Good morning, everybody, and thank you once again for joining us for our first annual I&I Summit. I'm Jeroen Warburg from the Biotech Team, and it's a great pleasure to moderate the next Fireside Chat with Dianthus. With us today is Marino Garcia, President and CEO. Marino, good to see you as always. Thanks for joining us.
Good to see you. Thank you for the opportunity to talk with you today.
Absolutely. The first thing is I have to congratulate you on the recent El Clásico win over Barcelona. We're talking about the El Clásico. Real Madrid is sitting on top of La Liga. Marino is a huge Real Madrid fan. I'm a Barcelona fan. We're trying to get the three points back.
Yeah, good luck to you. My God, 600 million people watched that match. That was insane.
Crazy. Barça is going to move to their new stadium very soon. Hopefully that brings luck. All right, let's talk about claseprubart. Specifically, I want to start by maybe focusing on gMG. For the audience, if you have questions, feel free to email us or just go ahead and put that into the chat right there in the Wall Street webcasting, and we can read it for you. Congrats on the gMG phase 2. The MAGIC study looked good, and you're now beginning to prep, obviously, for the phase 3 trial design. Have you had a chance yet to talk to FDA or really kind of fine-tune your thinking about the Q4 week dosing and what that might look like?
We haven't come to we haven't discussed it with the FDA yet. We're planning to meet with them to talk about the changes we want to make from phase 2 to phase 3 based on some learnings and the fact that we're going to add a Q4 week arm on top of the Q2 week arm for 300 mg, 2 mm. I really don't see the FDA having an issue with that. I mean, I think the data is very clear. And we have everything we need to support why we think Q4 week would be an equally effective dose.
At that point, you're thinking about the same, right? 300 mg Q week. It's going to be tested two weeks before.
Right. The idea here is to push on our differentiation even further and our advantage versus the competitive set even further. It would be testing placebo, 300 mg, 2 mm, one autoinjector, for example, every two weeks, and then one every four weeks. We always suspected, Jeroen, that every four weeks is on the table. We have a 60-day half-life. It is a very potent antibody. This targeting IC90 and MG was always something that was theoretical, but there was never any evidence that you needed to be above IC90 or to have such high levels of inhibition and to accumulate so much drug to be effective in conditions like myasthenia gravis. Between the some disround data from Regeneron plus what we saw in our open label data and the fact that we did not see a dose response between 300 and 600, we always said if these different bits of information come together and are consistent, we will be looking at once-a-month dosing. That is exactly what we are doing now.
Okay. In the phase 3, I think you're considering a screening criteria with a QMG score above 10. That was not done right in the MAGIC phase 2. What's the thinking there?
When you look at the C5s that are approved, Soliris, Ultomiris, they didn't have a QMG screening criteria. When you look at zilucoplan or Zilbrysq, RA Pharma originally and then was acquired by UCB Pharma, they had a QMG screening criteria of 12. We're debating, like, do we put one in or not in our phase 2? You have to remember, we're a young company. We wanted to execute quickly. The QMG, unlike the MG-ADL, which is eight questions, patients self-administer, it's very fast. QMG is something the physicians have to administer, and it could take up to two hours. We said, if we put this in as a screening criteria, it's going to be just another hurdle in recruiting patients into our trial. We're a young company. This is our first trial in this space. Let's not make it difficult.
We said, let's not have a QMG screening criteria. What we saw, if you look at the baseline characteristics, the QMG baseline numbers look a little low compared to historical trials. This amazing efficacy we had was probably with milder patients than maybe what C5s in the past have recruited into their trials. That makes the data even more impressive, right? I mean, the more severe, the more you expect to see a robust response versus placebo. That made us feel really good. At the same time, it made us think, like, what would have happened had we put in, let's say, a QMG of 10 in our screening criteria? What we saw is we don't lose many patients in each of our arms in our post-hoc analysis. Our MG-ADL differentiation or our MG-ADL score versus placebo becomes a 3.0.
That becomes a best-in-disease result, certainly a best-in-class in complement. It is a very small number that we saw that had a very outsized response on placebo. We feel like that will be a way for us to screen out patients that might have an over, too much of a response on placebo and really be able to shine a light properly on what claseprubart can do in MG.
Your MG-ADL's baseline were, again, about eight and a half or so, eight, two to eight and a half. Zilucoplan was anywhere in the phase 3, they were around 10-11. Champion was about 9. A lot of them are sitting at 9 and 10. UCB's were actually a little bit lower than the other one, sort of 8.1-8.4. Vyvgart was 8.6-9. In the ADAPT study, the ADAPT sub-Q was about 8.5-8.8. I think the screening criteria was MG-ADL above 6. I imagine that's going to stay in the phase 3?
