argenx SE (ARGX) Earnings Call Transcript & Summary

Good afternoon. Welcome. My name is Beth DelGiacco, I'm the Head of Communications and Investor Relations at argenx. I want to first give a warm welcome to everybody in the room. We have some of our analysts and investors. Many of our colleagues here today, our Board members. So yes, a big welcome. This is our first of what we're calling an R&D Spotlight, and we're spotlighting today ARGX-119. This is our MuSK agonist. And I want to spend a little time quickly on why we've decided to host these series of events. For those who have a bit of a longer memory, early in the days of efgartigimod development, we would host these pretty often. And it was usually around rolling out a new indication or ahead of data, and we actually had the request to bring this concept back. And we're happy to do it. I think it's very much core to who we are. We're very science-based. We're data-driven. So you're going to see more of these next week. It's also the right time to really be spotlighting ARGX-119. We had a go decision in this molecule into CMS in June. So we're moving into a registrational study in CMS. We'll have data, top line results in an ALS study in the first half of next year. We're also kicking off our SMA study. Also importantly, we're going to be retiring the ARGX-119 name soon and rolling out a generic name. So that's exciting. And I can't tell you what it is today, but I can promise you, it will be very hard to pronounce. If you do after this, have feedback on what topics you would be interested in hearing, we would love to hear that. We're going to be making forward-looking statements today. We have a lot of people in the room who we'll hear from. Tim, Luc, Peter. You're also going to be hearing from Roeland Vanhauwaert, our clinical scientist on the asset, ARGX-119; and Rebecca Shilling, our clinical development lead on the asset; and Jeff Guptill, who also presented last year during our R&D day, he is the clinical development lead for neuromuscular indications. We're very fortunate to have two key opinion leaders here. We have Professor Steve Burden, he is our IIP collaborator on ARGX-119 and a true pioneer in neuromuscular junction biology. He identified many components of the neuromuscular junction. So you'll hear from him during a panel. And Dr. Ricardo Maselli, he is a neurologist from UC Davis, and he treats and is a researcher in CMS. So you'll also hear from him during a panel session. We're going to start with, of course, where we always start, biology, and talking about how we designed 119, then we're going to move into the proof of biology from that Phase Ib study in CMS. And then we're going to look forward, looking at what this pipeline and a product potential could be across all indications. And of course, we'll save time to hear from Tim and our Q&A session. We rolled out our Vision 2030 last summer, and you'll all remember this, we want to be by 2030 in 50,000 patients across all of our medicines. We want to be in 10 labeled indications. And we want to be in 5 new late-stage studies with new molecules. And actually, this was on top of efgartigimod and empa. So 119, now that it's in Phase III is contributing to that 5. But really, the message today is that 119 is part of this vision. It's in a Phase III. It's in 3 indications, and we're working on bringing this transformational biology, potentially transformational medicine to patients. You know our innovation playbook. This molecule is really a prototype of our IIP. We like novel biology, and we like it around foundational components of the immune system. Why we like novel biology? It typically means we're first. So when you're first, that means that we can really bring transformational benefits to patients. Why do we like it to be around foundational components of the immune system? Because that often means that it's relevant in many different indications. So we have this pipeline and a product potential. We rely on external researchers, typically academics to source this novel biology because they have decades of experience around a specific disease target, around a pathway. This is experience that we could never recreate in-house. So this is taking with our antibody engineering capabilities, their depth of disease biology insights. Together, we can really drive transformational outcomes to patients. It also positions us not only to be first in class, but to be best-in-class, hopefully, with our molecules. All of our pipeline programs actually emerge from our IIP and there's amazing stories of collaboration around our IP, and 119 is no different. I think the key here is Steve brings his expertise in the neuromuscular junction biology. And you're going to see how this collaboration comes to life during the panel and that interaction, these are not transactional relationships. These are really partnerships through the entire life cycle of the molecule. And the reason we do this is because we believe there's power and the know-how of the person. So it's not just about the molecule. So Steve bringing his depth of biology. And together, we were able to kind of apply this know-how to how we're going to bring these medicines to patients. This approach is also creating an innovation ecosystem. This is something that is dynamic. We make connections between our collaborators. We can tap into the connections of our collaborators into their relationships around the science, around the disease and really the currency of this economic or the economic -- this innovation ecosystem is data. And we can take the data that we generate in the clinic. We can take the data that we're generating in the real world and bring that back into our development plans. And that's what that ecosystem really looks like in action. I want to spend just quickly, before I move over to the official program, on 5 takeaways that you should remember from today. First off, we are pioneering innovation with ARGX-119, pioneering. You're going to hear that word a lot today. We're pioneering a new understanding of the biology at the neuromuscular junction. We're pioneering in our development for CMS to ALS, SMA. And this is really important to kind of how we bring these molecules forward. We're going to be doing it, and we're going to be doing it fast. The second thing I want you to remember is actually around 119 itself. 119 is a MuSK agonist. It is serving as MuSK in this situation. And MuSK plays two important roles at the neuromuscular junction. One is that it serves to cluster and anchor acetylcholine receptor. So it actually primes the muscle to receive that signal. The second thing it does is actually at the presynaptic differentiation, where it induces the signaling. Why this is important is you're going to see how our development plan plays out that we are creating a development plan across both modes of action. The third thing is that CMS is severe, and there's no precision treatments. You're going to hear from a patient during this session. And we talk often about this disease being ultra rare. It is also ultra severe. And with no effective treatments, there's low awareness. There's a long path to diagnosis and there's also not enough systematic genetic screening. And I think that what we're hoping to do today is really talk about all the learnings we're going to generate that hopefully will bring more awareness, more innovation to CMS, which brings me to number 4. We're taking a very data-rich, bold approach to how we're developing CMS. Remember that with 119, we didn't have a Phase I where we could take a biomarker and check that proof of biology with our Phase I study. That's what we did with our Phase Ib. So not only do we have now the justification to go from the Phase Ib into a Phase III, but we also have proof of biology of the molecule. It's doing what we expected it to do. The other thing I want to mention on our development approach is we're really arming patients here with -- well, actually it's with digital sensors because we don't just want to get data from they go into the clinic and get their assessments. We want data from when they're home and going through the regular activities because that's going to be how we're going to learn the most about this indication. We're going to take this exact same playbook and apply it to what we're doing for the rest of our ARGX-119 because I think this is how we're going to maximize the opportunity and really elevate the positioning of 119 in our portfolio right alongside efgart and empa. So that is all I have for today, and I'm going to bring up Roeland to talk about the biology.

