Coya Therapeutics, Inc. (COYA) October 20, 2026 Conference Transcript

Okay. We're ready to start this next session. In this one, we want to kind of continue our discussion on neurodegenerative diseases. In this session, we're going to focus on ALS. My name is Kane Akai. I'm the Director of Research and one of the Senior Biotech Analysts at Shardin. Joining me for this discussion is Dr. Fred Grossman, the president and CMO of COIA, and Dr. Asa Abelovic, the CEO of LEO. So, gentlemen, thank you so much for joining us. ALS, you know, the first, well, we've, you know, it's kind of been slow progress, a lot of failures here as well. Derivome was approved, many would say, kind of limited, maybe best for early-stage patients, less effective for the broader ALS patient population. In 2023, we had the approval of Tefercin, I think notable in terms of a different approach, but obviously a much smaller class of patients with the SOD1 mutation, but also notably an accelerated approval based on a biomarker, NFL. So I think that was a nice milestone for the field. But again, I think we need better solutions for a broader population of patients with ALS. So let's dive into that with the two companies here. Maybe Asa, let me start with you. You know, in your case, you have an approach where you're looking to address the generation of the sphingolite livits and the ceramides, and so how do we keep that from becoming a problem and propagating progression of the disease? So maybe talk about your platform and how you're doing that, especially with your ASO.

Yeah, no, thanks so much. No, we're really, really excited about where we're at now in the clinic, in patients with ALS. So Leal Therapeutics is focused on basically the intersection between metabolic disease and CNS disease. And we basically go after really well-characterized, deeply validated metabolic enzymes and pathways with brain penetrant drugs. And so case in point here, LTX002, targets an enzyme that's involved in the synthesis of ceramide sphingolipids, and it's really based on human genetics. So there's people with disinhibiting gain-of-function mutations in this specific enzyme that's the first and rate-limited enzyme to generate ceramide sphinglob is this class of lipids that's known to be potentially toxic. And that is enough to cause ALS. It causes early-onset severe ALS or ALS and FTD in people. But what's really exciting about the story is that when we went on to look, it's a fairly robust literature that ceramide sphingolipids are generally increased in sporadic disease as well. And people have shown, and we now have shown as well, that inhibiting this pathway is protective. It's anti-inflammatory. It also suppresses a lot of the kinds of pathology that you see in ALS. So I think what's been missing to an extent in ALS is a treatment that's genetically validated where there's robust biomarkers with a deep understanding as well, but that can help sporadic disease. To your point, the one thing that we now know is that ASOs can work in ALS. There's TOFERSEN, but as you know, also in the last couple weeks, really exciting data for the FUSS program. Again, NFL validated. And then we're just missing something for the other 98%, 97%, 98% that don't have those mutations. So here, what's exciting is we can go after both the rare genetic mutations. Again, there's rare individuals that have exactly this mutation that causes disinhibition genetically in excessive ceramides, but again, a lot of data that this is relevant in sporadic ALS as well. And then, finally, we're now in the clinic and progressing well.

So maybe highlight some of the key observations from your preclinical data for 002.

Yeah. So we built particularly strong capability in Worcester, Mass, where with just expertise in developing intrathecal nucleic acid therapies like ASOs. and but also capability and BBB shuttle transport so that we could use ultimately use peripheral delivery so that's a transferrin based approach and we can talk about that as well but with it so ultimately we generated LTX 002 we could show readily then in NHP that this reduced the level of this enzyme it could suppress the production of ceramides phenylopids and it could do this safely and effectively you can't a hundred percent turn this pathway off but we see a very broad safety margin basically so and ultimately we even have long-term data that supports that supports taking this to open label extension and and and we actually are in that process now so yeah so so So basically, ultimately, after extensive other work, in NHP, we could show that we could effectively and safely reduce the enzyme, reduce the ceramide sphingolipids, and that allowed us to move forward. Separate from that, we do also have an oral small molecule that targets the same enzyme. It's the first-class brain penetrant oral small molecule. There have been tool compounds for over a decade. They are effective in models of ALS. Other people have shown that models that do or that have this mutation, but also models that have other mutation, they suppress TDP43 pathology. This is upstream of TDP43 pathology because inhibiting it suppresses TDP43 pathology, whereas knock-in mice, and this is work at the NIH, not our preclinical work, but at the NIH, people have shown that knock-in mice that have this mutation get typical TDP43 pathology they get increased NFL they get as they get motor findings that can be suppressed with with tool compounds what what's been missing is is drugs that can actually help patients and get in the so we do also have oral small molecules and and we've shown that they are effective again across other models

of ALS so in your first in human trial here I know you're currently enrolling patients in Europe. Maybe just give us a sense for what you're assessing and what, when you first report interim data, what will you report on?