Yes, that'll be consistent. Our MG-ADL baseline characteristics are not out of, they're not out of market or they're not out of those numbers. They're not significantly different. The QMG was a little low. That's the one that caught our eye. I think the combination of having a screening criteria of MG-ADL 6 plus or above and QMG 10 or above, I think will make sure that we have patients that are comparable to what C5s have recruited into their trials and the FCRNs. I think it'll bring the numbers more in line on both those measures, both the very subjective self-administered scale that is the MG-ADL as well as the physician-administered, let's say, more objective, although they're both subjective, but more objective scale of the QMG.
Okay, let's move to the ANA data that we saw in the MAGIC study. From the FDA review document, they don't believe there's a clear trend between autoantibody positivity and lupus symptoms. I know that's something investors were asking a lot about. Any thoughts about what that ANA signal, the small one, means? Is there anything you can do in the phase 3 to mitigate that at all?
What does the ANA signal mean? That's the issue. It doesn't seem to mean anything. We were not seeing any sort of clinical symptoms that correlate in any way with ANAs. The FDA asks classical pathway inhibitors or the C1s inhibitors, when they do their first trials in patients, they ask, what are you going to do to look for potential any autoimmune activation, any DIL? For example, drug-induced lupus, something that the anti-TNFs have, the blood pressure meds, the ARBs have in their label. There's really not much you can do except, say, we're going to look for the patients who display any symptoms. Except, okay, we'll test ANAs. That's what sutimlimab did, Enjaymo, which is approved on the market today. You can read the regulatory documents there and what the FDA thinks about that. That's what claseprubart did with their phase 2.
We did the same. Again, just like they found, it has no real meaning. There is nothing that that information tells us. You have to remember, ANAs are not used as screening criteria in the real world, right? ANAs are used to confirm if you have a patient who presents symptoms that look like lupus, you run an ANA lab test in order to help you get more confident that that diagnosis is correct. You cannot use it as a screening criteria. They have a very high false positive, for example, 25% of healthy people walking around the United States have elevated ANAs at any point, and there is no clear reason for it.
Interestingly, one thing we found in our trial that nobody knew, I think, is that MG patients walk around with elevated ANAs all the time, and they have no family history of lupus, no personal history of lupus, no symptoms of lupus. It was actually the number one reason for screening failures in our MG trial. We were really amazed at how many of MG patients tested very high for ANAs with no real reason for it. Just like rilzabrutinib did, we looked at it. We do not see anything that we can, there is no clear so what. We are going to talk to the FDA, and we are going to propose doing exactly what rilzabrutinib did for their phase 3 CIDP program, which is we are going to stop looking at it.
We're going to take some ANAs at the very beginning, and then we're only ever going to test a patient if they display clear symptoms that look like could be drug-induced lupus. Then we'll test them again. Otherwise, testing these patients every time they come in, I mean, the investigators themselves, they saw the data and they go, please, please don't make us do this anymore. It's useless. It's just adding burden to the trial and to the patients. We're not concerned about it. It's really irrelevant. We're going to tell the FDA we want to do the same thing as claseprubart's doing in phase 3, which is stop looking at it or testing for it and just move forward.
Right. So just the baseline and at the final endpoint, or just the baseline and that's it?
Baseline. If anybody ever presented with a rash or something that a physician goes, oh, I feel like this is drug-induced lupus, let's test for ANAs and see if that's what it's looking like. If it is, all you do is you stop the drug and the symptoms go away. That's what Anexon did with their C1q inhibitor. There are very clear reasons for why a C1q inhibitor would have drug-induced lupus symptoms. That's what they did. They just stopped the drug and the symptoms went away. You do not really, you do not see that with the anti-TNFs and other drugs that have this in their label. It is not that you are triggering lupus itself and therefore permanently giving people lupus. It is that you get symptoms that look like it. You just stop the drug, the symptoms go away. That's a theoretical risk with C1s inhibitors because it hasn't happened and it's never been reported with any of the three programs.
Yeah. Okay, so let's move to Captivate. Encouragingly, you said it's enrolling faster than expected. Now we're going to be having interim data in Q2 and not second half, which was your original disclosure. Rileprabard is having a tough time enrolling. Any sense what's going on there?