Hello, everybody, and thanks, Beth, for a nice introduction. I'm really excited to show you some of the science behind the program and also to explain to you why we've chosen MuSK to target. So let's dive right in. As you can see here on the left of the screen, the motor neuron is innervating a muscle. This connection is called the neuromuscular junction, as you probably know. The signals from your brain go to your spinal cord and then go into your muscle to have a muscle contraction. This allows you to breathe, to function and to walkabout, everybody has this. So argenx really pioneered development at that neuromuscular junction already with efgartigimod in myasthenia gravis; with empa, for example, in CIDP; and now also with ARGX-119 in CMS, ALS and SMA. Also importantly, some IIP programs are in the pipeline targeting novel disease biology at the neuromuscular junction. So how does this neuromuscular junction biology come about? It's just -- it's not there in 1 day. This was actually the work of Steve Burden. He uncovered from the 1970s onwards this full pathway on how a neuromuscular junction is formed, but also proteins that are important to play at that neuromuscular junction. He unraveled each of one of these and how they function. So there is, for example, rapsyn, agrin, MuSK, Lrp4 and Dok7. So why did we choose MuSK then, or muscle-specific kinase? Well, MuSK activation is really important for the formation, the maintenance and the maturation of a neuromuscular junction. I'll guide you through this. Upon MuSK activation, acetylcholine receptors will cluster together on the muscle surface. During embryonic development, these batches of acetylcholines -- of acetylcholine receptors are now primed to receive the innervating motor neuron. When MuSK gets activated more, this connection gets consolidated, and the neuromuscular junction will mature to a full functioning state. Also later in life, MuSK activation remains important because if it's -- if you don't activate MuSK anymore, the neuromuscular junction will disassemble, and you get denervation and all kinds of neurovascular diseases. So let's dive in into the signaling pathway. And what happens when you activate MuSK with a MuSK agonist antibody, for example, ARGX-119. [Presentation]

So I hope you understand the signaling pathway now and I don't need to explain it again. But it's crucial that impairments in every step of this signaling pathway cause disease. You might know myasthenia gravis. The autoantibodies, either block clustering of acetylcholine receptors or block MuSK activation in MuSK-MG. For congenital myasthenic syndromes, this is the same. Mutations in any of these proteins will cause muscle weakness. We are studying Dok7 congenital myasthenic syndrome and they cause truncated Dok7 protein, which can lower MuSK phosphorylation, and a MuSK agonist exactly repairs that kind of deficit. So I hope now that you understand that MuSK is a master regulator of the neuromuscular junction and that targeting with an antibody, the extra cellar domain is quite easy to turn on the system and activate MuSK and get mature functional neuromuscular junctions again. So how did we develop ARGX-119? Well, it was not that easy. It took us 3 iterations. And the idea came actually via the IIP program. Dr. Huijbers and Dr. Jan Verschuuren from the Leiden University came to us with an idea. They had an idea that they could turn the inhibitory MuSK antibodies from MuSK-MG into activating antibodies. So using the patient's own inhibitory antibodies, converting them to activating antibodies as a therapeutic treatment. That was quite innovative. So we jumped with them in that kind of antibody campaign. However, we hit some roadblocks. And then we started discussing the biology with Steve Burden. We'll get more into that into the panel. But this co-creation of the 3 groups together really laid the foundation to the development of ARGX-119. The biology expertise of the neuromuscular junction, the biology expertise of mask and the antibody capabilities of argenx that was really a true combination that made ARGX-119. So ARGX-119 is derived from the SIMPLE antibody platform. It is llama derived. And as you can see here, it will activate MuSK at the frizzled domain which in turn clusters the acetylcholine receptors. That clustering is important to have efficient transmission across the synapse. In addition, there's a second mode of action of a retrograde signal communicating with the motor neuron that will modulate the synaptic architecture. So we were quite excited of having this lead antibody candidate. So we went as quick as possible to dose the first healthy volunteers. That happened in 2023. Already then, we had in our mind that we wanted proof of concept in Dok7 congenital myasthenic syndrome because of the biology that I explained to you. So not to wait for that. We already started the natural history study, identifying these patients and characterizing their deficits. When we have sufficient safety data from the healthy volunteer trial, we immediately started dosing these patients with ARGX-119. And within 1 year, we finalized the treatment period and analyzed the data, and I'm super excited to show you today the clinical trial outcome data of that trial. But first, let me tell you a bit more about the biological rationale of congenital myasthenic syndromes. Congenital myasthenic syndromes is a -- that is a genetic disease. It's characterized by mutations in any of these neuromuscular junction proteins. There are over 35 genes identified. And here, we've shown you the most commonly identified proteins involved. For those, genetic tests exist and are used to establish the diagnosis in parallel with the muscle weakness observed in these patients. We will focus on Dok7, LRP4, MuSK and agrin because those are involved in the clustering of acetylcholine receptors and likely amendable for treatment with ARGX-119. As I told you, mutations in the Dok7 gene will result in lower MuSK phosphorylation. And activation of ARGX-119 actually fully compensates for that. We've tested that in a mouse model. This was the work done together with Steve and his postdoc in the lab. And as you can see, he had already developed a mouse model for Dok7 congenital myasthenic syndrome. These mice bear the most common patient mutation inside the genomic locus of these mice. So it's a humanized model. Importantly, because they have lower MuSK phosphorylation, they have smaller neuromuscular junctions, which is also observed in patients. As you can see in the lower -- in the graph on the left, if they're untreated, these mice have a very severe phenotype and die within 2 weeks after birth. If you now dose these mice pups at postnatal day 4 with ARGX-119, you see an immediate increase in body weight, suggesting that we activate the neuromuscular junction, these mice get stronger, can eat again and also increase in body weight and just a single dose has a potency to keep these mice alive for over 60 days. What we also saw is that these smaller neuromuscular junctions now get healthy again, fully matured. However, after 60 days, the antibody is cleared out of the system, the neuromuscular junction collapses again, and you see this little drop in the middle. The mice lose their ability to walk around, lose their ability to eat and they drop in body weight. Upon that time, we give them a second shot of ARGX-119. And within the week, they start to eat again, increase in body weight and live for another 100 days. Interestingly, on the graph on the right, you see that prior to the second dose, these mice cannot run on a rotating wheel. However, within 4 weeks, these mice are fully capable of running on this rotating wheel similar to healthy mice. Do keep this graph in mind when you look at the Phase Ib clinical trial data. But first, let's have a conversation with Steve. Steve and Peter, do you want to come up?

Thank you, Roeland. Welcome, everybody. Special thank you to Steve, our collaborator, so shortly after your surgery. And glad to see that your neuromuscular junctions are still fully functional. Steve, Roeland already introduced you as the world's expert in neuromuscular junction biology. He's covering a lot of these important elements. How did you start that research? And how much do we know more now than when you started?

Well, I've been working in this field a long time, almost 50 years. And at the time that we began, we knew a lot about synaptic transmission. And as Roeland and Beth have pointed out, that involves acetylcholine release nerve terminals and activating acetylcholine receptors and ultimately causing muscle contraction and muscle movement. But we really knew very little about the mechanisms around forming the synapse and maintaining the synapse. And as has been pointed out, over that time, we've identified and studied how -- identified agrin, LRP4, MuSK, Dok7, rapsyn and not identified them but tried to understand as much as we can about how they work. And understanding in detail about how these molecules work is critical for developing, understanding disease and developing therapies for treating them.

And you immediately have a therapeutic focus in the back of your mind when studying all these components.

Sorry?

You immediately have that therapeutic applicability...

No, no, no. At the time we were carrying out these studies, the type of translational studies and clinical studies that can be done now were not in our mind at all. We were just trying to understand the basic mechanisms for building a synapse and maintaining a synapse. And I think the clinicians who are working on diseases, and you point to -- we've talked about congenital myasthenia and Roeland has pointed out their mutations in multiple genes that cause congenital myasthenia. But at the time that we were working on this, the clinicians were simply classifying these patients as having some form of congenital myasthenia, and they didn't know what the cause was. And at the time that they were studying them, unlike other diseases like Duchenne muscular dystrophy, where people were for genetic mapping and walking, trying to identify the culprit genes, for whatever reason, those clinicians didn't do that. And so it was a [ follow ] area where the disease was described, but the cause was unclear. And I've learned from the clinicians who work in this area, what they did is they waited for labs like mine and others to identify genes like MuSK and LRP4 and Dok7. And once we had identified those genes and studied them in mice, they then used as candidates to go back to those patients and ask were they defectively mutated in those patients. So the discovery of nearly all of the patients with congenital myasthenia came from these basic science studies that we were doing in the absence of any real pursuit of the diseases.

