All right, great. Thank you, everyone, for joining. My name is Frank Brisebois. I'm one of the biotech analysts at Oppenheimer. The next company presenting here is Sensei Biotherapeutics. From the company, we have the pleasure to have John Celebi, the CEO of the company, to present. What we'll do in terms of format here is we'll have John present for about 20 minutes or so the company, and then we'll jump in with some Q&A. If you have any questions at all, feel free to send them in the tab, if not by email. With that, John, thank you very much for joining, and I'll let you take it away.
Frank, thank you so much. The timing could not be better. Sensei has a major milestone coming up in the second quarter of this year, so we are excited to bring investors up to speed on where we are currently. We are happy to keep this conversation with you, Frank. If you have any questions as we present the story, do not hesitate to interrupt. Just a quick note that we are a publicly traded company, and we do have some forward-looking statements. I want to give everybody just a 50,000-foot view of the company before I jump into our lead program, which I think is where we will focus today given the time limitations. We do have a lead program that is in the clinic. It is early in the clinic.
It's completed dose escalation, but it is now in a dose expansion study, and that's the milestone I referred to earlier on. It's a very innovative program called solnerstotug. It's a conditionally active antibody, and I'll explain what that means. It's targeting a very important immune checkpoint called VISTA that's had some challenges in the past that we've been able to overcome. We have a platform that generates our antibodies called TMAb, and this is what we use to generate these conditionally active antibodies and open up new targets that have been very difficult to drug in the past. In terms of financials, we have cash runway into the second quarter of 2026, which is sufficient to complete our phase one. As I said, we'll be focusing on our lead program today. I want to give everybody a little bit of background on VISTA.
If you've been involved in immuno-oncology, you're familiar with the concept of how VISTA works. As you can see from the cartoon on the lower left, it's roughly analogous to how PD-1 and PD-L1 work together. That is to say, when VISTA and its T-cell receptor are engaged with each other, they tend to create an immunosuppressive environment within the tumor. T-cells cannot proliferate. They cannot activate. If you disrupt that interaction within the body, you now unleash the brakes, so to speak. You now have T-cell proliferation and activation. However, there are two very important features of VISTA that I want to highlight here that are important. One of them is that the expression profile of VISTA is quite different than PD-1, PD-L1. Yes, VISTA is found on tumors, but it is found to a very heavy extent on myeloid cells, so neutrophils, monocytes.
That's important because these are some of the most widely expressed immune cells within the body and can create issues, toxicity or pharmacology. I'll talk more about that in a moment. The other important feature of VISTA that's important to remember is that it is pH-sensitive in its own right. From the crystallographic cartoon below, you can see we've got certain amino acids circled. These are important amino acids within VISTA because they become more positively charged under low pH conditions. That's what enables VISTA to interact with its T-cell receptor, PSGL-1. It interacts with these tyrosines on the surface of PSGL-1 only when those histidines are positively charged. When they're not positively charged, it does not interact, and it is not a checkpoint. That's very important to remember, and I'm going to come back to that in just a minute.
Before I do, I think it's important to highlight the commercial potential of VISTA. We already know that immunotherapies are quite successful. PD-L1 targeted therapies are one of the largest classes of drugs within all therapeutic areas. Last year, the top five PD-L1 targeted therapies generated close to $49 billion in sales. That's important because VISTA is also an immune checkpoint. It is combinable with PD-1. These pathways are orthogonal to each other, but it's actually more widely expressed than PD-L1. As you can see from the chart in the middle there, it's found at higher levels than PD-1 and PD-L1 and all the other immune checkpoints on this chart. This is from a recent publication. As you can also see from the chart there on the right, most solid tumor indications have VISTA within them.
A lot of that has to do with, again, its expression pattern. It's found in myeloid cells. Myeloid cells are not scarce within tumors, which can sometimes be the case with T-cells. So VISTA does play a very important role. It does seem to have a large commercial potential, and it's combinable with PD-1. However, VISTA does come with some baggage. There has been prior work within the space, most notably from Johnson & Johnson, but there was also a French pharmaceutical company called Pierre Fabre. Both developed antibodies targeting VISTA, and both encountered difficulties early in development. One of the main difficulties was dose-limiting toxicity. Johnson & Johnson actually terminated their phase I study after treating about a dozen patients for grade 3 CRS-associated encephalopathy. It was done at a very low dose of 0.3 mg/ kg. Within the world of antibodies, that's a pretty low dose.
