Your Relight trial that's coming next year. Remind the audience how these trials are designed in adjunctive MDD. When can we expect to see some data here and help frame expectations?
OK. Well, let's start from how the trial is designed, right? So we come from some experience in running phase III clinical trials. So we did two in the past. The design is somewhat similar. Now, we made a lot of amendments and changes to improve, right, especially the placebo control. But in terms of design, it's really simple. So it's two doses, placebo and 25 mg of REL-1017. And the trial lasts four weeks. And the primary endpoint is the delta between placebo and the drug at day 28. Everything else has been already done. So there is no more need for the long-term safety. The 12 months has been completed. The phase I special population, renal, hepatic, is all done. So that's made the design very, very simple. So primary endpoint, 28 days, difference between placebo and the drug.
Maybe you'll remind us or maybe help provide some context. You've referenced your prior phase III trials here. I think in both of those, you did see an elevated placebo response. What have you implemented in these trials now to try to control for that?
Well, thanks for the question because, Andrea, this has been the real focus of the program. As everybody knows, the results of the other two trials have not been good enough to file. The [Inaudible] , the second one, is a per protocol at a p-value of 0.051. It was really borderline and missed by a hair, but in any case, not good enough for a filing. The focus of the two phase III clinical trials they're running now was to be able to control the placebo effect. We made changes across the board on many aspects. We do believe we kind of figured out what did not go right in the other trial. Something was more controllable. Something else was really COVID-related.
So, to have oversight on a large number of sites, with 100 sites in two different clinical trials, without the ability to go into the sites. Most of the interviews for the rating of the patient were done via Zoom. It was a kind of like a difficult environment where we didn't know at that time, but it really impacted the data of the studies. We made changes in three different areas. With the old goal, right, to enroll the right patients and to control the placebo effect. The protocol, so the three changes are protocol, site selection, and patients. Let's start with the protocol. The protocol was the first phase III trials program that we did.
It was, I don't want to call it complex, but it was like it was aimed to detect a lot of information, especially about safety and cognition and other aspects. Now, all that is done. The protocol has been simplified quite a bit in terms of like there is no more requirement of all the safety like ECG and because we have the data. We don't need to do all that. That translates in the patients having a lot less contact with the investigator and spending a lot less time in the site. And that by itself, it should help to control the placebo effect. So simplified protocol. At the site selection, there's no magic because you only see if a site is a good site or not in terms of controlling placebo effect after you see the data.
But we have a luxury that we have data, proprietary data on over 100 sites that participate in the previous clinical trials. And so this data, we use this data, and we can clearly see the sites that have been able to control placebo effect and the sites that were not able to control placebo effect. In particular, as you remember, there were two sites that had particularly like not very good data. And so clearly, these sites are not part of the site selection anymore. So site selection has been somewhat optimized. We do have a certain number of sites that are new. And we do as much as possible to keep like an oversight on these sites. So looking at the blinded data that clearly don't tell you anything about efficacy. But you can see there is some anomaly, right?
If you see over the four weeks where the MADRS scale is detected or the patient is interviewed, if you see like zigzag or like data that are outside of the normal behavior of placebo and a drug, you can call the sites and verify it was the wrong patient. So something happened related with having data that are not like in line with what you expect for placebo or for the drug. And so we try to keep an eye as much as possible on the site's behavior. We are limiting the number of patients. There was an issue with the previous clinical trials, right? We had a few sites that had an oversized number of patients recruited, like over 20 per site. And clearly, that can have an oversized impact on the overall trial. So we are trying to avoid that.
So we try to limit without giving a specific number of patients maximum for each site. But between the oversight and limiting like the impact of every single site on the trial should definitely help to avoid what affected us in the other trials. The last that is the most important is the patient selection. For a variety of reasons, we do believe that at least a sizable part of the patients enrolled in the previous trial were not really affected by biological depression. It was mostly a consequence of the COVID situation where there was a lot of reactive depression that clearly goes away with placebo and maybe sometimes just with the contact with other people and the reassurance that you're not sick. And so to avoid that, we put a certain parameter. And we do have an adjudication process that is rather severe.
