Good morning, everyone. Welcome to our Piper Sandler Healthcare Conference. My name is Yasmeen Rahimi. I'm a biotech analyst here at Piper Sandler. Excited to have Shattuck Labs, here with us. Excited to hear the presentation as 2026 is going to be a big year. Thank you again for being here.
Great. Good morning, everybody. My name is Taylor Schreiber. I'm happy to present on behalf of Shattuck, and thank you to Yaz and the team at Piper Sandler for including us this year. These are my forward-looking statements. This is a view of our current pipeline. In today's presentation, we'll focus simply on the lead antibody, which we call SL-325. This is a DR3 blocking antibody. There's also a half-life extended version of this antibody called SL-425 we won't touch on. There is a variety of bispecific antibodies also leveraging DR3 targeting with one arm that we also won't have time to get to today. What I'll start with is a brief overview of the biology of the TL1A/ DR3 axis. This is likely a familiar axis to many folks in the audience, but you've probably heard mostly about TL1A.
TL1A, is the ligand of this axis, and it binds a single receptor, called DR3. You have learned about this because of the increasingly positive data that is accruing from multiple companies, with TL1A blocking antibodies in ulcerative colitis and Crohn's disease, where we are already seeing potentially best in disease efficacy with this axis, with also potentially best in disease safety. Some KOLs, have described the emerging data with the TL1A blocking antibodies as having JAK-like efficacy, without the toxicities that come with JAK inhibitors. That is wonderful. The two aspects of the TL1A/ DR3 axis that I would like to focus on and which really underpin our belief that thinking about blocking the receptor might represent an evolution is the known immunogenicity liability with TL1A blocking antibodies and the potential for more durable inhibition of the axis by going after the receptor.
On the left, this figure both explains where our antibody binds and gives you a very simplistic but accurate overview of the axis. As I mentioned, TL1A is the ligand. TL1A, is primarily expressed by antigen-presenting cells, located in tissues, and I'll go through the kinetics of that in a moment, whereas DR3 is the receptor, and it's expressed primarily by lymphocytes found both in the blood and within tissues. TL1A and DR3 only bind one another from a signaling perspective. There's no reason to believe that you will have greater specificity by blocking TL1A instead of DR3. Because of that, we believe that the safety profile that is emerging with the TL1A blocking antibodies is likely to inform the safety profile of a DR3 blocking antibody, particularly if that antibody is engineered to lack the ability to bind Fc gamma receptors, which ours is.
SL-325 is a high affinity, highly specific human DR3 blocking antibody. It binds with close to a 1-picomolar binding affinity, and it binds an epitope which was selected, number one, because it prevents trimerization of DR3. Number two, it prevents trimer-to-trimer binding between TL1A and DR3, and it is an epitope that is not shared with the decoy receptor, DCR3. The natural role of DCR3 is to facilitate clearance of TL1A, so we leave that alone. The figure on the right here begins to speak to the efficacy hypothesis. What is shown here is data from a publication where these authors were taking biopsies from patients with Crohn's disease and biopsying both the actively inflamed part of a Crohn's disease patient's gut and then the adjacent uninflamed tissue.
Simply asking, in those two locations, what proportion of cells stain positive for TL1A and what proportion of cells stain positive for DR3? What you can see is that in the actively inflamed biopsy, this is a panel of flow figures on the top, about 8%, of cells stain positive for TL1A and about twice that many cells stain positive for DR3. If you go to the margin of inflammation, you find that there is no upregulation of TL1A. You have even abundance of DR3, however, both within the actively inflamed area and at the margin of inflammation. Why is this? As I said a few minutes ago, DR3 is expressed primarily by lymphocytes. When a lymphocyte has matured in a manner where it has decided to express DR3, that cell will continue to express DR3 for as long as it lives.
It's a stable target. TL1A, on the other hand, is not expressed constitutively. It's an inducible ligand. This is a figure that's showing you that in human antigen-presenting cells, the primary source of TL1A, those cells do not express any TL1A whatsoever unless you have stimulated them with an innate immune stimulus. In this case, they've tested both lipopolysaccharide, a bacterial cell wall component. This is probably what causes TL1A upregulation in an IBD patient or immune complexes. In both cases, you see that those cells will rapidly turn on TL1A, but just as rapidly turn it off. You get this pulse of ligand that will be expressed and then get shut down.
The point here is that when you're trying to block TL1A, you are trying to block a moving target, and you are not trying to do that when you're going after DR3. Those are some of the data that underlie this cartoon here where in an IBD patient, you have these are migratory diseases. You have inflamed parts of the tissue and then adjacent uninflamed tissue. In the inflamed area, you have TL1A. You have locally more abundant DR3. In the adjacent uninflamed tissue, you don't have TL1A. You still have DR3. The problem that this sets up for TL1A blocking antibodies is that there is no means of retention for a TL1A blocking antibody right outside the actively inflamed part of the gut.
