I chime in with a quick question here?
Sure.
We can get into the data and make it, like, more interactive too. Like, if you're a clinician and you have someone who has stable positive symptoms but, you know, residual negative symptoms, like, how do you know that it's safe to take them off the antipsychotic? That the probability of a positive symptom relapse is low, right? I mean, this will kind of get into, you know, how your drug could be adopted in the real world for patients with negative symptoms. If they use that as a monotherapy, they need to be comfortable, right, that the positive is not going to sort of spike. Or at least confident that your drug can maintain stable positive. How would a provider feel?
In your question, there are several responses and several elements to take into account. Yes. First of all, I mean, it is very clear that you have different subtypes of patients inside the ecosystem of schizophrenia. Yes. It is very clear now, and a lot of people are working on it and are publishing on it, and you have also long-term data, even going to 20 years. Yes. I mean, follow-up of patients where you clearly show that, yes, maybe acutely, maybe subacutely, you know, over 1 year, antipsychotics are protecting against relapses and hospitalization.
Long term, if you're going beyond one year, if you're going to five years to 10 years, definitely there is a population who does not need antipsychotics because relapse rate is very similar between treated patients and non-treated patients. What is also important to keep in mind is that, when you're looking to functioning, these patients who are not continuously treated basically are functioning better. This notion of intermittent treatment, which is now becoming more and more popular, and bringing back the antipsychotics to why they have been developed. Yes. They have been developed as major tranquilizers. If you have an acute episode of psychosis, there is no discussion. You need definitely an antipsychotic to bring down the patient to his baseline. Let me put it like this.
The second reason why I think the prescriber can be reassured is the pharmacology of the drug. Yes. This drug is not blocking dopamine, roluperidone, and this was intentional because of all the reasons I was just giving in the previous slide. You need nevertheless to have a drug which is not only improving negative symptoms, but which is also keeping patients stable in terms of positive symptoms. Yes. Basically, when you're looking at the pharmacology, this is a drug having a 5-HT2A antagonism, a sigma-2 antagonism, and some alpha-1A antagonism. It's very well described in the literature that 5-HT2A antagonism is definitely helping to keep someone stable in terms of their positive symptoms.
At minimum, in the patient population we have targeted, which is a patient population where we are looking to stability of the positive symptoms over the last 6 months before thinking about monotherapy with roluperidone and obviously they need to be very negative symptoms. Last but not least, we have more than 800 patients who have participated to our trial. I can also mention Lundbeck, who did a trial with the same patient profile. What happens is very clear is that the relapse rate is extremely low, yes, in this patient population.
Last but not least, I will come to this when I speak about the trial design we are currently working on, and we agreed with the FDA, definitely we will get some additional data to reassure the prescriber about monotherapy. Just coming to the fact that if you want to demonstrate an effect on primary negative symptoms, you probably have only one way, which is monotherapy versus placebo, because there is no approved treatment currently for negative symptoms. Because we have not a lot of time, I allowed myself to put here the integrated analysis of the two studies we have carried out, registrational studies. Yes. This is the primary endpoint, which is coming out from the PANSS scale, which is called the Marder negative score.
The data I'm showing you here, Excuse me, are the ITT population, and the way this is analyzed is the MMRM approach. These are changes from baseline. Primary endpoint of efficacy was week 12, and what you can see very nicely here is that, already after week 2, you could start to see an improvement compared to placebo for the 2 doses, blue being placebo, green being 32 milligram, and red 64 milligram, things are building up over time, and at the end of the 12 weeks, you have p values for the 2 doses.
You can also see that 64 milligram is definitely having a stronger effect than 32 milligram, and this is also seen in the effect size here as you have with the two doses. Just one word about the two studies. The first study was extremely positive with P values of 0.0001 with effect sizes of 0.6, 0.7. What happened in the second study is that and I will come to this in a second, placebo picked up a little bit more. I'm not, how to say, blaming placebo here. I'm just saying that placebo picked up a little bit more.