Yeah, so, yeah, huge upside to looking at a well-caratized metabolic enzyme. We can directly measure the product, so we can look at ceramide sphinglypids and show we engage target approach. We can also actually measure the enzyme, SPTLC1, on the subunit in CSF, we can measure that. So we can look at PKPD very directly. But most exciting is NFL. Again, that's what's really enabled the Tofferson program, the FUS program. And we have a lot of data, and again, published data also, other people's data, that this pathway directly is upstream of NFL, upstream of TDP43 pathology. So, you know, and then, of course, we'll get ALS, FRS, the ALS rating scale data, as well as other motor and clinical endpoints.

Okay. Well, great. Well, I'll come back to you, Asa, with some other programs that you have under development and your shuttle delivery. But, Fred, let's shift to what you're doing at Koya, basically addressing the inflammation and the problem that that causes in the propagation of the disease using recombinant low-dose IL-2. So tell us how that works and impacts Tregs, and then your initial compound, 3O2, is a combination of that and CTLA-4. So start with the low-dose IL-2 and then the combination and the synergies that that's affecting.

Sure. ALS is a disease of inflammation. We've heard that before, and we know that the inflammatory activity is quite high in ALS. And the cornerstone of this inflammation, we believe, are regulatory T cells. It's been shown time and time again that regulatory T cells are dysfunctional in ALS as well as other neurodegenerative diseases. And the reason for that is that, you know, obviously in a healthy body, when there is a foreign invader, the innate inflammatory system increases and you need something to hold it back. Otherwise, it'll continue to destroy itself. And that's where regulatory T cells come in. Now, the problem is it's very hard to harness something that can affect regulatory T-cells as a single agent. We know that low-dose IL-2 can raise regulatory T-cell number and suppressive function, which is really the goal here. And at Coria, we believe in combination therapies because the immune system is very complex and redundant. We don't think a single agent is going to be effective. So we're really focusing on combination therapy. So let's go back again to regulatory T-cells and ALS. In ALS, regulatory T-cells are dysfunctional. They're lower in number and lower in suppressive functions. Low-dose IL-2, and by low-dose, extremely low-dose, can increase regulatory T-cells as well as increased suppressive function, and that's great. However, in the inflammatory environment, when you have a highly inflamed environment, regulatory T cells can become dysfunctional again. So what we've found is that combining low-dose IL-2 with CTLA-4 that tamps down the effector T cells, You can have prolonged elevation of regulatory T cell in number and suppressive function. And that's the basis of COIA 302, which is combination therapy in ALS. So the goal that we have is not to reverse or cure ALS. It's to take individuals who have just been diagnosed, who have some deficits, but a good quality of life, and hopefully initiate this therapy and stop the continued reduction in ALSFRS or the continued progression so that they can continue to have a good quality of life. And by the way, that is associated with survival. So that's the initial goal that we have. And we can talk about data if you want.

Yeah, so, you know, obviously you're in a pivotal study right now. But before you got there, maybe highlight some of the key observations from your preclinical data. But maybe also, importantly, you also have some actual human data from disease patients from an IIT conducted by Dr. Appel. So maybe start with the key findings in the preclin and then the validation of that in Dr. Appel's patient.