We learned a lot. I think we identified a lot of these components mutations as a cause of several diseases. What are according to you, the biggest unknown on that neuromuscular...

The biggest unknowns. I'd say there are -- in my view, two, is we pointed out that once agrin binds LRP4, stimulates MuSK and Dok gets recruited, then magic happens. Then the synapse gets organized, assembled, stabilized. But we really don't understand the mechanisms that follow the activation of MuSK and lead to Dok7 recruitment and activation. That's really a black box.

So that's what we scientists that's where magic happens. We don't know it.

That's one. And I'd say the other one has also been pointed to is we found some years ago that you need to activate MuSK and you need to cluster LRP4 in order for LRP4 to serve as a second role. It's not only a receptor for agrin, in itself, the protein signals back to motor neurons and stimulates the differentiation and attachment of motor nerve terminals to the muscle. But we don't understand how that works. We know LRP4 is critical, but how motor neurons respond to it, recognize it and then stimulate differentiation and attachment. That's a big mystery to me.

And if I may -- can add, now it's quite interesting. We have these two big questions. And we're not developing this therapeutic because of the therapy, but we're still doing the research in the lab to really understand these mechanisms. So hopefully, more to come soon.

So that to me is the power of IIP. I think you start with a very concrete project. And based on that collaboration, you can expand and further pioneer the biology. Roeland, how did we get connected with Steve?

How did we -- connected with Steve. I know you so long already, it's kind of hard to think when we first met. But I think the connection came when Maartje Huijbers from Leiden came to that roadblock where unexpectedly male mice started dying. And she worked in your lab. She was a postdoc there. So she said, "Yes, I mean, let's just go talk to Steve." And I think we few -- we even visited you in New York and we started chatting and you even had these synthetic antibodies already laying around. And then we did then the second antibody campaign. So yes, it was quite natural actually as scientists to discuss new data and unknown data.

And what that moment is -- why did we decide to continue the journey despite that setback?

Well, Steve had a reasonable explanation for why the patient-derived antibodies did not work because they bind to this upper most domain and they block likely the agrin signaling. And you already had new synthetic antibodies to this lower frizzled-like domain, which does not block that signaling. So it was kind of -- yes, you move on, you build on to the new knowledge.

Steve, anything you remember argenx knocking on your door?

I concur with Roeland. I mean, again, this comes back to just basic fundamental hard science. We had done a lot of work trying to map what pieces of MuSK did what? What was the function not only of the molecule, but the different domains. In the first IG-like domain, it has 3. We knew that was critical for binding LRP4. Whereas we had studied in mice, we had mutated and deleted the frizzle domain, and that was dispensable. So for us, this was an aha moment. If you're going to tinker with MuSK by stimulating it and using antibodies, you want to target the frizzle domain, it's dispensable. You don't want to touch the IG-like domain. It's critical for it to bind the LRP4. So for us, it was obvious that if you want to make agonist antibodies, you want to target the frizzle domain and avoid the first IG-like domain.

And Roeland, what do you experience in that collaboration? What made it unique for you?

I mean, I came from academia where you just discussed everything openly. And I had the same experience with Steve. We -- although we were on different continents, we would just hop on a call whenever and discuss new data. I think we're sometimes on the phone 4 or 5 times a week. It was quite exciting. We had a lot of data coming out and trying to interpret all those kind of things.

And Steve, for you, did we bring any value in your research?

No, absolutely. I've had a lot of interactions with biotech companies over the years, quite a lot with Genentech and quite a lot with Regeneron. This was unique. So the nature of the communication, both being regular and being very open, neither of us was holding our cards close to our chest. We were very open about what we were doing and what we should try to do. We were very open to changing views. It made it very easy not only for myself, but in an academic lab, you depend upon your graduate students and your postdocs who are both heavily involved in the work and committed to the work, and they need to feel that they're involved and they're not being set aside. And they're aware of what's going on both in the lab and outside the lab. And in this case, it worked really well, and it hasn't always worked like that. So this was a real delight. And this was with Roeland, but also with everyone in Roeland's group. These are just really excellent scientists and easy to communicate with, pleasure to talk to. And...

And it progresses the insights. I think you've -- we now have the 119 data. You talked about the unknowns. I kind of understand we're doing further research there. Something you can tell about that?

I'll let you start.

I mean, yes, it's -- we're doing further work. And out of this new ideas come and so new discovery programs are set up. It's interesting to see that feeding also back to new discovery programs. Yes.

Yes. So basically, a 219 molecule is coming our way.

You're the CSO, you can say that.

Well, the one thing I can add to that is we're -- we've mentioned other diseases other than CMS or other forms of CMS as well that are reasonable targets for 119. I've also mentioned that I think one big black box area, an important area is how nerves respond to LRP4 to stay attached. And there's -- we're working on this together with argenx to try to find out more about what that receptor is, how it works and whether it can be targeted and help sustain attachment of nerves to muscles. To me, that's a big area.

Maybe I would end with this question, Steve. So you've seen the CMS data, obviously. What was your feeling? And how do you see the opportunity for 119 in CMS and beyond?

Yes, that's interesting. We published a paper and Roeland referred to it an earlier version where we had generated our own in-house antibodies to MuSK that had -- they rescued the mouse model, but they had these problems of aggregation and having off targets. But even at that time, there are patients with CMS who read the literature and there were several who would write me e-mails. And they are touching because as Beth pointed out, this is a very severe disease. People really have a difficult time with this. And so when I started to see the clinical data, there's just very little substitutes for them. It's just very humbling to think that anything you can do in basic science to really help patients, it's powerful. So was I surprised? The preclinical studies we had done, I thought were really, really solid in every way, both for safety and effectiveness. And there are a lot of similarities between -- maybe not in all the diseases, but in the neuromuscular junction between mice and human. So I really believe it worked so well in mice in so many different situations that it was going to work as well in humans. But you never know until you do the experiment. So was I shocked? No. Was I delighted? Absolutely. Yes.

And how do you see the opportunity beyond CMS? What's your opinion on the biology in different indications?

So I've worked a long time on ALS. And our first actually adventure in trying to tinker with MuSK was not in Dok7, CMS, it was in ALS. And we published a study in 2012 showing that if you activate MuSK in a mouse model of ALS, you can improve the phenotype of those mice. The disease is less severe. And then we went on to use our -- an antibody to MuSK to activate it in the same ALS mouse model. And it did the same thing. It lessened the severity of the disease symptoms. So we all know this is a terrible disease. So I certainly have great hope that the ARGX-119 can provide some benefit to ALS patients. As we all know, there is really nothing that's available for ALS patients right now.

Okay. Thanks. I would like to thank you, Steve, for being here in this panel, of course, for being a very important collaborator to us. Also thank you, Roeland. I will invite Luc Truyen, our CMO, on the stage to talk about the clinical data. Thank you.