If disease migrates, if inflammation migrates, it will set off a race as to whether that newly expressed TL1A is immediately neutralized by a TL1A blocking antibody or has an ability to bind locally abundant DR3. You do not expect that race to be won on either side in absolute terms. You would expect less inflammation but not no inflammation. That is simply not a kinetic battle that you are fighting if you have instead decided to block DR3. The other complex, the other problem relates to immunogenicity. This is data from the Pfizer, antibody that Roche, is now developing, known as Afimcovart. Every TL1A antibody has shown the same thing. At baseline, patients have about 100 picograms per mL of TL1A in their blood.
Two weeks after they get a shot of, in this case, Afimcovart, the concentration of total TL1A in the blood goes up two logs. What every TL1A antibody also does within the same interval is show you that the concentration of free TL1A goes to zero. You have no free TL1A, but total TL1A goes up two logs. What that means is that you're measuring an immune complex between the therapeutic anti-TL1A antibody and soluble TL1A. This is the reason why every TL1A blocking antibody has antidrug antibodies in excess of 65%, of patients that are treated. At least in the case of Afimcovart, published in Lancet Gastroenterology this year, those ADA, facilitate accelerated clearance of the antibody, and with accelerated clearance, they sacrifice efficacy. These are the two aspects that we believe DR3 targeting could improve upon.
With DR3, because it's a membrane-restricted receptor, you simply don't expect immune complex formation. A couple of very simplistic data slides on our antibody. There's much more on our website and from our corporate deck and prior presentations. This is a super high affinity antibody. That's important when you're going after a TNF receptor, because it helps reduce the chances that there would be any residual agonism with a receptor-targeting antibody like this. You also engineer the Fc domain, so that it doesn't bind Fc gamma receptors. That's another way to remove that risk. You pick your epitope really carefully. Those are the reasons why folks probably focused on TL1A blocking antibodies first, because it's a greater engineering challenge to build a receptor-blocking antibody that doesn't have those liabilities. We've tested this antibody preclinically against all of the established TL1As.
This is an example of a simple ELISA, assay where you're asking, at what concentrations does a DR3 blocking antibody prevent binding of TL1A to DR3? And it compares favorably to the Merck and Roche antibodies. One of the most sensitive signaling assays you can do preclinically to both ask yourself, am I sure that my DR3 blocking antibody doesn't have any residual agonism, and am I sure that it can completely inhibit TL1A-mediated cytokine production from effector T cells, is to take lymphocytes from the, in this case, from ulcerative colitis or Crohn's disease patients, stimulate them with an antibody cocktail that mimics a T cell receptor signal, and then titer in your antibody alone in orange or in the presence of TL1A in blue.
When you titer in your antibody alone, you look at, you hope that your antibody doesn't cause any residual activation of DR3. The lead antibody didn't. Others that we had in the original campaigns did, and you weed those out. In blue, this is the amount of cytokine production that is stimulated when you add TL1A to the assay, and then you add in your blocking antibody, and you can bring that back to zero. Before going into phase I, we did a typical acute toxicology study in monkeys. We dosed between 1 and 100 mg per kg, and 325 began to differentiate from the TL1As in this study. Some of the TL1As had lethal toxicity that was secondary to immunogenicity in acute tox studies like this.
We had no changes in clinical chemistry, gross path or histopath, and so the no adverse event level was defined as the top dose of 100 µg per kg. When you're going after DR3, you can also measure receptor occupancy in the peripheral blood. This is very helpful in picking dose, both going into humans and for phase II. DR3 is more abundant than TL1A, so we were pleased to see that we achieved full RO even at the lowest dose of 1 µg per kg. This is a human antibody given to monkeys. We did have a few animals with ADA. We had 8%, of our animals that had ADA. That's typical for an antibody like this.
and that was actually a helpful data point because in those animals, we had a loss of receptor occupancy, and that helps us begin to peg a trough concentration that we might want to maintain in humans to exceed full receptor occupancy. In this case, it was pegging somewhere around 1 microgram per mL. This is very similar to what Vedolizumab saw in monkeys in their original studies. If you had a DR3 agonist in monkeys or humans, the way you would know that is by looking in the peripheral blood a week after you give your antibody for proliferation of any T cell subsets. We also looked at cell surface markers of activation, serum cytokines, and all of the monkey data demonstrated that, as we expected, this was a pure DR3 antagonist. We are now in the middle of a phase I study in healthy volunteers.
This started several months ago. It's going exactly as we had hoped. The key questions that will be answered in this study are: number one, is this as safe as a TL1A blocking antibody? Number two, do we achieve full RO at low doses, and how durable is the RO? Number three, is there no evidence of DR3 agonism in humans? Fourthly, do we have a best-in-class immunogenicity profile? Those questions will be definitively answered by the phase I data, that we will share in the first half of 2026 and set the stage for our phase II trials, which are now funded and expected to begin also in 2026. This is an incomplete view of the core of the competitive landscape on the TL1A side. To our knowledge, it remains a complete view of the competitive landscape on the DR3 side.
We are really excited to share the data from the phase I trial early next year and get into the phase II. The company is funded through those phase II trials at this point, with cash into 2029 as a result of the recent pipe financing that we did. Thank you very much for your attention. Dosing interval? We have not yet disclosed the dosing interval. The target in maintenance is roughly Q8 weekly, but we'll have to see the phase I data to pin that down.