I mean, we're just going by what they've said, which is they're having trouble recruiting. I think objectively, if you look at, well, first off, I do think we have a great team. Our team is singularly focused or has been up until now and will continue to be on claseprubart for a while. Of course, once MG recruitment was done, a lot of our focus, more of our focus shifted to CIDP. We're very engaged with the sites and the investigators and so on. Kudos to the team for continuing our very successful track record of execution to date. More objectively, I think you just have to look at the trials, right? If you look at our trial, what are you asking of patients? A, go into an open label and you are definitely going to get drug.
You only have to come in every two weeks. Only if you respond, then you'll be randomized into the blinded portion. You have a 2/3 chance of getting drug. If you do get placebo and you start to relapse, we'll very quickly put you back on standard of care and potentially claseprubart, this drug, because we know you responded to it in part A. As I mentioned, you're only coming in every two weeks. By the way, it doesn't matter. You can be refractory, you can be stable, you can be naive to standard of care. Come on in. You have these other two trials from Sanofi. One is only refractory patients are allowed and it's blinded from the start and you have a 50% chance of getting placebo.
The other trial is blinded from the start and you have a good chance of getting placebo only. It is only for patients who are stable on standard of care. We have all comers in our trial, just like the phase 2 trial for claseprubart. I think if you're an investigator or a patient, and by the way, they have to come in every week to get two shots, I think it is. It's 600 mg, 4 mm every week for them. If you look at these trials, which one would you as a patient want to be in? Which one, if you're an investigator, do you want to be more involved with? It's the easier one.
It's the one where you're getting an active C1s inhibitor that's more potent that people, you only have to have patients come in every two weeks and everybody gets drug right off the bat in the part A open-label portion. I think that's probably, objectively speaking, it's just a more attractive trial to put your patients into. We knew that when we decided to execute on this design, we knew that it was going to make it more attractive and easier to enroll patients as well.
To your point, there's a couple of differences from ADHERE. One of them is in ADHERE, it did require patients to have been IVIG responsive at baseline, but you're not doing that.
I think I would just turn it around a little bit. You couldn't be refractory and go into the ADHERE trial. Because if you're refractory to IVIG, it's very unlikely FCRNs are going to work because FCRNs are less effective than IVIG. In our trial, because of what we saw, the impressive data with claseprubart in their phase 2, where they had 50% response rates with patients who were refractory to the current standard of care, which is IVIG, the gold standard, we allow refractory patients. In the ADHERE trial, it was only if you were stable on IVIG. They removed the IVIG and you had to relapse. You had to get sick. Only those that relapsed, once you remove the IVIG, then were put on FCRNs in part A. Remember, claseprubart did not make patients relapse.
They just switched them immediately from IVIG to claseprubart within seven days. Two points I want to make on that. A, you really enrich your trial by making patients relapse. So that 60-something percent response they had with efgartigimod in part A, that's after patients got really sick. That means still about a third of patients did not recover, did not get back to where they were on IVIG before they were made to relapse. If you look at the claseprubart data, these patients were standard of care stable, the same type of patients. They were switched immediately and half got better than they were doing on IVIG. That's really impressive. That is telling you that somehow active C1s inhibition can extract even more efficacy than what IVIG could in CIDP patients. Very, very different design.
By the way, Immunovant also made patients relapse and they were having trouble with recruitment. I think in today's world where people are more comfortable diagnosing CIDP and where there are more treatment options, conducting a trial where you make patients relapse first before you put them on your drug is not going to be feasible anymore. You have delays as Immunovant reported out earlier this year. Essentially, investigators are saying, if you are going to make me relapse patients, we are not going to participate in the trial anymore. That is why also we are able to recruit compared to other CIDP trials where they are making patients relapse.
Yeah. You are not doing a washout. Once a patient responds in part A, right afterwards, they right away get randomized to either of your two drugs. Why not do a washout?
That's what the placebo arm is essentially. If you get randomized to placebo, you're essentially being washed out of claseprubart. And there's a good chance that within the 52 weeks, you're going to relapse. Part B is really looking at a difference between patients who continue on treatment versus patients where the treatment is removed and potentially relapse.
Yeah, right. So you're saying that even if they have drugs for 60 days on board, they still have another 10 months, so to speak, to relapse.
Exactly.
The placebo. And just remind us, what have you said about powering for Captivate for the actual phase 3 component?
