Okay, our next session is with Taylor Schreiber, MD, PhD, CEO of Shattuck Laboratories. Dr. Schreiber is leading the efforts to advance a new target in IBD known as DR3. I say new target, even though the pathway itself is a very prominent one that we are all familiar with, TL1A. And so, Dr. Schreiber, if you'd like to say a couple of introductory remarks on Shattuck Labs and TL1A.
Sure, well, thanks, Martin and David, for hosting the session today. Shattuck, as you mentioned, is focused on developing what is the first in clinic and potentially first in class death receptor 3 blocking antibody. DR3 is the sole signaling receptor for TL1A. As you mentioned, Martin is a well-known target to probably most people who are listening in today, and there are a number of reasons we believe why targeting DR3 may provide superior efficacy to what has already been super exciting data seen to date from multiple TL1A blocking antibodies, and I suspect we'll get into that.
Fantastic. And so, DR3, any reason to believe that targeting the receptor for TL1A could be substantially different from targeting the cytokine itself?
Sure. so, as I just mentioned, one consideration you always have to be mindful of when picking to block a ligand, in this case, a soluble cytokine like TL1A versus a receptor, is one of them promiscuous, i.e., does one side of the axis bind multiple targets, and does the other side only bind a single target? There's many examples of receptor ligand pairs where one end is promiscuous and the other end is not. and so it can be very obvious to go after the receptor or the ligand. In the case of TL1A and DR3, this is a monogamous receptor ligand pair. TL1A only binds to DR3, and DR3 has no other signaling ligands.
So, from a safety and specificity standpoint, there's no reason to believe that what has been learned in the clinic so far to date regarding the safety profile of TL1A blockade will not predict the safety profile of DR3 blocking antibodies so long as your DR3 blocking antibody lacks the ability to engage with Fc gamma receptors, which ours does not. And so then, you know, what are the reasons to believe that there could be differential safety and efficacy or efficacy profiles and tolerability profiles between the two? There's two prongs to that answer. And one relates to efficacy, the other relates to immunogenicity. So I'll answer first by giving our perspective on why the efficacy could be different. And what that relates to are significant differences in how and where TL1A is expressed versus DR3. So TL1A is not a constitutively expressed ligand.
It gets turned on in short waves, and it only gets turned on in response to different innate immune stimuli. So if you were to take a biopsy of someone's gut who doesn't have inflammatory bowel disease, you would find that there would be little to no expression of TL1A in that biopsy. And if you do the same thing, you take biopsies from a patient that does have, let's say, Crohn's disease, and you biopsy the actively inflamed part of a Crohn's disease patient's bowel, what you'll find is that about 8%-10% of the cells in that biopsy will be expressing TL1A. And in that same patient, if you move 2 cm down the bowel, so you remember Crohn's disease is a discontinuous inflammatory disease, so you have inflamed followed by neighboring uninflamed followed by downstream inflamed tissue.
If you biopsy the uninflamed tissue, you'll find that TL1A is not upregulated, just as is in the case in a patient without inflammatory bowel disease. In that same patient, though, if you look for DR3, what you'll find is that about 18%-20% of the cells in the actively inflamed biopsy will be positive for DR3, so about double the amount of cells that are positive for TL1A, and that same proportion of cells will be expressing DR3 in the neighboring uninvolved tissue. Now, the reason for, and so DR3 is both more, it has a broader pattern of expression and is more abundant in patients with IBD. Now, the reason for that is that the cells that express DR3 are primarily lymphocytes and innate-like lymphoid cells.
And when a lymphocyte has decided that it is going to express DR3, it expresses it for as long as that cell lives. The amount of DR3 on the surface can fluctuate over time, but those cells generally don't turn DR3 off. So it represents a stable target. TL1A gets turned on, as I mentioned, in response to innate immune stimuli in short pulses. So it gets turned on for 12- 24 hours at a time, and then it gets turned off again. And then if there's another inflammatory stimulus, it can go up again and come down again. And so you're constantly chasing these waves of inflammatory ligand with TL1A, and you're not fighting that sort of kinetic battle if you're instead blocking DR3. So this is analogous to a number of other receptor ligand pairs.
And for those folks who have followed oncology, as an example, you're probably familiar with the fact that there are both PD-1 and PD-L1 blocking antibodies. And the efficacy of PD-1 blocking antibodies is generally higher than the efficacy of PD-L1 blocking antibodies. And one potential reason for that is that with that receptor ligand pair, PD-1 represents the more stable side of the axis. Just like TL1A, PD-L1 is an inflammatory ligand that achieves pulsatile expression patterns just within the affected tissue, in that case, tumors. So just to summarize, we think DR3 is a much more stable target than TL1A, and that could lead to higher efficacy.