Sure. Well, most of our data are, in fact, in humans and come from human cells and human studies. Much of that work, as mentioned, is by Dr. Appel at Houston Methodist. But also our scientific board has Shimon Segaguchi, who won the Nobel Prize in Medicine last year for establishing and understanding regulatory T-cells. So that's our focus as a company. No matter what disease we've looked at, whether it's through IITs at Houston Methodist, whether it's Alzheimer's, FTD, and others, we find that this combination mechanistically does exactly what we want it to do, meaning that it increases regulatory T-cell number and function in a sustained way. And in small numbers of patients, in fact, there is, in the case of FTD, there were nine patients. And in fact, there was a stability in the MOCA and cognitive tests. However, we did find, importantly, increase in regulatory T-cell number and function. So let's get back to ALS. This combination of low-dose IL-2 and CTLA-4 were given to four patients with ALS. These patients came in with an ALSFRS score that averaged around 33-34, which is typically what you would see when someone is first diagnosed with ALS. And over a six-month period, remarkably, there was little change, little to no change in the ALSFRS. Now, we know that in ALS there's typically a reduction as the disease progresses of one point per month. So after six months, you would expect a reduction of six points on the ALSFRS. And by the way, six points is significant. You know, that may be the difference between a patient who's walking and someone who winds up in a wheelchair. Well, we found flat lines in the ALSFRS over six months. There were really no changes in that six-month period. Now, that's eye-opening, but it's only four patients. Continued those patients for another six months, 12 months. and at the end found very little difference in the ALSFRS. So we have a clinical observation here where patients who should have had a reduction of even 12 points had barely any change over a year period, but it's only four patients. So importantly, we looked at the key variables mechanistically, And in fact, within a few weeks of therapy, regulatory T-cell number went up, regulatory T-cell suppression went up, and it maintained through the course of the illness. So in those four patients, we see a clinical finding that was, I mean, quite frankly, eye-opening, but only four patients, and biologic confirmation mechanistically of what we wanted to see. There were no safety concerns because the dose of IL-2 is extremely low, and even CTLA-4 is half the dose that's typically given. And importantly, the formulation is sub-Q. It means that if this will eventually get approved based on data, this can be self-administered by patients and their caregivers. So that was the basis of moving forward.

Okay, so you're currently enrolling in the All-Stars Phase 2. So maybe briefly highlight the design, but maybe more importantly, in doing your design, how did you incorporate your learning from your own studies, but also the failures of other studies to increase or handicap your chances of success there?

So this is a controlled study that consists of three groups. One group is the combination at the dose sequence that was given to those four patients. The second dosing group is half that dose, given once a month versus twice per month, and placebo. So it's placebo-controlled. It's powered well over about 80%. 120 patients which have been enrolled already. However, there are patients in our screening period, and we're allowing them to move through. So it'll be higher than 120 patients. The primary endpoint is a six-month endpoint, and it's the difference in the ALSFRS. Of course, we have secondary NFL endpoint as well. And then we have a very interesting second half to the study. So the six-month period will be the primary endpoint, and quite frankly, if it's very positive, then we will be in discussions with the FDA to submit the file. The second six months are blinded, and so we have three groups, two treatment groups, one placebo. The treatment groups will continue blinded on their sequence therapy, but the placebo group will be re-randomized to one of those other two groups. And why is that interesting? Well, we just said before, a six-point reduction in the ALSFRS is significant. And you would expect that placebo group, that most of the patients will have had that reduction. So now we'll be able to test, when they go on either of the two doses, whether it can stabilize even the more severe patients with ALS. And we were going to complete the study after 12 months, but there was really a lot of pushback. Patients wanted to continue, and we don't, it's blinded at this point, we haven't locked the database yet. Patients wanted to continue and PIs were requesting it. So we did add an open-label extension at the end of that study, so if patients choose, they can continue on that. The other innovative thing we did was to stratify by NFL, and we'll talk about that in a moment. NFL, surely, is an established biomarker, but it's not a great biomarker, quite frankly. It's just a marker of axonal injury, and so it doesn't track with the ALS very well. And we could talk about data that we've done in 100 patients where it showed that it did not track. But we used NFL to stratify across the three groups. Why? Because in prior studies, they did not do that. And in fact, it's possible that some groups may have had a very high NFL, which would mean that they probably have had so much damage already that no intervention would be helpful. So stratifying was something that we did. We also had the entry criteria to ensure that we get the majority of patients who are reducing their ALSFRS by history by just 0.5 to 1.5 because the fast progressors are likely not going to have enough time to respond to your treatment. We have an entry level of 35 on the ALSFRS, which, again, is getting at those patients at that sweet spot when they're first diagnosed. Because, again, we're not looking to reverse ALS. We're looking to stabilize it so patients and people, you know, can live a good quality of life. So those are some of the things that we've done based on prior studies.

Okay. Well, you are on track to report the data, top-line data in Q2, so we'll look forward to that. also let me come back to you because the the idea of the problems that excess ceramide sphingoide lipids it's not just ALS it can apply to other neuro neurodegenerative disorders but also an enabling technology that maybe is going to allow you to proceed to addressing those others as your delivery technology and your conjugate to you know as a shuttle as you call it to get more to across the blood-brain barrier. Can you describe what you're doing on that?