The effect size or the improvement with treatment was the same, but what happened is that only the highest dose was hitting a p-value in the second study. Because we used, we were optimistic, maybe too optimistic, we used to control for type one error, we used the Hochberg correction. We did not do a top-down analysis like you might do in phase 3. We wanted to have the two doses hitting. Unfortunately, only the highest dose was hitting. We can only speak about a nominal improvement of 64 milligram in the second study. Again, maybe a statistical artifact, clear that we have a real treatment effect for 64 milligram in the two studies. This is what we have as of today.
Now, what is important and what I was telling you in the last previous slides, functional improvement is important. Here I'm just showing you the. I could show you exactly the same for the first study, yes. This is the PSP result, which by the way was the sole key secondary endpoint in the second study. You see that when you're looking to PSP total score, you see that these patients are really improving in terms of functioning. Best of my knowledge, I think roluperidone is the only drug in the space improving patients' functioning. If we would have time, I could go into examples of patients who have dramatically improved and were able to go back even to have a small job. Yes.
I mean, which were observations which came out from the studies. Just to mention that definitely the 64 milligram is improving functioning as well. This is a quite quick summary of all what we have seen with the two studies. Improvement of primary negative symptoms, improvement of functioning, improvement of avolition, which has been demonstrated in several ways, in several approaches, several analyses. No increase in terms of positive symptoms. Monotherapy was associated with a very low relapse rate. What is also important, I think, to keep in mind is that we did not see all what you see with antipsychotics, particularly dopamine-blocking molecules, in terms of side effect profile. Yes. The drug is extremely well-tolerated. No sedation, no EPS, no prolactin increase, no nausea.
I think it ticks a box about the pharmacology we are using here and the patient population, you know, who has participated to these trials. Based on this, I mean, after you know several back and forth with the FDA, and I wanted to thank once more the FDA about all these very helpful discussions we had. Remember, they even organized a public meeting in August 2024 to speak about how to design a negative symptom study, how what would be the endpoints. I mean, we came to this overall agreement. Yes. As you know, there is always a discussion going on with the FDA, but clearly we came to this conclusion.
The first thing is that we agreed that the primary endpoint should be at week 12, like in the 2 previous studies, with the same primary endpoint, which is the Marder negative score coming out from the PANSS scale. Like in the phase 3 study, the previous phase 3 study, the sole key secondary endpoint is PSP looking to functioning. This is a double-blind, placebo-controlled in monotherapy. Before including the patients, same elements as in the 2 previous studies. Patients need to be stable in terms of positive symptoms. They need to be stable in terms of negative symptoms over the last 6 months, and they need to have a minimum score of negative symptoms in order to really quantify impairment.
This is really, let me put this in brackets, but this is quasi a copy-paste of the two previous studies. The big difference here is that we are only testing 64 milligram versus placebo, yes, and not the two doses versus placebo. Afterwards, this is I think the one part also of the answer to your question, you know, we have already a lot of data analyzing existing databases like the CATIE database, showing that the patient population we are targeting is definitely not keen to relapse a lot. It confirms what we have seen also with our data.
What I think the agency would like to see, in addition, are data in this study where we are comparing really the relapse rate of roluperidone over a longer period of time compared to Antipsychotics. We came up, I think, with a very innovative study design here. Yes, you can see that patients are re-randomized to either roluperidone or antipsychotic. Basically we are using three antipsychotics. The three most prescribed antipsychotics are in this arm and because even the agency recognized that, I mean, the blinding is a little bit complicated due to the side effects of antipsychotics. We came up with a double dummy study design, so really to try to blind the study as much as possible.
This part is purely here on a descriptive level to check relapses in these patients over a longer term period of time. Primary endpoint of efficacy, week 12, 64 milligram versus placebo, and afterwards, you know, a comparison on a descriptive level between relapses of antipsychotics versus 64 milligram-
Hey, Remy, can I chime in and ask a couple quick questions?
Sure.
Sorry. I didn't mean to interrupt. Just a couple quick things. One, do we have a good understanding of what is a clinically meaningful effect size for the FDA? Like, do you just need to hit a P value in the acute portion and, or do you need to show something above a certain threshold? Second, maybe just talk a little bit more about the relapse assessment portion. Like, what constitutes qualitative non-inferiority in this con-
Yeah. Clinically meaningfulness. This was obviously a debate since we received the CRL. Because it was written in the CRL and it is true that the FDA was writing this in the CRL. This was one of the discussions that we had over the last quasi two years with the FDA. The first thing you have to keep in, and what I'm saying here is something we have discussed, and we have agreed. Just to be clear. The first thing is that because there is no approved treatment, it is extremely difficult to have any benchmark. I allow me to take always the same example.