Great. And one point that we also wanted to note about TL1A is because of that transient expression, you could potentially have high local concentrations, whereas with DR3, you would be able to suppress across the board broadly in all tissue locations. The binding. Would that be a correct way of assessing it?
Yeah, I think what we have modeled and what our data are showing is that because DR3 is more abundant, your first dose target-mediated clearance effect is higher than what you see with TL1A as the less abundant target. But then with our antibody, once it binds DR3, it doesn't cause DR3 to internalize, and binding is highly durable. So what that means is that by the time you give that second dose, there is no free DR3 left. So you have rapidly decreasing clearance with DR3 blocking antibodies with repeat dosing, whereas with the TL1A blocking antibodies, because TL1A is rapidly turned over, you see more steady clearance with each dose.
Great, and we were hoping you could also talk about that free versus bound concept, so with the DR3, it being bound, does this prompt any additional receptor to be expressed on the surface? And then with TL1A, as that gets bound, how does that affect the concentration of TL1A, total concentration?
Sure. So with DR3, we don't see changes in the per-cell expression pattern of DR3 when that cell is fully occupied with our antibody 325. So it seems to be essentially inert other than blocking the ability of DR3 to interact with TL1A. And there is no shed form of DR3. So our antibody does not bind any soluble proteins that we can detect. Just as a note for those that are familiar with the axis, there is another member of this axis, which is called decoy receptor 3. That is a soluble protein, but decoy receptor 3 is encoded by a separate gene than DR3. It's not a shed form of DR3. And our antibody does not cross-bind decoy receptor 3, so that remains untouched. Now, decoy receptor 3 doesn't exist in rodents. It evolved in primates.
And whenever you see new evolutionary changes like that, you have to ask yourself why. So what decoy receptor 3 does is it binds to and facilitates the clearance of soluble TL1A-like TNF ligand. Those are all TNF ligands that can be shed. So when TL1A is expressed, it's initially anchored to the membrane, but then there's a membrane-proximal protease cleavage site that leads to shedding of the extracellular domain of TL1A. And that is still capable of signaling. So it's likely that humans evolved decoy receptor 3 to rapidly get rid of shed TL1A. So that tells you a little bit about how important this axis might be in driving different inflammatory processes.
Now, when a patient is treated with a TL1A blocking antibody, those antibodies, depending on whose antibody we're talking about, can bind either trimers of TL1A or trimers and monomers of TL1A, both when that TL1A is expressed on cell membranes as well as the shed forms of TL1A. And so what all of the developers of the TL1A blocking antibodies have shown is that within a week or so of getting the first dose of a TL1A blocking antibody, the total concentration of TL1A in the blood increases by 2-3 logs. So it goes from folks predose from about 100 picogram per mL into the microgram per mL range. And what is actually being measured in that case are immune complexes between the shed TL1A and the anti-TL1A antibodies. And so your total concentration of TL1A goes way up.
Now, immune complexes are a dominant generator of immunogenicity, and so when immune complexes form, they often are internalized, processed, and presented on MHC, and that results in the generation of antidrug antibodies against whatever the protein components are in that immune complex, and this is the reason why every TL1A blocking antibody developed to date has seen rates of antidrug antibody formation in excess of 65% of patients. In some cases, it has been 100% of patients, and so this is a concern. With the Pfizer antibody that Roche is now developing, they recently published a paper in The Lancet that demonstrated that even though there wasn't a particularly high rate of neutralizing antidrug antibodies, that high rate of ADA led to accelerated clearance of Afimcobart over time, and that accelerated clearance was associated with loss of response.
So this is something that IBD clinicians are very used to from their experience with TNF alpha-directed agents that cause the same sorts of immune complex-related immunogenicity. It's the reason why patients often cycle from one anti-TNF agent to another. And it's a reason for loss of response. So with DR3, you don't have to worry about immune complex formation.
Great. So if I could summarize the three differentiators that we are hearing so far in terms of targeting DR3 versus TL1A, first one would be increased broad expression of DR3 allows you to intercept the signal before you start to have that inflammatory spike that you would with TL1A. The second one would be that you are able to shut down, you're able to allow decoy receptor 3 to continue performing its natural inhibitory function on TL1A. And the third would be reduced immunogenicity potential because you're not generating those soluble immune complexes. Would that be correct? Would there be any other differentiators that we're missing?