So we had four different steps where the patient is evaluated and is considered for enrollment. So the sites, we have the CTNI SAFER, that is a second site that has contact with the patient and rediagnoses the patient using a different questionnaire called SAFER. Then we have an internal team that reviews all the data without contact with the patients. And the last, we do have an external independent adjudicator that signs off on the patient. So it's a four-step process. And there is a lot more oversight on the patient selection, which is now possible because people can actually travel. And I do believe that the most important amendment that would be expected to make a big difference in the data is the requirement for medical and pharmacy records. So this is an adjunctive treatment.
The patient has to be on treatment with an antidepressant and not to respond to the current treatment. Then he can be enrolled in the study. In the previous trials, there was industry norm. There was no requirement to provide a proof that you actually have been taking antidepressants and that you have been diagnosed with depression. And we are now requiring that. Every patient to be enrolled has to provide his family doctor or whoever diagnosed him or her with depression and provide a pharmacy record that proves that he has been at least purchasing antidepressants from a pharmacy for a certain amount of time. That should support, right, that the patients that are enrolled are really affected by depression.
Got it. As you think about these patients, just remind us what the inclusion exclusion criteria are for the trial?
Yeah. Well, the protocol inclusion exclusion criteria, but there are a lot of them. From the most detail. But in general, the more stringent or the inclusion exclusion that the patient has to be affected by depression, it has to be treated and has to be taking an antidepressant for at least six weeks. So that's the key criteria. Should not have an improvement since he started the treatment of, there has to be less than 50%. Technically, he does not respond well to the current treatment. That's the basic inclusion exclusion. Then, of course, the other clinical trials, like they are not compatible. Diabetes and concomitant have an exclusion criteria that is taking any kind of drugs like amphetamine, methamphetamine, opioids, but also THC, marijuana.
I mention it because, not surprisingly, but there is a large percentage of patients. I believe it's close to 10%. They present and they are taking marijuana. So that's unfortunately easy to exclude. So it would be risky to include the patients. So concomitant medication is one of the also the inclusion exclusion criteria more than inclusion.
What is the baseline cutoff for your HAM-D score for these patients?
Well, we use the HAM-D, yes, is 19.
How does 19, I guess maybe if you could provide some context relative to other clinical trials, because I think that's been a focus that if you have maybe less severely depressed patients, it may be more difficult to maybe tease out that signal. Help us understand how you came to 19 as the cutoff.
Right. Well, it's a good question. 19 is kind of the industry standard. It doesn't mean that you enroll a lot of patients with 19 HAM-D. So we use HAM-D for diagnosis and then MADRS for baseline and during the trial. So 19 HAM-D corresponds very roughly about 24, 25 MADRS scale. It's kind of industry standard for depression clinical trial. I do, as you said, and I do fully agree because we have seen the data. The more severe is the patient and the more the effect of the drug shows up. And placebo, if you're really severely depressed, only does that much. But the 19 is the cutoff, the minimum. If you look at the average blinded mixed data between placebo and drug, I do believe is around 33. That is somewhat in line with the previous trial.
The severity of the patient enrolled is much higher than the minimum required to get in.
Why not have a higher cutoff maybe to help maybe ensure that you are getting a more severe patient population?
Well, t he sites, they want to have more room in the patient. We don't know because we haven't really. They may want to have like a larger range of patient population that reflect more the real patient population. So there was no specific reason, but it's more to have a widened pool of patients where to pick the patients for recruitment. So no specific reason. It's just you need to have a cutoff. And also the concern, if you have put a cutoff very high, the sites say, OK, you know it's too restrictive. And there is competition too on running clinical trials. So you have to keep some included in the criteria. They are not put at a disadvantage to competition.
Maybe when you've spoken in the past and you've disclosed 80%, and that seems to be well above at least your prior trials, how does that compare to the extent that you're aware of other clinical trials in depression here? And what would you maybe pinpoint as the nature of those failures?
Yeah. So I can give you a lot of details on this because, as you may imagine, right, the attention and the focus is on enrolling patients that are suitable to be included in the trial. And , how can I say, we don't really know all the, these are proprietary data. We disclose it because we tend to be very shareholder- friendly and investor- friendly. So we tell as much as we can. And so, but we do believe that 75%, 80% is on the higher side of the screening failure. We are trying to work on that. But in the previous trial, what I can tell you, it was around 50%. And it was lower. We have been told by the expert that it's more in the range of the industry.