Yes. Well, isn't science cool? So I'm going to digress for a second. When I was in the process of talking to argenx, of course, you get information about the assets in the portfolio. I knew about VYVGART. I learned about empa. But then the third one that I -- it was just a lead selected, I believe, at that time. And I said, argenx did what? They built an agonistic antibody for MuSK. And I thought to myself, if they actually did that and that works, that has a lot of potential. Well, that was 4 years ago. Really exciting. Roeland already spoke about why we studied Dok7 CMS. But let me give some color as to what it means to be a patient with Dok7 CMS. The name, congenital myasthenic syndrome already contains a lot of information. It's congenital. That means these people don't actually know what normal is. Myasthenic fatigability of the muscles and the reasons why it should be clear from what you heard before. Now these syndromes are very rare, but they are very severe, as I -- was already said. It can start at younger age or in adolescence, but it does progress slowly, and it ultimately leads to significant disabilities. These disabilities will often require the use of aids, wheelchairs and so forth. And what also creeps in there is respiratory problems, difficulties talking. And you will see an illustration of that in the patient video. And I would like you to look at that video and know that there's no FDA-approved treatments for this particular patient. Can I have the video, please? [Presentation]

Well, if this doesn't give you a sense of urgency, I don't know what will. So using our playbook of innovation, building this molecule, thinking about how can you accelerate development we embarked on the timeline that already was shown, to make sure that we could answers as fast as possible too, can CMS Dok7 be helped by 119? And we used our playbook that we talked about last year's R&D show. And if I apply that to what we did in this clinical trial, so there are very few patients, right? So how do we maximize the impact of every single patient in our evidence generation? First thing is, of course, because there were no dedicated measures, we borrowed from our learnings in autoimmune MG, we took those scales with us. But we felt that might not be enough because we actually don't know what they will pick up really in a setting of an interventional trial. So then we said, we want to have more documentation of what happens outside of the clinic. Let's apply digital sensors, actigraphy of these patients. We also thought, well, ambulation is a problem. Let's use an instrument that's more well known, the 6-minute walk test. But let's pair that with some more sensors to see, well, what is actually at the source of any change. Part of our playbook is also empowering patients through the whole arc of setting up the understanding what the need is, how we're going to now measure with an intervention and success and get the story of the patient back. We first, as was already mentioned, invested in a natural history study to better understand how do these scales evolve and therefore, get some more information for our intervention part. And then we also bookended that with exit interviews after the intervention as to what patients really felt. I already spoke about small numbers. At this stage of development, typically, you try to do dose ranging, trying to find minimum effective dose, maximum dose and those things. Given the low numbers, we said we cannot do that in this traditional way, and we are going to do intra-patient dose escalation. Our [ Clean Pharm ] people assured me that they could get to an exposure response relationship. And therefore, that's what we did. Now this natural history study, it's not the first time we make an investment in this. We think it's an essential tool to understand the disease you're actually trying to effect. And what we learned here, and this goes to the normalcy of these patients, is these patients are actually chronically severely disabled. And 85% require devices to help them around. And therefore, that realization to us said, "Wow, yes, yes, I hope we can see an inflection point with our treatments." We're going to evolve this natural history study. We're going to include other genotypes. Also look earlier, as Taal said, it started as a baby, of course, to get more information. You're going to see the results of this Phase Ib clinical trial. And at first blush, you could look at this and say, "Well, that doesn't look very innovative at all, two parallel groups, small sample." But it's again because of the factors I said before, a large number of measures, high frequency of measuring, intra-patient dosing, that we can get real answers out of this small, I'm not going to say it's big, experiment and that allowed us to make -- well, the decision we made you already know about. But Rebecca will tell you more about the data that underpin this. Rebecca? Thank you. Thank you.

All right. Thanks, Luc, and welcome. It's great to show you our data. So I'm going to, on behalf of the team, walk you through our Phase Ib study results. So in this study of Dok7 CMS patients, we enrolled 16 patients, we were able to see a favorable safety profile and looking at measures focused on mobility and weakness, we see what we think of as proof of biology. And taking these data, we believe we have the reason, and we can also convince the regulators that we're ready to advance into our registrational study. So on this slide, I'm showing you the demographics and the baseline characteristics. So we enrolled 13 patients in the 119 treated group and 3 placebo patients. And what you can see is the majority of the patients were stable on beta agonist as the patient Taal mentioned, albuterol is one of the anecdotal medicines that's used in these patients, patients do require assisted devices and the patients were enrolled, we targeted patients who had evidence of weakness. And on the next slide, I'll show you a little bit more about this. So using the quantitative myasthenia gravis score, the QMG, you can see the components that are most affected at baseline, shown here by each component and then the colors signify the severity. So you can see on the left, looking at Orange was just moderate, or severe was red, that the patients were most affected in their arms, legs, and the weakness that they had in these components as well as the head, whereas the speech and swallowing components of the QMG were less affected. So we focused on where these patients are most affected to see if we can have an effect with 119. And on the next slide, what you can see here is the levels of weakness in the legs and the arms. So in the dark blue is the 119 treated group and in the gray is the placebo patients. And so we see, over time, as we're dose escalating within the patients, an increase in the ability of these patients to hold up their legs and their arms. So in the legs, the patients are getting towards what is considered normal on the QMG, which is 100 seconds. And on the right, with the arms, there is a doubling of the amount of time they're able to hold their arms up, but they haven't quite reached the normal range, which would be 240 or above. But this is really showing that the areas of this limb-girdle phenotype that we're improving, and their patients have less weakness. But we wanted to understand more about how they're functioning, how are they doing with walking. As Luc mentioned, we used the 6-minute walk. So on the next slide, I'm showing you the total distance walked. And we -- when we saw these data, now we're focusing here on the ambulatory patients. So patients who didn't report a need for wheelchair at baseline, but were able to -- and were also able to walk at the beginning of the study. And we see that over time, in the dark blue in the 119 treated patients were getting a median distance of 75 meters. So in other neuromuscular diseases, which I have here on the light blue is the clinically -- what's been considered a clinically meaningful range, which is between 15 and 50 meters. So this was quite exciting for us to see this magnitude of a change in the small study in these patients. And the other thing is that we don't think that we don't see a plateau, so we're continuing to following these patients off drug. So we'll have more data in the future. But we're also encouraged that if we can in the next phase when we treat for longer that we'll have an even greater magnitude of improvement. Now the other thing we did, as Luc said, is we had patients wear sensors so that we could really measure different parameters of their gait. So we can understand not just that they're walking longer and maybe actually having more endurance, but also are they moving better. And so one of the measures that we looked at is cadence. So this is a measure of the steps per minute. And so we see that in the 119-treated patients that we're getting a difference in the cadence relative to placebo. And again, extrapolating for other diseases where a clinically meaningful change of at least 5 steps per minute is considered clinically meaningful. So this also shows us that we are really starting to see patients moving better and able to improve their gait parameters. Now in order to put all the data together, we did also look at a correlation between what we see with the distance in the 6-minute walk. And with the leg weakness, and we see that the patients treated with 119, there is a positive correlation, whereas with the placebo patients, we do not see the same pattern. So this helps us understand how the data are all being synthesized in these patients and the coherence in the data. Now another way that we wanted to look at these patients because as you heard from Taal that she has a hard time with her day is what is happening outside the clinic. Because it's this one thing for us to measure how they're doing in the clinic, but it's another thing to take a look at what's going on at home. And so in the next slide, we -- this is data from patients wearing just a very simple actigraphy watch that could measure multiple different parameters of activity at home. And the -- each column is an individual patient, and the A marks the patients on active drug 119 and the P marks the patients on placebo. And this is a hierarchical clustering that has separated the patients into different groups. So you can see on the left are the patients who reported wheelchair use and one cluster and on the right are the patients who didn't report any need for a wheelchair. And so what we see just overall is that in the patients who were on 119 that we actually see in the red, an increase in activity. The blue is a decrease in activity. So that was very encouraging. But then when we look at specific clustering in the green box in the patients who reported wheelchair use, what we see is an improvement in short-term activity parameters, such as the best 1-minute cadence. So that was very interesting. And then when we look in the dark blue at the patients who didn't require wheelchair assist, we see an improvement in the what are called longer-term activity parameters like their best 4-minute effort. So this is interesting that we saw these patients clustering differently depending on their baseline activity, but in both sets of patients, we're seeing improvements. And then the last thing I'm going to show you on here is the rectangles around the patients who are the 3 placebo patients. And we really don't see that they are really changing much. And if anything, some of them are actually decreasing. And this is consistent with what is reported in the literature with these patients that over time, they're relatively stable and also consistent with what we have seen in our natural history study. Now this is exciting data for us to see the activity of 119, but also important for this -- the preliminary of Phase Ib study is the safety profile. And we actually saw really favorable safety profile in that 119 IV was well tolerated. We didn't have anybody discontinue due to an AE. There were no serious or severe AEs, nobody developed anything that would suggest that we were inducing a myasthenia type phenotype. So this is very encouraging for us as we move forward with 119. And this last data slide, I'm going to show you is we want to understand in any study, especially in a small study in an ultra-rare condition where we don't have established end points is what are the patient -- what's meaningful to the patients, is what we're measuring actually something that matters to them. And so we used a qualitative analysis of the patient exit interviews to understand if they felt that the study was capturing what mattered to them, did they feel like they were improving, were they satisfied, was the watch acceptable to use and if they were having increase in mobility. And all of these responses were positive, and you can see some very lovely quotes that are quite inspiring just like that patient, Taal. So putting this all together, we feel that we have the confidence and the data and the story to tell to the regulators so that we can move on to targeting Dok7 CMS patients aged 12 and up in a pivotal trial, and we're going to expand to the additional subtypes that its biological rationale. We're going to focus on the 6-minute walk as the primary endpoint, but we also want to go discuss with the regulators about our digital health technology-derived end points. And we're also going to expand our natural history study so they can really understand fully the disease. And with that, I would like to bring up one of our collaborators as well as Luc for our panel. And thank you.