Those are the key things, Martin.
Fantastic. Okay. So that is the background on the TL1A versus DR3. Let's talk a little bit about what you're doing here. So SL-325, tell us a little bit about this antibody development history and why has nobody else done a DR3 blocking antibody?
Yeah. So 325, as I've alluded to, this is a human DR3 blocking antibody. It's a very high affinity antibody measured at less than two picomolar. And this antibody lacks any Fc gamma receptor binding activity. So it is completely null from an Fc binding perspective. It does still bind neonatal Fc receptor. And the major concern when you're developing a TNF receptor-targeted antibody as opposed to a TNF ligand-targeted antibody is that a TNF ligand-targeted antibody like the anti-TL1As, your only design consideration really is to find an epitope that interferes with TL1A binding to DR3. There are other small design tweaks that certain folks have applied about whether they bind trimers and monomers and whether they bind a site that also interferes with decoy receptor 3 binding or not. But really, the major objective is just to bind an epitope that blocks DR3 binding.
When you're going after the receptor side of the axis, in our opinion, the design goals are to pick an epitope that interferes with trimerization of the receptor, an epitope that also interferes with TL1A binding to DR3, pick an epitope that is not shared with decoy receptor 3, and fourth, and perhaps most importantly, and this represents the biggest challenge, find an epitope and overall antibody characteristics that do not cause residual agonism of DR3. This is the principal challenge, and I would assert the principal reason why we are the first to develop a DR3 blocking antibody is because many antibodies that bind TNF receptors, all antibodies that bind TNF receptors, will at a minimum be able to bring together two subunits of DR3 in this case.
Whenever you are clustering multiple subunits of a TNF receptor, there's always the risk that those cytoplasmic domains come into an appropriate configuration where you can trigger residual signaling. So you can inadvertently end up with an agonist when you're trying to develop an antagonist. This is something that you can study and make yourself aware of with very sensitive preclinical assays, which we did. Many of the antibodies that we developed ended up having residual agonist activity and were triaged at various points in development. We found several candidates that did not have those activities that we chose to bring into non-human primate studies.
And if you were to have a DR3 agonist, the way that you would see that in vivo, in primates, meaning both non-human primates and humans, is that within 5-8 days after administering your DR3 targeted antibody, you would see proliferation of various T cell subsets. And you might see increases in cytokines. You might see changes in activation markers on T cell surfaces. And so our data confirmed that none of those signs were seen with 325 in primates either. So we have a pure DR3 antagonist on our hands. And you can look at this. This is not the first time that this challenge has been encountered in the industry. If you look at another TNF receptor ligand pair, BAF and BAF receptor, right? Benlysta is an anti-BAF antibody that has been commercially available for over a decade, worked in some indications, didn't work in others.
Now you have companies like Novartis and Jade who are coming very late to the scene with BAF receptor-targeted antibodies. And those antibodies seem to be active in Novartis's case and Sjögren's, for example, where Benlysta previously failed, right? You can look at OX40/ OX40 ligand where you have rocatinlimab that is the OX40 targeted antibody that we'll see may have a differentiated profile from some of the OX40 ligand-targeted antibodies. And in general, the receptor-targeted antibodies come a little bit later than the ligand-targeted antibodies, but there is a pattern emerging that they can be more efficacious.
Fantastic, and so far, are you aware of any other developments in progress for DR3 antagonists, barring any surprises from China, but any other competitors that you are aware of that are looking at the same mechanism?
Yeah. Nobody has a disclosed program. And we are not aware of anybody going around and talking about it. We are aware of one highly experienced antibody development pharma company that did prioritize DR3 over TL1A for the same reasons that we have. And the antibody that they developed ultimately had some of the agonist issues that I was speaking about earlier in the same assays where our antibody does not. So there are efforts, no doubt, ongoing at multiple companies. But I think evidence that one of the world's most experienced antibody developers encountered that issue gives us some degree of confidence that this isn't going to be a simple fast follower story.
Great. And speaking about fast followers, so you are now, tell us about your timeline. So when can we expect to see first clinical data from this program? When can we expect to see it starting to advance into IBD? What are your thoughts on IBD advancement in terms of selecting between UC and Crohn's disease?