The other issue that they ran into, in my opinion, was a PK profile that was quite challenging and really not commercially viable. As you can see from the chart on the right, you're looking at levels of the antibody in human patients over time. You can see that the x-axis only goes out to about 26 hours. Most antibodies are dosed every two weeks or three weeks or longer. For example, pembrolizumab or Keytruda is dosed once every three weeks. That's about the cadence that you want to dose an antibody. A 26-hour half-life is too short. We looked at this data very closely, and we surmised that these were class effects of targeting VISTA rather than a property of the antibody itself because both J&J and Pierre Fabre ran into these issues.
We believe that these issues were both driven by the fact, again, that VISTA is found quite extensively on the surface of myeloid cells. Myeloid cells, when an antibody binds to VISTA on the surface of a myeloid cell, will release cytokines. It is no surprise that you could run into CRS. Quite often, the antibody is internalized, which removes the antibody from circulation, which can create a poor PK profile. In order to overcome these issues, we created what we call a conditionally active antibody that was specifically designed to take advantage of VISTA's pH sensitivity. We created an antibody that was very pH sensitive, and it only really binds to VISTA in its low pH form where it is more positively charged.
As you can see from the chart, the table on the left, VISTA is quite a potent binder, picomolar-level binding at pH 6. There's virtually zero binding at pH 7.4. That's a really important feature because it allows us to avoid binding to VISTA in the blood, but it will bind to VISTA within the tumors where there is a more acidic environment. We also confirmed that VISTA does disrupt the interaction with PSGL-1 quite potently. We co-crystallized these VISTA and PSGL-1 together with the antibody and confirmed the binding was quite unique. We were very pleased with the progress. We went on to test this antibody preclinically. As you can see from the cartoon on the left, the idea was that the antibody would bind exclusively within an acidic environment such as the tumor. Solnerstotug will bind VISTA when it's more positively charged.
It will not bind VISTA when it's more neutral pH. As you can see from the data on the right and the upper panel, we've got s olnerstotug compared to both the J&J antibody and the Pierre Fabre antibody. You'll notice those curves shifting to the right, the gold and the blue. That indicates binding. This is done in monocytes and neutrophils of mice, but the solnerstotug antibody does not. It stays put, which means there is no binding at this neutral pH environment. However, in the tumor, in the lower panel, you'll see in the tumors of these mice, or different mice, the antibody accumulates quite rapidly, and it stays within the tumor for an extended period of time. This data was quite exciting for us to see preclinically. What was important at this point was to design a clinical experiment that could confirm this data in humans.
Before I get to that, let me just say a brief word about the competitive environment as we get this question asked a lot. As you can see here, there are a number of competitors in the space. From the column on the left, you can see the competitive space as we define it. We think there are three important features for an antibody to be successful targeting VISTA. One, it has to inhibit the binding of its T-cell receptor, PSGL-1. We believe it has to be pH sensitive to be successful, which so far the data to date has confirmed. It has to have an active FC. There's only one other competitor that has these sorts of profile, and that's from BMS, and that's a single preclinical stage competitor with this type of conditionally active approach. We were on to the clinic.
We dosed our first patient roughly 18 months ago. It was very important for us to rapidly confirm that solnerstotug had the properties we believed it needed to get to an active pharmacological dose. That is well above 0.3 mg/ kg. We actually started our dose escalation study at 0.3 mg/ kg. Recall that is the same dose where J& J had to terminate their study. We designed this dose escalation to be an all-comer study, and we did that quite consciously, knowing that we would be able to enroll the study quite rapidly, which was really critical because we wanted to de-risk as quickly as possible the CRS and the PK issues that I have already highlighted with prior competitors. We were able to dose escalate quite rapidly, and we completed enrollment within about a year, both the monotherapy and the combination aspect of this dose escalation study.
There was really nothing unusual about the patients on board here. I won't spend a lot of time on this slide. You can see we had 16 patients enrolled in the monotherapy and 18 in the combo. I think based on what we saw in this dose escalation and everything we've seen since, we've updated this safety information from time to time, we can say that solnerstotug was well tolerated. There were no dose-limiting toxicities observed during dose escalation. The majority of the adverse events were grade 1 or 2. We did see some CRS, but it was minor CRS. Actually, that gave us some encouragement that we were hitting the target and hitting it well, but we were probably restricting the binding to within the tumor. CRS was quite limited. This was grade 1 CRS. I don't believe any hospitalization was required.
We also saw some infusion-related reactions, which can sometimes look like CRS. That was grade 2, but pretty mild overall. We were encouraged again that this toxicity, which is a class effect of VISTA antibodies, was present, although at a very low grade. In terms of the PK profile of the drug, you can see that the drug has a very long half-life. Again, just looking at the x-axis on the left for the mono and on the right for the combo, you can see that this goes out for thousands of hours, supporting what we're doing currently, which is a once-every-three-week dosing regimen, could be even potentially longer in future studies. The phase one dose escalation affirmed what we set out to do. It was well tolerated. We believe this drug has best-in-class PK among all the ISTA competitive fields.