So while we're getting arranged, I'll just introduce the session. So for this part of the session, we're going to have a panel discussion about -- a little bit more about the CMS patient journey and experience as well as talking a little bit more about our data and future strategy. So I think we've already met Luc and Rebecca already, but I think maybe first, we'll start with introducing our guest here, Dr. Maselli, and I've been very fortunate over the years to learn a lot from Dr. Maselli from prior talks that he's given educating neurologists about congenital myasthenic syndrome. So maybe you can just introduce yourself and talk about your practice.

Sure. Thank you, Jeff, for such a nice introduction. My name is Ricardo Maselli. I'm a Professor of Neurology at the University of Chicago -- University of California, Davis. I was trained at the University of Chicago. And so most of the things -- and that's where my interest in congenital myasthenic syndrome started more than 30 years ago. So I've been very active in the not only in the clinical practice of congenital myasthenic syndrome but also in the research of congenital myasthenic syndrome. And in my interaction, I've been interacting a lot with Steve Burden for many years.

It's a small community, I guess.

Yes.

Can you tell us a little bit about your clinical practice and the patients that you see in clinic?

Yes. So most of my patients are neuromuscular patients. And although I do have -- as a general neurologist, I do have responsibilities of seeing patients in the hospital as well when I'm on call. But most of my patients -- my outpatients are patients with congenital myasthenic syndromes. And I see all kind of congenital myasthenic syndromes, including the most rare diseases, extremely, and we're really talking about, I would say, Dok7 is a rare, but it's not that rare, actually. I would say it's probably one of the most common form of congenital myasthenic syndromes. But I've seen patients with very, very rare, ultra-rare diseases that we are talking about 3 or 4 cases reported in the whole world. And some of my patients are contacting me, some of the parents are so motivated because the reason that we have seen that this is a terrible disease. And some of the patients are really so invested to the point of contributing to my research. And so in answering your question, yes, most of my practices in congenital myasthenic syndromes.

And before I joined argenx, I actually remember referring a patient specifically, this is Dr. Maselli for a second opinion. So maybe you can talk a little bit more about the diagnostic journey of that patient's experience with congenital myasthenia.

Sure. Well, in doing so, I'm kind of aging myself because I've been doing this for so long. And so that remind me when I was a fellow at the University of Chicago. And that was -- I think that was the golden age of congenital myasthenic syndromes, even we didn't know the generics of congenital myasthenic syndromes, up to, I will say, the 1980s. Congenital myasthenic syndromes, we knew one thing that were completely different from the autoimmune type of myasthenia. There were patients that they're clustered in families with congenital myasthenic syndromes, we knew that autoimmunity playing overall in that disease. And ergo, we didn't treat them with immunosuppression. But we didn't know the types. So the first challenge was to separate all these bunch of patients that were referred -- were reported over the years and say, well, what is this and what is this and what is medication for each type of the congenital myasthenic syndromes. And the only way that we had those days to make the diagnosis is to do a muscle biopsy, take the muscle actually, and keep the muscle alive actually, you can do that. And in the muscle alive, [ get ] microelectrodes on the neuromuscular junction, stimulating the nerve. And by doing that, you can establish the amount of [indiscernible] or synaptic vesicles being released and the amount of what was the impact of a single vesicle at the neuromuscular junction. And by establishing that ratio, we can determine if the condition was pre or post synaptic, which was a major accomplishment. Then we also had electron microscopy, so we can correlate that with electron microscopy. And so we established in doing so a group of patients with pre- and postsynaptic and we knew the medications that those patients will respond depending where the condition was pre or postsynaptic. So that was really the first accomplishment and was the first challenge that we faced back in the 1980s and beginning of 1990s. Was not only until 1995 that the first linkage to a genetic defect was established and was the variant that is called the slow channel syndrome. And up to that point, from 1995 and after that, there were really not too many laboratories in the world that can do the genetic. And there was the Mayo Clinic and most of the things that we are learning was from the Mayo Clinic, from Dr. Ed Lambert and Andy Engel. And then it was our group at the University of California, Davis that we had the technology to determine the genetic defect. And that was almost I would say for 10 years, we were really the only two laboratories in the United States that have the interest and the technology to determine the genetic, what was the genetic cause of these syndromes. But the challenge that we had at that time that only about half of the patients that we perform the genetic testing, we can determine the condition. There was a huge amount of patients with congenital myasthenic syndromes we cannot determine what they had was not only until the 2010 that next-generation sequencing became available for clinicians. And that made a tremendous difference in the field of congenital myasthenic syndrome. There's a tremendous impact, just to tell you, and I think that in the past, we interacted about a patient with presynaptic congenital myasthenic syndrome. And at that time, before next-generation sequencing, we knew linkage to only one. And now we have 15 conditions that we know associated with presynaptic congenital myasthenic syndrome. So that made a tremendous impact. And I think that the next challenge is what we're really facing now is the treatment.

Yes. So maybe, Luc, do you have a comment on that?

Yes. Just in your experience, how long does that typical journey last for the patient to get the diagnosis. So the technologies have evolved, but from a patient journey, how -- what do you think that is?

Yes. No, I can echo what the patient mentioned that many of these patients, I would say, really even maybe patients with DOK7 syndromes as the children has been characterized as lazy that they really didn't want to practice. And some of them we were talking about some of my early patients that they were really punished at school because they were really -- they interpret as they were lazy that were not really wanted to do any physical activity and those sort of things. And so there is a big range. You can see patients like that, that they have been a big, big journey without any diagnosis until -- and then in comparison, we have patients that we establish a diagnosis now with next-generation sequencing in months. It's born in three months after we have the diagnosis.

So I think we can also say it's underrecognized though, still even with the genetic testing available.

Very, very so. I mean there's a lot of patients and that the awareness of congenital myasthenic syndrome is not universal.

So maybe we can just turn quickly to treatments that are available currently. So can you talk a little bit about the limitations of treatments that are currently available for CMS patients?