We are trying to see if there is a way without increasing the risk to improve this screening failure. We look at the reason for screening failure. There are some more legitimate, like THC is a big one. Previous clinical trial enrollment is another sizable one. I would say half of the patients, they fail for blood work, right? Thyroid hormone, uncontrolled diabetes. There is not much you can do about that, right? Unfortunately, there are patients that in the real world, you can treat. In a clinical trial, you want to be more careful about, right? Especially like the thyroid because thyroid, one effect is depression. But it doesn't respond to drugs because it's related with the thyroid hormone. So we are trying to improve it. More recently, I believe we touched on that in one of our past conversations.
We expanded a little bit the range of allowed medication. But as a reaction to the previous trials, we put parameters that were very restrictive. So then with the feedback with the sites and we only listen to what the site tells us, there was certain restriction. There was a lifetime ban for ketamine. And the sites said, well, if somebody has been treated with ketamine like three years ago, it's not going to affect your trial. So now that restriction has been lifted. And there was like a ban in the use of benzodiazepine because they are depressed, right? And we changed that angle. And now benzodiazepines are allowed as long as they are taken to sleep and they are not taken all day and to stay in the bed all day. And so there is some small modification that should help over time.
But look, the goal is really to have a positive study. To have a positive study, we know we have to enroll patients that are affected by depression.
Maybe on that point as a follow-up, if you are enrolling patients who are taking, say, benzodiazepines, maybe even at a low level, how are you confident that that won't skew your data or maybe skew the response since benzos are used as antidepressants?
Yeah. So you look at the dose, right? If benzodiazepines, they don't last a long time. So if you take it in the evening before to sleep just to induce sleep, then it's fine. One thing that you don't want to do during the four weeks is to change the habit of the patients, right? It would be easy to say, OK, you know I enroll you as the patient if you stop taking benzodiazepines for four weeks. But that would be a change in the habit. It probably would have a much bigger impact than just to leave the status quo. And if you're taking the same dose of benzodiazepine on a daily basis just to sleep, right, the drug is not really having an effect. So maintaining the status quo is very important.
As you think about these protocol amendments that you've put into place now to control the placebo response, I guess how much risk is there that 30% of your patients in Reliance II were actually enrolled before any of these were implemented? As you think about the data that's coming later this year.
Right. Yeah. As you may imagine, we get that question quite a bit. And so we look into these patients. Of course, they're blinded. So there is no way to know if they are good patients, not good patients. But the level of confidence is that in two different parts. One, half of these patients were enrolled after the COVID restrictions were lifted, where the data that we have in the other studies that were very good. And so of the 80-90 patients enrolled pre-amendment, half, only half have been enrolled during the COVID restriction. So the number gets down to like 30-40 patients that really could have been affected by the COVID restriction. And all in all, I mean, we cut this data in a million different ways.
But in reality, if you get into the bottom line, we were two sites that had bad data and affected the whole trial, right? Excluding these two sites, the data would have been nice. The study would have been statistically significant. These two sites, they've never been in study 3 or 2. You put all together. That's one aspect. The second aspect is that we look at the data post-COVID lift of the restriction, how the data looked like. There was a 6-point delta between the placebo and the drug. Unfortunately, it was only 60 patients because we stopped it early. But the data were looking very good with a non-amended protocol. Now with the amended protocol, we are enrolling 200 patients.
And if you add the 30, 40 enrolled after the COVID restriction from the pre-amendment, so you get to like 200+ patients that they should have, expect to have good results. And the delta from placebo, assuming no delta, so zero effect of the drug in the previous 70, 80 patients, the delta in the following 200, 220 has to be like 2.5 points or 3 points. If it's flat, it's 3 points. So it is feasible. And the bar is not very high. So with all we have done, all we are doing, everything we are doing, it should be able to overcome even if these data are not that great. If we assume that the first 70, 80, 90 patients pre-amendment, they have a delta of 2.3 points. That was the overall delta.
Then the following 200 patients, they should have a delta of 2.5 points. We are fine.