Yeah. So the key questions that we have to answer are, number one, is it true or not true that targeting DR3 has a superior immunogenicity profile than targeting TL1A? Our phase I study is moving along swiftly right now, which is great. And our phase I data will definitively answer that superior immunogenicity question. We expect this phase I study to be complete in the first half of next year. And we are prepared to move immediately into multiple placebo-controlled phase II trials. The second question, obviously, is how does the efficacy of DR3 targeting stack up to anti-TL1As? And we'll get into the other areas where other indications that are exploring the activity of anti-TL1As. But today, the only clinically validated place for anti-TL1A is in ulcerative colitis and Crohn's disease. So one of our placebo-controlled phase II studies will be in IBD.
And we will use that data to definitively answer the comparative efficacy question. Folks will have to do cross-trial comparisons, but we're going to size and design our study in such a way that hopefully that is not a challenge for folks who are looking at the data. So it'll be at least one study in IBD and then one study in another indication.
Got it. Okay. So first half of 2026, we will keep an eye out for that data. And then this does follow along with what Dr. Dubinsky was saying in terms of comorbidities. The more indications that a drug is allowed in, it is useful in, the more attractive it becomes to clinicians. What are your thoughts on the most promising new indications for TL1A? And do you have any reason to believe that in certain indications, DR3 might be more efficacious than TL1A?
Yeah, so the reasons why development of TL1A blocking antibodies began in IBD and asthma, which a lot of people forget, but are related to the fact that there is a well-known risk of developing Crohn's disease, not so much ulcerative colitis, if you are born with specific single nucleotide polymorphisms in TL1A that lead to aberrant expression of TL1A throughout your life, and there are other autoimmune diseases where there is also a known risk with those same single nucleotide polymorphisms. That includes things like psoriatic arthritis and psoriasis. It includes things like primary biliary cirrhosis and a number of other diseases, and when you look, and so that sort of is one filtering criteria is where are these TL1A SNPs potentially driving pathology.
If you just look at the difference between Crohn's and UC, though, and the knowledge that anti-TL1A works in both Crohn's and UC, UC doesn't have an association, a risk association with TL1A SNPs. What it does have is patients are known to have elevated serum concentrations of TL1A at the time of diagnosis relative to non-IBD patients. And so that elevated serum TL1A is a second filtering criteria. And you can find elevated serum TL1A in a pretty wide range of autoimmune diseases. It includes asthma, UC, Crohn's disease, psoriatic arthritis, psoriasis, rheumatoid arthritis, again, primary biliary cirrhosis, and a number of other diseases, axial spondyloarthritis . And so you can use that too. And then you can look at where are there preclinical data that interfering with TL1A DR3 signaling is protective from disease.
And again, of those diseases, there's strong preclinical data in mouse models of inflammatory bowel disease, asthma, multiple sclerosis, interestingly, and also the arthritides. And so those are all criteria you can look at. You can also ask, okay, where of all of these diseases do the downstream effector cytokines, where are they clinically validated? Because TL1A DR3 signaling is upstream of the production of multiple inflammatory cytokines, including IL-17 and IL-23. And so you can look at the places where those antibodies are already clinically validated and hypothesize that the risk of success or the chance of success for a TL1A DR3 blocking antibody are high. And so when you do all of that, you end up with a long list of diseases where, in many instances, there remains very high unmet medical need.
So when you look at the fact that Merck and Roche and Spire have all chosen to go and begin phase ii studies in rheumatoid arthritis, I agree. That is a good choice of a disease to go into. TL1A DR3 is also known to play a role in the development of fibrosis, right? So there's a number of diseases there. Merck going into their SSC-ILD study is probably the first chance we'll have to really look at data that might speak to the fibrotic aspect of this axis. Atopic derm is another place where Roche is going. And that is another disease which ranks highly on the list based on the criteria that I went through. But so does asthma, right? And that's why Teva started in asthma before they went into IBD.`
I would urge caution, more caution in thinking about something like asthma because one of the commonalities between diseases like asthma and atopic derm is that the antigens which drive disease are not ever present, right? In IBD and RA and psoriatic arthritis, they are. And I think interfering with this axis in diseases that have a more seasonal or relapsing-remitting course is less of a sure bet than going into diseases where tolerance to either self or endogenous microbial antigen has been broken and that that antigen is perennial in the patient.
Fantastic. Well, thank you so much for sharing your thoughts on this, Dr. Schreiber. And David, I do apologize. We're out of time for this session. However, we are running now into our Q&A. And so there will be.