We haven't said anything about the powering, but again, CIDP patients, if they're responding to a biologic and you remove it, these are very sick patients. They're going to relapse. For us with a 60-day half-life, it's just a matter of time. That's why we made it 52 weeks, the part B. We want to make sure that the placebo patients have enough time to have the drug completely wash out. They're going to relapse. We're feeling really good about that. The interim responder analysis is just for us to make sure that we're seeing the efficacy we expect in part A with 300 mg every two weeks.
That's 40 patients, right?
Yeah. That's another difference to ADHERE to the efgartigimod trial. They did it after 30. We're doing it after 40 just to give us a little more of a robust data set.
Would you consider after the initial part A, is there any chance you'll consider adding a Q4 week arm to the study?
That would really significantly delay the program because that's a major, major change to the trial. For this trial, no. I think for us looking at Q4 week in CIDP, honestly, Jeroen, I think I would want to see the Q4 week in MG first. Because if we have Q4 week dosing in MG and Q2 week dosing in CIDP and MMN, that puts us in a very strong competitive as well as pricing and payer dynamics perspective.
Yeah.
A situation I probably would be a little bit more challenging is what if it's Q2 week dosing in MG, but we find Q4 week dosing in CIDP? Okay, how do you price this with the same auto injector? Look, Q2 week is a winning dosing strategy. Every two weeks, like a Dupixent, is really great. Q4 week is really upside. Right now, I see that as helping us really differentiate and take our strengths even further in MG.
Okay. All right. Let's go to the Momentum MMN study. Again, this is the phase 2 top line study expected in the second half of the year. This is going to be 36 patients, 24 treated, 12 on placebo. There's obviously a loading dose. Again, you're doing Q2 weeks, 300 and 600. There are slight differences from the AMPA study. You're doing 17 weeks. The AMPA ARDA study did 15. You're allowing sub-QIG. They did only IVIG. Again, your study is 36, 24 patients. I mean, your primary obviously is safety. Then you're looking at time to IVIG retreatment, time to relapse, and muscle strength and grip as well. Given the ARDA study, what do you expect from your data?
We expect to see similar data to what Enjaymo saw. We believe we have a more potent classical pathway inhibitor. That's really what's driving the efficacy, not the lectin blocking. It's all about the classical pathway. We believe we have a more potent antibody. Again, are they already at the top end of the curve, no matter what level of inhibition they're achieving with the classical pathway? We don't know. We are looking for at least similar efficacy. They had like a 90% response rate. It's going to be very tough to say that you got better than 90%. We are just looking to get a similar efficacy signal and a clean safety profile like we did with MG.
Then the differentiation, I mean, they're going to read out the phase 3, second half. We'll have both sets of data. They're in IV Q4 weeks. You're an auto injector Q2. You do have a loading dose. The loading doses have been IV. Is there any chance to make the loading dose a sub-Q auto injector at some point?
Oh, yeah. No, we're doing all that work. We're already talking to the FDA about that. Our eventual label, I would urge people to look at the Dupixent label. That's really what we're aiming for, where they have a, we're going to do the work to translate what does the IV loading dose mean in terms of auto injector shots. If you look at Dupixent, it says day one, give yourself two shots, and then at day seven, start taking one shot every two weeks. It will be something like that. It will be a number of sub-Q shots, maybe day one, day two, something like that. The patient would do it with the nurse or the physician to make sure that they know how to use the auto injector and so on. Then day seven, they start giving themselves that one shot every two weeks. That's what the eventual label will look like. I don't believe we're going to have, we're not aiming for an IV loading dose. We're aiming for sub-Q shots that patients can conduct and potentially in the office just to make sure they learn how to use the auto injector properly.
Yeah. Is that something that you can then introduce into the gMG phase 3 ?
Yeah, we have a full strategy of how to implement the auto injector, not just to prove bioequivalence in one trial, but then make sure that patients are using it as part of our clinical programs. All the things that the FDA wants to see before they approve an auto injector with your antibody. I mean, it's almost like a combination product. It is a combination product strategy. So we're doing everything that the FDA requires to make sure that we get that approved. Look, the good news here, Jeroen, is we're not testing some theoretical new auto injector. This is the exact same auto injector Dupixent uses, SHL Molly. It's been used for many, many years by millions and millions of patients. We're not taking any, we're not breaking any new ground here from a regulatory perspective.
We know exactly what we need to do to get it approved.
Yeah. All right. Terrific, Marino. Thanks so much for joining. We appreciate it. Good to see you. We will follow closely.
Jeroen, thank you again for the opportunity. Really appreciate it. Take care.
You too. Thanks, everybody.