We did get to inactive pharmacological dose levels. Based on our preclinical models, we think anything at 3 mg/ kg or higher should be an active pharmacological dose. We were quite pleased with that. Because we enrolled all-comers, most of the patients that came on to dose escalation had patients with what we call cold tumors. Cold tumors are those without T-cells within the tumor. While there is reason to believe that a VISTA antibody could have activity in cold tumors, it is not easy to understand why, given the lack of T-cells. We did look at the efficacy component of this study, but 85% of the patients had cold tumors, as you can see from these waterfall plots. Unsurprisingly, perhaps there was not a lot of activity in the cold tumor setting. However, we were fortunate to enroll six patients with what we call hot tumors.
Those are more likely to respond to immunotherapy. The patient numbers are too small to quantify, but qualitatively, you can see there's a difference here. We had a patient with an endometrial cancer patient that had a partial response, a renal cell patient that had significant tumor shrinkage, and even one monotherapy patient that had a head and neck patient that had significant tumor shrinkage. This gave us encouragement that we were potentially on the right track. We then proceeded to go to the next part of the study, which was dose expansion. Here, we had two components of this study. We had both a monotherapy component focused in microsatellite stable colorectal cancer, and we had a combination therapy component at two doses, 3 mg and 15 mg per kg.
That was more of a basket study across indications that are hot tumors, again, patients with T-cells in the tumors that are more likely to respond to immunotherapy. Why did we enroll patients with microsatellite stable colorectal cancer? These are among the coldest tumors out there, and the response rate to PD-1 and PD-L1 is roughly zero. It's because we saw this one patient here. It was quite interesting to us that had a close to a PR, about a 27% tumor shrinkage. We wanted to understand more about the mechanism of action. We proceeded to enroll about 10 patients in the monotherapy component of that study. That is a little bit more of a science experiment. We are taking biopsies where we're possible, trying to learn about why that could be, why are we seeing activity there.
If we do, it could open up additional avenues for clinical development. I really want to focus on the combination arm because here is where we're enrolling the hot tumor patients, and that's where we're more likely to see activity. What we updated on very recently was that we had enrolled about 75% of the study and that we can anticipate completing enrollment in the dose expansion by the end of Q1. We are exploring two dose levels in the combo to optimize the dose for a phase two study. One thing that's really important to emphasize is that not only do these patients in the combination setting have hot tumors, they also have prior experience with PD-1. Not 100%, but the vast majority of these patients have prior experience with PD-1. Why is that important?
It's important because most of the patients that have had prior PD-1 and then progressed on PD-1 have a single-digit response rate upon re-challenge to PD-1. Because this is a combination study, it's important for us to be able to say the activity that we might see is coming from our drug so we can separate that activity out from the PD-1 activity. That's the milestone that we plan to announce in the second quarter of this year. We're looking forward to sharing that information. Frank, I'm going to stop here. I'll hand it back to you if you have any questions, or maybe you have some from the audience.
No, that's great. Thank you very much. I just wanted to, I guess, a couple of things. The conditional activity, is this kind of the way of the future here in oncology? Why did you guys do this? And is this something that, whether it's pH dependent or not, is conditional activity kind of a hot space?
Yeah, it is. It is an important feature that is sort of among a class of next-generation approaches to developing antibodies. Of course, antibodies have been known for quite some time. I think the low-hanging fruit has been plucked. The question is, how do you take advantage? How do you take selectivity to the next level? Creating binding that is more selected to certain compartments is the way to do that. This approach has been known for some time. It's just been incredibly challenging. There's actually an FDA-approved drug that uses this feature outside the oncology space, but still based on pH-sensitive binding that improved the pharmacology of that drug compared to its first-generation competitor. These principles are well known. They're well understood. They've just been very difficult to apply. I think Sensei has been fortunate to identify antibodies so far that have these really exquisite pH-sensitive binding features.
Great. Another question that comes up sometimes is the tumor acid environment. Is that an environment that's, how much, you showed the difference between pH 6 and pH 7.4? Is that heterogeneous? Is it variable, or is it pretty consistent across the tumor microenvironment?
What I would say is it's widespread. There is some heterogeneity. It's not like a light switch flips on and you go from pH 7.4- pH 6. It's not entirely clear if there's a gradient or sort of a border around the tumor. It's also not entirely clear the conditions that drive that. What is known is that many, many tumors, probably the majority of tumors, have this low pH environment within them. It is driven by tumors that are rapidly proliferating. It is called Warburg metabolism. That biology has been known for nearly a century, believe it or not. It is a different way to metabolize nutrients within the tumor.