Sure. And I think that this is really relevant to the introduction of this new therapy. And the treatment of congenital myasthenic syndrome is different from the one that we apply for autoimmune myasthenia. And autoimmune myasthenia gravis, we use immunosuppression and inhibitors of fetal receptors like efgartigimod. And -- but for continual myasthenic syndromes, the only strategy that we use is to increase the amount of acetylcholine. And we use very old medications, inhibitor of the cholinesterase in other words, acetylcholine is just taken back to what has been explained. Acetylcholine is being released and interact with the receptors. And once that interaction occurs, then the muscle is activated. So -- but if there is a problem with the receptor or the problem with the aggregation of receptors, one way that we can correct that is by increasing the amount of acetylcholine. So one strategy that we use is to inhibit the enzyme that hydrolyzes acetylcholine. That strategy sounds like innovative, but it's nothing new, have been around for almost 100 years. It was developed by Mary Walker, a Scottish neurologist that established that medication inhibitor of the cholinesterase work in myasthenia and in congenital myasthenic syndromes. The other strategy is to force the nerve to release more acetylcholine. And we do that by inhibiting another channels, potassium channels, and that increases the polarization time and forces the neuromuscular junction to release acetylcholine. So those are two things, and we tend to use those in presynaptic form of congenital myasthenic syndromes. Now the problem with that is there are certain forms of congenital myasthenic syndromes that acetylcholine is not good. And in a long run, it produces more damages than good. Because there are two forces. I think that Dr. Burden has alluded to that. There is the acetylcholine and the forces that are acting to get the receptors aggregated, the nerve itself, agrin, MuSK, [indiscernible] MuSK and all of that. But there are some forces that tend to disperse the receptor. So they're kind of counteracting. And acetylcholine, believe it or not, is one of those negative forces, right? So by treating patients with DOK7, LRP4, agrin, with these medications that are available now, we are really in the long run, producing more damage than good things, okay? So this is something that needs. It is a realization that we can little by little because patients initially, if you give an inhibitor of the cholinesterase or 3,4-diaminopyridine that releases more acetylcholine, they really like the medication because they have an immediate gratification. They have more strength. Then in the long run, it has a very negative impact. So I'm so really so impressed and so related about this new way because it's the first time that we see something that goes away from that equation of increasing the amount of acetylcholine, because acetylcholine in this particular group of congenital myasthenic syndromes is not really good. There is another group of -- there is a specific form, you know that, that is of congenital myasthenic syndrome, which is the deficiency of cholinesterase. There is a condition of a congenital myasthenic syndrome that results or is characterized by deficiency of the cholinesterase, okay? So evidently, if you give more acetylcholine to a patient that doesn't have the enzyme to process the acetylcholine, those patients then get worse. Whether this medication will work in those patients, I do not know, but this is in the future, if we -- if you are thinking about expanding the...

Maybe something to think about in the future. So maybe I'll turn to Rebecca now. So, Rebecca, have you -- so up to this point, you showed the data up through 12 weeks of the Phase Ib. Can you talk a little bit about what the team hopes to learn like from the rest of the data that's accumulating now?

Yes, yes. So one of the things we learned from the mouse model is, as Roeland showed, you dose the mice and then the activity, at least in terms of looking at how well the mice function on -- was the road around, right?

Yes.

Yes, yes. Going back from mouse to human, that it got -- it was sustained and their weight was sustained. So we believe, and as Roeland explained, the biology, if you activate the neuromuscular junction in these patients that the activity of the antibody may outlast the actual ability of us to develop -- to measure the antibody in the blood. So we're very interested to see how long these patients maintain the level of function that we've seen them improve to. And then that will help us as we build on the plan for the next study.

Great. So maybe, Luc, now a question for you. So now as we think ahead to Phase III, like is there anything in particular that you would take from our learnings from Phase I and the relationships that we've established with Dr. Burden and Dr. Maselli and others as we plan for Phase III?

Yes. So the obvious one is science always leads, and we have been able to demonstrate that in a very particular model and that connection with the animal model is to me, that is a beautiful strong connection that we can also bring forward to the regulators in terms of plausibility that together with this is ultra rare, we shall forge a part a path where we're going to continue to have the focus on how do we demonstrate this in the most efficient way, the benefit so that a benefit risk assessment can be made by a regulator. Now we need to negotiate that. And there have been recent pathways alluded to that might be an opportunity for this particular setting like the plausible mechanism and all those things, which here is evidently 100% present, and it is a severe disease. So we're going to have this push to work in the argenx way, data-driven, science-based and learn from this that high-density measurements, digital measurements contribute to our understanding of the benefit.

Thank you. So I wish we had more time to have more questions, but I think we'll move on to the next segment. I think Roeland will be coming back up to speak to us. So thank you, Dr. Maselli and Rebecca and Luc for joining us.

Well, that was interesting. So I guess you're all wondering now what is the opportunity for ARGX-119. You know that we're developing an ALS and SMA. So let's first recap that a little bit, and then I'll tell you more. So ALS is this progressive motor neuron disease. It is characterized by the loss of motor learning from the muscle. And as you remember, ARGX-119 can slow down this denervation pathway. So there's a high unmet need and activation of MuSK by, for example, ARGX-119, slows down muscle denervation and improved motor function. As Steve alluded to, he has shown in ALS mouse models that MuSK activation slows down the denervation and improves motor function. We took a slightly different approach. We took ALS patient-derived motor neurons. We culture these in a dish and let them innervate onto human muscle cells. Now in this kind of setting, we can use an electrical stimulus at the motor neuron and look at the muscle contractions on the other side. And in the ALS setting, the muscle contractions are reduced by 50%, showing the muscle weakness observed in patients. If you now give a therapeutic treatment with ARGX-119, these muscle contractions are fully restored, suggesting that the neuromuscular junction in that dish is fully functional again. So I'm very excited that in the first half of 2026, we'll have this first top line readout in ALS patients. And we do this with the trials experts in the trial. So how did this trial look like again? Our Phase IIa trial is a dose-finding trial, but more importantly, it's a proof-of-concept study. We include all the typical ALS measures in the field, but we will look at how ARGX-119 functions in these patients. We will capture its mode of action by using a novel innovative endpoint that is called M-Scan. I'll explain that to you. We will use M-Scan as a precision tool to quantify motor unit loss in ALS to really capture if ARGX-119 can slow down denervation. So M-Scan is a surface myography approach. It's a neurophysiology approach, where as you can see there in the M-Scan setup, you use stimulating electrodes on your wrist. Those stimulating electrodes will excite the full nerve bundle that is innervating in your muscle here and your tongue. We then record all the excitability from that muscle. What is then -- what can then be measured there? So you see there a recording from an ALS patient at baseline, you see this S-shaped curve and every dot there is a stimulus. So every dot represents an innervated neuromuscular junction. From that kind of graph, we can measure several parameters, for example, the MUNE or motor unit estimates. That patient has, if I look right, 30 neuromuscular junctions innervated in its tongue muscle. You can also measure the maximum CMAP and CMAP is the compound muscle action potential, so the signal across the neuromuscular junction. If you -- if this patient waits now 7 -- after 7.5 months, you see this curve drop down, the signal across the neuromuscular junction is lower. But also you see the MUNE here is only 8. That patient lost a significant amount of motor neurons over 7.5 months. You also see this step-wise staggering in the graph. That means that the remaining motor neurons have expanded because they need to compensate for all the loss of the other motor neurons. So MUNE or innervated motor units can be tracked over time. You see in the left graph below the progression rate over time. So we will have a 24-week study. There's a significant decline after six months. We hope to see with the treatment of ARGX-119 that the slope has a positive deflection. More interestingly, this biomarker can be used as a kind of prediction for clinical outcome as there is a correlation with the clinical outcome scores in ALS, for example, the ALSFRS-R score. So a positive effect on M-Scan can result in a beneficial effect in the clinic. SMA, you all know spinal muscular atrophy. It's affected in children. It's very severe. It's characterized by neuromuscular weakness and motor impairment. You probably think there was this massive breakthrough with SMN upregulating therapies and the disease is cured. It's not. It was a great breakthrough. These patients don't die anymore. But in partnership with Cure SMA, they reported that patients are still feeling weak. They want to improve in muscle strength in motor function and they want to reduce their muscle fatigability. If you look at the science, that all comes back to a defective neuromuscular junction. There are several publications show that even on these SMN upregulating treatments, they have reduced neuromuscular junction maturity and transmission defects remain. So we hypothesized that activating argenx -- activating MuSK with ARGX-119 can mature these neuromuscular junctions and improve the transmission. So together with the SMA Foundation, we've conducted mouse experiments. We've used the SMA delta 7 model, the very severe model. We gave them SMN upregulating therapies and looked how well they can use their muscles. We've measured the muscle force in a specific muscle. And as you can see on the graph there, the black line is the muscle force in these mice that were treated with SMN upregulating treatments, it's still very weak. With ARGX-119 treatment on top of that, you fully rescue the force in these mice. Interestingly, we've seen that the increase in muscle force preceded the increases in muscle weight, suggesting that muscle function might be more important than muscle size. So we're going to initiate a Phase II study pretty soon. So ARGX-119 and what other diseases can it work? We think it can really be a pipeline and a product. As you know, we've discussed these two modes of action. In the blue there, it can really increase the clustering of acetylcholine receptors and increase transmission across the synapse. That's applicable in neuromuscular junction diseases like we've shown in CMS. You could think beyond that. That's also applicable in muscle diseases, muscle dystrophies, muscle myopathies. The second mode of action, this retrograde signal, keeping the neuromuscular junction together, restructuring that is very applicable in motor neuron diseases like ALS and SMA and others, but also in peripheral neuropathies. So we're doing all that work now and more to come soon. So currently, this is the path forward for ARGX-119. I'm very proud that we've shown the proof of biology in our Phase Ib trial, and we're going to get ARGX-119 to patients as fast as possible with a straight to Phase III approach. In ALS, we have this exciting data readout in the first half of 2026, and the M-Scan data set something -- will be something to look forward to. In SMA, we already have a name for our study. It's called SPARKLE, and it's going to initiate soon, and that will actually broaden the age range from adults going to pediatrics. So with that, I hope we've given you a lot to think about. And therefore, we're going to have Tim on the stage for some closing remarks.