What are you powering your study to detect?
It's powered to detect maybe 2 points, 2.5 points delta depending on the standard deviation. That would be the minimum, right? We hope we can do better than that. But that would be the minimum to detect the 0.05 p-value.
Maybe on that point, because you have had some, I think, very robust data in the phase II trial, maybe remind us the profile that you've seen there. How important is the rapid onset of action? You're seeing that be touted as a selling point for some other drugs that are on the market right now. Where does Relmada Therapeutics 1017 eventually fit into the treatment paradigm?
Yeah. So the rapid acting is a very good feature to have and differentiate from most of the other drugs. And especially from the antipsychotic, they are indicated for adjunctive treatment. Remember that there is no antidepressant approved as adjunctive treatment of depression oral. And so we discussed that with the FDA. We actually wanted to leverage as much as possible the rapid acting. But I mean, the FDA seems a lot more oriented. They do believe rapid acting, if it's on the label, it's a good feature to have. But they are very much oriented in having duration of effect, right? And so if we would have a primary endpoint of 7 days was not acceptable by the FDA. They're nice to have, but we want to see if the drug works for at least 3-4 weeks when the patient is on drug.
So they wanted the primary endpoint to be day 28. So nice feature to have. It is not critical on the, at least for the FDA. Commercially, yes, it is. You want to have a rapid acting as much as possible. But it's very important the drug lasts over time. For the European, it's even more important. European duration of effect and the impact on the economics, if you can go to work earlier, it's even more important than in the U.S. So duration is important.
As the profile of REL-1017 has emerged, whether it's the phase II study, your open label study, what has been the receptivity from the medical community as to the profile of this drug?
Yeah. Well, the feedback has been, if you can do something that doesn't need to be the same, but even close to what the phase II data look like, it would be very well accepted. W e are not emphasizing it because efficacy is critical to get approval. But the tolerability profile of REL-1017 is something that on the marketplace, according to the KOLs and doctor, it's very, very important. I mean, the antipsychotic, they're used widely as adjunctive treatment. And the efficacy, there is some efficacy, but it takes a while before you can see any efficacy. And the tolerability profile is not as ideal because diabetes, metabolic changes over time, these are chronic drugs. So to have a clean safety and tolerability profile like REL-1017 on the marketplace will be very important, assuming that you show efficacy at that. That's why we are focusing on efficacy first.
Maybe remind us the work you've done with the human abuse potential studies?
Right. Yeah. I do believe that Relmada Therapeutics' REL-1017, we have done the most extensive abuse potential program ever done for a drug. We have published the data, right? There is a placebo-like profile. So we feel very comfortable. We have a very, very good team that specifically has done follow-up for the whole program. One of the members is the former head of the DEA. So they are very well. We have two former FDA members, retired, that help us out. And they all said with this profile, initially, it should be Schedule IV/Schedule V. And eventually, after like one or two years of experience, there is no sign of abuse. It could even be a non-scheduled drug.
What I can tell you is that in all the data that we have seen in the abuse studies that were specifically designed for that, but also in the whole program, phase II and especially the phase III long-term, we haven't seen one single case of attempt of abuse of the drug. Consider that we did propose to the FDA two weeks in the long-term safety. They only come once a month for the visit. We're coming once a month to the clinic for the visit. We thought about giving the patient to go home two weeks of therapy. So it would be like 14 tablets and/or 15 tablets. And the FDA said, "No, you don't want the patient to come in every two weeks. Give it 30 tablets". So that's a sign. That was before data.
That's a sign that the FDA, of course, taking everything very seriously, but they don't seem very preoccupied about safety or about the abuse potential of the drug. 30 tablets is a lot.
Maybe given the clean signal that you've seen there and the lack of an abuse potential, remind us why the assumption is that you would start with a Schedule IV or Schedule V.
Well, it is mostly for the parent drug, right? It's the isomer of racemic methadone. So I mean, some cautiousness would be required. That's according to the team. But again, there is no sign of any potential for abuse. And so dextromethorphan has a long history. So it is an over-the-counter drug. And it's not that different from the mechanism of action. So there is to be cautious in Schedule IV or Schedule V.