Tumors switch to that sort of metabolism when they are proliferating very rapidly. One really important feature of VISTA that relates to your question that I should mention is VISTA is really only acting as a checkpoint within a low pH tumor microenvironment, which tells you that it is very likely widespread within tumors. It also suggests that there may be this biology that occurs at a low pH environment and makes it probably pretty unlikely that this is the only target that acts that way. We are interested in other targets, and we're looking across the space to see what other targets can we identify that act as checkpoints or have other unique features in low pH environments.
Okay, great. I think it's been a little bit of a tricky market in biotech for the past couple of years, I'd say. We kind of get out of it, and then we get some data coming up in the second quarter. I was just wondering if you could take a second to maybe set expectations a little bit around what people should hope to see here or what you guys would consider successful, just because I think there's been some misunderstandings with the press releases, and just the market itself has been difficult.
The market has been very challenging. I think there are a lot of biotech companies out there that would agree with that statement. I think the really important thing to remember about VISTA is it does have really massive commercial potential. Many investors have switched their focus to late-stage, highly de-risked assets. Makes perfect sense in a resource-constrained environment. There are a number of targets out there, a number of drug development programs out there like VISTA, where these are not niche drugs.
These are large, potentially very large drugs along the lines of Keytruda. What we have been able to demonstrate so far in terms of de-risking is that we have overcome the safety hurdles of prior-generation efforts in terms of safety, in terms of the pharmacology. Our next milestone is squarely focused on the efficacy component of these drugs. That is what we are focused on. The initial goal was, let's show that we can overcome the hurdles that J&J ran into, the safety, the PK. Now we're focused on activity. I think that's what investors want to see in oncology at the end of the day is activity.
Okay. Great. You talked about the cold and the hot tumors. Can you just remind us here in this phase one, the dose expansion data, how much is that as cold versus hot?
In the dose expansion data, we're projecting to enroll roughly 60 patients. Fifty out of the 60 patients, roughly, round numbers are going to be hot tumors. We only enroll about 10 in the cold tumor setting.
Okay. The colorectal case study that you said is turning into a little bit of a scientific experiment was interesting in terms of its potential for the future. Can you just go over that one more time and just kind of show what you had shown and the thought of going forward with that specific patient and the learnings from?
Yeah. I think the interesting thing about microsatellite stable colorectal cancer is that they're completely unresponsive to PD-1. I didn't even say it's low single digits. It's 0% response rate to PD-1. We saw a patient that had 27% tumor shrinkage. The question is, why? Why are you getting a tumor? Why is an immunotherapy drug having 27% tumor shrinkage if there's no T cells in the tumor? We want to understand that. We wanted to enroll 10 more patients to see, do we see this again? If we do see it again, what's driving that? What's the mechanism at play? If we can understand that, it could be quite meaningful for future drug development.
I think in terms of prioritizing, we have to, like any company, we have to prioritize what is the most likely in terms of the first pathway that we're choosing. What is the most likely area we're going to see activity? That's going to be in the hot tumor setting. That's where we're going to explore first. For the future, which we're keeping an eye on, we want to understand what are all the unique features of this drug. That's what we're doing there with the cold tumors.
Okay. No, I don't need to push too much on this either. The phase one dose expansion data that's about to read out here, do you give any kind of, you share what you're hoping to see in terms of endpoints on the efficacy side, anything there? Is it, look, we're looking for activity at this point?
Yeah. I mean, this is too small a study to focus on overall survival or progression-free survival. I think there's really two things that we'll be looking for in this dose expansion study. One of them is obviously going to be the activity. Are we seeing tumor shrinkage? If we're seeing tumor shrinkage, how many of those are PRs or CRs? That's important. Everybody wants to understand that early on. However, the other important thing, and I can't emphasize this enough because this was a feature of Keytruda and Opdivo as well, is how long are these patients staying on study? Remember, these are patients that have been on and progressed on a previous PD-1 inhibitor. If they're staying on study for quite some time, that's meaningful for these patients. That's a potential. It's a hint that in a future, larger randomized study, you could see statistically significant progression-free survival or overall survival.
Interesting. It's also a good look into safety as well.
Yes.
Excellent. I think that kind of wraps it up. Is there anything else in terms of closing remarks that we didn't talk about that you wanted to mention here, John?
We do have a pipeline of programs beyond SNS-101. Those are on our website. We didn't have time to talk about them today, but they're all quite interesting as well.
Excellent. All right. Thank you very much for your time today.
Thank you, Frank.