First of all, welcome to the company. Here, you're in the belly of the beast. So whenever we meet you in your offices, you want to hear about IIP and we get lots of interest in IIP. I think today, you saw IIP alive. I could not think of a better example to show you today live with our trusted collaborators, how really IIP works. I'm very proud today that you gave us the time to talk about science because it always starts with the science. I don't know how you feel about the patient which we saw, but I'm deeply impressed by the patient, so weak and still so strong. And then IIP being the formula of connecting deep science with that fundamental patient needs and trying to create the innovation, which, frankly speaking, is our raison d'être, the reason of existence. The only reason we exist as a company is because this is the type of work which we're doing with your full support over these past years. What I would like to do now is get into Q&A. We have the huge advantage of having Steve here, having Ricardo here. Thanks again, guys, for joining us today. But this is your opportunity to get into the questions you may still have after this session. We will reserve the questions only for our investors and analysts. Employees can hold the questions for the reception, okay? So why don't we start. Yaron, thank you for kicking off.

Good afternoon. Thank you so much for hosting this event. I'm Danielle Brill from Truist Securities. I guess my question is pertaining to the CMS trial. From my understanding, CMS symptoms fluctuate. They can wax and wane at different times. Can you contextualize for us what impact this might have on metrics like 6-minute walk distance and what risk that may pose to a Phase III?

Thank you for the question. But unlike autoimmune, MG, you don't really have the waxing and waning. I think what we tried to say with the natural history study is that when we started to study these patients over certain periods, they were pretty stable. And I would like to give the floor to Luc to maybe give a bit more color to what that natural history study has actually shown in terms of the stability of these patients because it is a key question you're actually touching on. Luc?

But you already gave the answer. No, that was a remarkable part that, by and large, they stay quite stable, which is why one of the other reasons we thought this could be useful is to show that inflection point, right? So you have a certain so before and then after treatment. But they were within the confines of that natural history what we measure, quite stable. Was there variability between patients? Yes, of course.

Maybe a couple of questions. Yaron Werber from TD Cowen. Maybe the first one, I don't know, Luc, if you want to take it or somebody else. In the Phase Ib, did you look at 6-minute walk test? And can you share with us kind of what that data showed? And as you think about sort of powering the Phase II based on the natural history, what's -- it sounds that patients are stable in 6-minute walk, so you're looking for a net gain from baseline. You're not looking for -- you're not expecting deterioration in placebo. And then maybe for one of the KOLs, what percentage of patients have postsynaptic CMS of the overall syndrome that you're targeting? And how many patients are there in the U.S.

That's a four-part question, I think. So the 6-minute walk test you saw, it's definitely not stable, and it's actually our biggest signal we have of the effect with the effect being of a clinically relevant size. So within the natural history study, we wish we had had 6-minute walk test, so we would have that comparison, but we didn't do that because -- well, we just didn't do it, okay? And in -- and I'm being told in the amendment, we are adding it. So -- that was actually the most encouraging part. And for me, but this, Rebecca, you actually beautifully show this, the connection with cadence also going up, right? So these patients upped their speed of walking also, not further, but within the distance also faster. So we found that as a pretty compelling that something is happening in one of the most affected limbs.

Yes. And thank you, Luc. So a lot to unpack and pioneer in CMS, big learnings is this is a limb-girdle phenotype, weakness in arms, legs, and head that translates into that walking test that translates into that cadence of step. And I think we're very well equipped with this data set to now actually go and meet with the regulators and discuss what suitable endpoints could be in the registrational trial. Your other question is also very relevant. Maybe Steve, do you want to address the question or Ricardo, what percentage of all these congenital myasthenic syndrome patients would actually qualify for the MuSK agonist therapy.

Well, if I understood the question correctly, the question was congenital myasthenic syndromes, what is the proportion of postsynaptic patients. And that is actually is a majority of patients with CMS, they have postsynaptic defects because defects of the genes that encode the subunits of the receptors and rapsyn and Steve alluded to rapsyn, those two molecules -- well, five molecules, four subunits of the acetylcholine receptor and rapsyn, those are the most common. Actually, it's more than 50%, I will say, 60% of the patients resolve from that. Now if the question refers to do DOK7, specifically DOK7 as agrin MuSK and LRP4 has a two component, the pre- and the postsynaptic component. We know a lot about a postsynaptic component because the technology that we have is suited to study that, but there is a lot of questions about the presynaptic component and I was so glad that Steve was mentioning that they are really focusing on that because this is really very, very important. And I can tell you, the more severely affected patients with CMS doing the microelectrode recordings are the one that has a very severe presynaptic component. So in answer to your question, yes, they have the two components, the postsynaptic and presynaptic component, postsynaptic is by far the most common.

Steve, anything you want to add to that?