I was involved in trying to demonstrate clinically meaningfulness of SSRIs and SNRIs, and guess what? It took us 10 years to come up with some consensus after the molecules were on the market. So it is extremely complicated to come up with clinically meaningfulness, you know, when you have the first drug improving the symptoms you want to treat. This is one aspect. The second aspect is that having all these interactions with the FDA, we were able to share with the FDA a lot of data. First set of data was the responder analysis. You know, this is what is usually done when a drug gets approved. When you're looking to our responder analysis, it is largely in favor of the drug. Yes.
Much more responders with roluperidone in the two studies, by the way. Yes. It's not specific to the first study. The two studies. We did also an anchor analysis, anchoring the primary endpoint to CGI-S, which is well-recognized. 1-point improvement of CGI-S is recognized to be clinically meaningful. When you're doing this analysis, again, highly significant in favor of the drug. Yes. We have shared all this with the FDA and they really agreed that we have already a lot of data here and what we need basically is a p-value and, you know, we will do this responder analysis, a classical thing you're doing. Last but not least, I think there is an external source which also was.
is very helpful to keep in mind that you know we have published I think now 12 papers on the data. We are very open. We have shared the data with Correll, and Professor Gregory Strauss has published on this. Yes. What he did, he did an analysis between the BNSS scale, which was a scale we used as a secondary endpoint in the first study, and was asking the patients and the caregiver what for them on the scale is clinically meaningful. Yes. When he is comparing this outcome of this study data to our data, roluperidone is as good or even better than what the patients and the caregiver are expecting. This is an important piece of information.
Long story short, the answer is a P value is enough, yes, in this case. No counter con-
Yeah. No, that's great context.
Concerning the relapse.
Yeah.
So, uh-
Please.
I guess a question you have, first of all, non-inferiority, a formal statistical analysis is impossible. Why? Because the number of relapses we had was a handful of relapses, and if you're doing a calculation of number of patients you would need, it is between 2,000 and 3,000 patients. Yes. We agree that, I mean, this makes no sense. What is interesting here is what we agreed with the FDA is that we will use some psychometric endpoints coming out from the PANSS scale.
Like, you know, this consensus, the few papers which have been published recently, Siafis, for example, where they explain that, an increase of 12 points and more on the total PANSS score is probably signaling, a relapse and, is also, related or correlated, I would say, with a number of hospitalizations. We will have this psychometric endpoint and because we're measuring PANSS, we're measuring CGI, we are measuring, you know, PSP, you can analyze this psychometric part as much as you want in different ways. But for the moment, what we are going after is consensus currently, in the scientific community and the medical community.
What the FDA also wanted, and I think it's completely fair. They want also what they call hard clinical endpoints. That's in addition to psychometric endpoints. Basically looking to hospitalization, looking to you know attempt of suicide, aggressivity towards a caregiver. All this daily you know things which might happen when a patient starts to get worse in terms of
Right
the positive symptoms. This is what we'll be, you know, monitored in this Part B of the study.
Okay, great. We are out of time, but maybe just very quickly, the timing of this readout, Remy?
Yeah. I'm switching. This was a slide which I think was an important slide why I think it's not how to say testing that we have a drug. It's to execute in order to demonstrate the gains that we have a drug here, yes. The timing here. As we speak, we have you know obviously selected a CRO to help us. Multiple sites activated. First patient will be in the study very soon in Q2 of this year. Top line results of week 12 will be second half of next year. Relapse assessment will be obviously a little bit afterwards. We will keep the FDA you know updated as much as possible.
When we have the top line results of week 12, we will definitely re-engage about, you know, NDA submission. Yeah. This is, what are the timelines, Paul.
Okay, great. Awesome overview. Thank you guys very much for joining. Appreciate it.
Thank you, Jacob.
We'll have to end now, but look forward to catching up soon. Thanks everyone for listening.
Thank you.