What are you hearing from the medical community on that front? A Schedule IV or Schedule V versus no schedule at all, not a controlled substance, how would that impact how they think about using this drug?
Yeah. Well, I'll tell you, we spoke with a lot of doctors and KOLs and not only KOLs, also prescribing doctors and in the sites. The common answer is that many of the doctors don't even know if it's Schedule IV or Schedule V. And the issue would be Schedule II would be restrictive. Schedule III is already well more accepted. Schedule IV or Schedule V is not. It's not the focus. You can repeat the prescription. So that doesn't carry any burden for the patients or the doctor. So IV or V is very, very benign. And Schedule II would be the one that you definitely don't want. Although all stimulants, and they are used for kids, they are Schedule II. And they sell a lot. So even Schedule II, if you have a good drug, would not be like a deterrent.
Schedule IV or Schedule V is definitely not, would not be an issue.
I mean, maybe a follow-up there. If you were Schedule IV or Schedule V, which, recognizing it is a much lower bar than a Schedule II for these physicians, but how would that impact it? I guess maybe as you think about other antidepressants on the market that really, minus Spravato, don't have a schedule there, what particular patient will be prescribed REL-1017?
That's a good question. Well, REL-1017, if approved, would be indicated for adjunctive treatment. So it is a second line no matter what. So it would not compete directly with the current antidepressant. Real competition would be the antipsychotic. And they are not scheduled, but clearly they have their own baggage.
Maybe in the last couple of minutes, we can jump over to your REL-P11 program.
Right, right.
The psilocybin. Remind us the genesis of this. Why there's this interest to explore this agent in obesity? Why does that make sense right now?
Yeah. And I'm glad you asked the question because sometimes serendipity also helps a lot. But you have to be able to recognize when you have data that could potentially have a big reward. And P11 was originally studied as a neurogenic, as a neuroplastogen, and similar to an NMDA. So it helps to grow nerves. So the indication could have been, and could be, very various. Anything related with neuroprotection, like one thing that we have been looking, we are still looking, it's stroke, occipital stroke, where the recovery after stroke. But during these studies, neurogenesis, it is much slower, or it is slower in obese people, but in obese animals as well. So the team, the Chief Scientific Officer, to better detect the neurogenic effect of P11, of psilocybin, he thought about using rodent, the artificially made fat, made obese, with the high fructose, high fat diet.
But the goal was to show the effect of the drug as a neuroplastic agent. That has been demonstrated before, but we wanted to have our own data. And the neuroplastic effect was seen. But what has also been seen is that in the rodents who were taking placebo and the high-fat atherogenic diet, nothing changed. The rodent stayed the same weight. And in the rodents that were administered very low dose, we are talking about 0.5-1 mg versus 20-25 mg of the psychedelic used for depression and for a psychiatric indication. We used 1/20th or lower dose, so where there is no seen psychedelic effect. And the rodent treated with P11, they lost weight about 10%. Glucose went down of an amount similar, if not more than metformin. And the liver that was clearly white and full of fat, it got clear.
They became pretty much normal after a few weeks. Based on these data, we repeated the study because you don't want to have by chance same results. We look at the combination with GLP-1s as well because they are very popular. And they work. And they have their own issues for loss of muscle. So there is something more that can be done in that space. But we spoke with quite a bit of opinion leaders in metabolic, which we were not really experts. And they all said, look, this one, there is a huge need for a metabolic drug that can help or that can be combined with GLP-1 or can be by themselves. And the indication, we haven't chosen the specific indication now because it works on glucose. It works on weight. And it works on fatty liver.
We are planning to start the phase II proof- of- concept. Hopefully, everything goes well. We start this year. Data sometime next year, probably first half. Based on the data in human, it would be obese patients will decide which is the easiest indication to pursue to get approved.
Perfect. Well, we are just up on time. But Sergio, any last remarks as we await Reliance II Relight data?
Well, we really hope this time we'll get it right. A patient deserves to have a drug like 1017 approved and make it available. So we really hope that this time we are doing everything we can and is possible to do to get positive data. We believe we will.
Great. Well, with that, thank you so much, Sergio.
Thank you. As always, it's a pleasure. Thank you, everyone.