Yeah, no, I think this is a tremendously exciting time in CNS right now in that shuttle technology, which has been around to some extent for a couple decades, is really maturing. And we've worked on this for the past four or five years, really trying to optimize different aspects. We've looked at different kinds of shuttles. These are antibody-like shuttles. And we, as some others, have ended up focusing on transferrin receptor-binding antibodies like shuttle. We've iteratively really optimized these in terms of their affinities. We've had to actually have bespoke different shuttles to deliver SINAs versus delivering antibodies. And what's really exciting is that, you know, we have two that are now in NHP and really getting lined up. One, a MAP-TAU program where we selected a candidate actually delivering an sRNA as well as an APOE program. And particularly with the MAP-TAU, we, you know, this is with subcutaneous injection of this conjugate. subcutaneous injection, which ultimately could be every few weeks, we believe, maybe up to even three months. Remarkably, we get better knockdown and quite deep knockdown in the CNS across the brain than we get in the liver peripherally. And again, that's taken a lot of optimization in terms of the nucleic acid, including the linker, and in terms of the affinities involved. And even Even looking head-to-head at some of the other shuttles, we do have a strong case that this could be best in class as a shuttle to deliver nucleic acids. So really excited. We've seen very clean tar-ability and moving that forward now. We'll have NHP data this quarter, actually, and then very soon we'll start to push that into IND-enabling studies to get to the clinic. We do also have a shuttle, again, that can deliver antibodies, like A-beta antibodies, and that we now have—we just completed an in vivo study that's reading out to select a definitive candidate. And there also, we're particularly excited because we believe we have optimized to avoid some of the off-target issues, as you know, with respect to delivering antibodies, one of the very active issues is how do you avoid toxicity, because you typically do want to have an active FC, you want to have an antibody that will promote phagocytosis, but if that's excessive, if the FC activity is excessive, or if it's polarized incorrectly, you can get toxicity, particularly reticulocyte drops, because these transferrin antibodies will also end up binding to reticulocyte. So just optimizing this has been really rewarding, and we really do think we have something that's potentially a bit more optimized than what we see out there. Again, and we'll have the in vivo candidate selection completed this quarter and also pushing that forward.

Okay, and Fred, maybe I'll squeeze in a final question about your pipeline. Obviously, 302, a lot of versatility. Give us a status update on 302 for FTD.

Yeah, so we have FDA approval to move forward in FTD because like ALS, FTD is a highly inflammatory disorder. and we have some preliminary data I mentioned before where this combination can be useful. We will be initiating the FTD trial in 2027, and, of course, you can imagine if some of this data continues to come in positive, we'll be moving into other neurodegenerative diseases with this combination.

Okay, ma'am, you had a question? Yeah.

Yeah, so in the non-human primate, And so particularly, for instance, with the tau or APOE, we have a really pretty deep understanding at this point in the field that tau plays a key role. We know how much we want to reduce it. We know that even we can reduce it in active disease from a lot of models. So really what we're doing in the NHP is showing, first of all, safety. That's critical. and second of all, deep target engagement and getting a clear sense of the PKPD, just to be clear. So in NHP, we're not modeling the disease per se. This is really validating. But there's a strong argument with targets like MAP-TAU and APOE that there's so much human genetic, human biomarker, and preclinical data, and such a deep understanding that the more critical aspect is really showing target engagement and showing that you can do it in a tolerable way.

All right. Well, great. Go ahead, ma'am. Yeah.

Yeah, yeah, yeah. No, it's very efficient, to your point, and absolutely. I mean, I think the same way the liver targeting, you know, was hugely beneficial safety-wise, no question. The targeting absolutely looks well-tolerated. I think where people have run into trouble is with delivering with fully active FC. there and there are a couple different ways to to work that but um other than that the doses are with the targeting the doses are lower and and they're much better it's absolutely you don't run into some of the issues that you see with the untargeted?

Sure.

That question is to me. Yeah, so I can just answer it very easily. Remember, this mechanism is agnostic to whether it's familial or sporadic. It doesn't matter because we're reducing inflammation. It also doesn't matter what treatment they're on. So So effectively, if they have high inflammation, which they will, and they meet our criteria, this combination can be effective. We're also looking at oxidative stress markers. We don't have time to talk about that, but that's something we're looking at also that may be a little more sensitive than NFL.

All right. Well, great. Thank you so much for joining us for this session.

Thank you.