Mostly just a similar view. I think about -- my guess is about 25%, 30% of patients have mutations in DOK7 and a similar number in rapsyn. And so as you say, adds up to more than 50% of the patients. We've started to look at mouse models of mutations in rapsyn and COLQ, which is the attachment protein for the cholinesterase that Ricardo mentioned. And it looks like from our preliminary phase that a mouse model of congenital myasthenia caused by mutations in agrin might be rescued by the agonist antibody as well. And again, anything that's upstream from us, it acts to activate MuSK you might expect would be treatable by the agonist antibody. So the first evidence of the mouse model in agrin is that, that works. We've looked at a no mutation in COLQ, so there's no COLQ at all. And that looked like it wasn't affected. And that was consistent with the way we were thinking. And again, with rapsyn, which is downstream, again we've looked at a no mutation, so there's no rapsyn and that looks like it wasn't rescued by the agonist. So that's kind of what -- that fits with the biology.

Which is logical if you understand the biology, right?

Which would suggest still leaves open the possibility of the patients with mutations in LRP4 and mutations in MuSK could be, so it had the potential of being treatable with the agonist antibody.

But ballpark, roughly based on what we know today, 50%. Yes.

Leland Gershell from Oppenheimer. I want to ask just in terms of dose, do you have clear visibility on dose entering the Phase III in CMS. If you could remind us in the Phase Ib was that ascending dose or different doses? I ask because at some point, you start to see an inverse dose response from -- given the dimerization mechanism.

Thank you for the question. And maybe, Luc you want to comment on the key question on dose finding and how we went about it.

Yes. Again, so we're in a situation where the classical paradigm of placebo, low, mid, high dose really wasn't feasible in a reasonable amount of time in our sense of urgency that made us say, we can get to this information with intrapatient dosing and some very clever modeling. And so we have a PK/PD model that gets at the exposure response relationship from which we will select a dose for the Phase III. Again, in conversation with the agency, they will have something to say about this as well.

And it goes back, Luc, if I can say that to have an innovation mission and the way we design innovative clinical trials, based on that intra-patient dose escalation with that exposure model, we can really triangulate what exposure we think we need in order to fully activate MuSK and have the maximum benefit from a functional point of view.

Okay. And then just secondly, I know you've been focused on DOK7. Is there any opportunity for use outside of DOK7 CMS based on biology or would that require different strategy?

So I think that's what Steve said. So basically, in that pathway, anything which is leading up to MuSK activation and which has a defect could be overcome by forcing the activation of the antibody. Thanks for the question.

Tazeen Ahmad from Bank of America. I wanted to get a sense about how you're thinking about additional indications. So you obviously are moving forward with CMS. You've talked about moving potentially in many directions, but you've highlighted SMA and ALS specifically today. What is the sense of confidence sort of that path that's leading you to highlight those two because there are other therapies available right now for those. And historically, argenx tries to go where there's a big opportunity, but not necessarily too crowded. So would love to hear your thoughts on that. Second, have you had any initial conversations with FDA regarding the path forward. You talked about all the innovative ways you've tried to collect data. but I think it's standard protocol now to ask how conversations are going with agencies and level of confidence in what they're telling you. And then lastly, when could study for SMA that you're starting yield data?

Three great questions. I will try to quickly address them because we're short in time. So unmet medical need and ALS, we don't need to elaborate. I mean that's obvious, that's a pretty high-risk indication to venture into and that's why we really wanted to show you that precision instrument we're using to venture into ALS and really get a clear cut answer, so whether we move the needle on the biology, yes or no, based on this, you would advance in your clinical development. So I think that's a reasonable derisking. And I think that's a responsible use of capital to do it, strong belief in biology, clear way of measuring effect, but we need to do the experiment. If you move the needle, even if you're just slowing down disease, I think for ALS patients, that would be big. We have never seen a clinical trial, which enrolled so fast as the ALS trial. [indiscernible] maybe did a lot of work with the specialists and with patients because indeed, from the outside, it looks like in SMA, the unmet medical need has been mainly satisfied. That doesn't happen to be true. There's a ton of function which still needs to be regained from a muscle point of view. So we're really looking at the mechanism of action, which can be additive or maybe even synergistic to the therapies out there. So we don't seek to compete with them. We seek to complement them and further regain muscle function. And then to your last question, we're leading up to an interaction with the regulators. I think we're very well equipped with data. But let us talk about an interaction after we've had it. And I get a signal here that Steve, you would like to add to the comments.

Yes. Just to clarify, for ALS, the idea is that in ALS, as many people don't appreciate the earliest first step in the disease before motor neurons die is that nerve terminals withdraw, and it detach from the muscle. So it's that phase, the early phase of the disease, which is sufficient to cause the paralysis that we're trying to address with the MuSK agonist antibody. We're not trying to save motor neurons. We're trying to keep the nerve attached to the muscles so it can function. And in mouse models, if you play genetic tricks and keep the neurons alive so they do not die, but the synapse still detach, the phenotype is the same. So it's a very important feature of the disease.

Thank you, Steve. Maybe we go to the final question of the session.

Yes. Patrick Culliton, Stifel. I know you've shown at least preclinically that 119 doesn't interfere with the agrin pathway. But is there any evidence that chronic MuSK activation could result in some sort of down regulation of the agrin-LRP4 pathway? And then -- just considering that? And maybe just more broadly, how are you guys thinking about dosing and maintenance treatment with 119.

Two great questions. Roeland, do you want to comment on the longer-term DOK7 studies, which we have been doing in the run-up to the Phase Ib study.

Yes. So we've done sign-off studies and a 26-week tox study at very high doses, very frequent dosing. And there, we have not picked up anything suggesting that the molecule even with chronic dosing is safe.

Yes. And then on the second question, the durability of the effect is, of course, speculation, but it's being studied in the OLE based on the suggestion of the mass data, right?

Yes, correct. So the patients after 12 weeks stop dosing, then the PK will go down, as you know, it has a half-life of about 30 days. That's why it's a 7.5 month follow-up so that the PK is fully down. And if we then see still function remaining that will give us a massive insight into further development of the dosing strategy, yes.

Maybe the final question.

Delma Caiati from Guggenheim. Thank you for this great event, really insightful. So Dr. Maselli was making a very interesting point about some subtypes not responding to acetylcholinesterase inhibitors. How well is that understood? Like is it associated to specific genetic subtype first and what is the mechanism underlying that? Is it associated to the down regulation of the muscarinic receptor -- to the acetylcholinesterase receptors themselves? And is there any risk that activating MuSK can over long term lead to a similar risk, similar effect? That's the first question. And then if I may, on SMA. For the study, are you targeting both type 2 and type 3 patients?

Two great questions. I'm so happy we have the experts here to answer question #1.

Yes, I will take only the first one because I don't -- so yes, there is a condition that is actually the first congenital myasthenic syndrome described completely 50 years ago by Dr. Engel was deficiency of cholinesterase, in other words, you don't have the cholinesterase, as it turned out it is not due to a defect of the gene of the cholinesterase but the gene that holds the cholinesterase at the end plate. And that gene is called COLQ, okay? So the deficiency of COLQ is a very well-established congenital myasthenic syndrome that evidently doesn't respond to inhibitor of the acetylcholinesterase because they don't have acetylcholinesterase to start with. So this is a condition that is being treated, and there is a well-established linkage to a gene defect, but the other condition that Steve mentioned, LRP4, agrin, MuSK and DOK7, those do not respond to -- the only medication that we have available is salbutamol. Salbutamol is a medication -- it's not an approved medication, but it's a medication that is effective. We actually don't know exactly the mechanism how those patients respond, but this is the only medication that is available. So I don't have any comments about the other component.

Thank you so much, Ricardo. We need to be respectful all of time. We're going to close the formal session here today. The experts are here in this section. We have drinks, so what I would suggest we do next is we close the formal session. We go into the mix and mingle, and you really use the opportunity to talk to the experts and continue with Q&A, okay? Thank you so much.