Today we're thrilled to have John Celebi, CEO of Sensei, here with us. John, I wonder, just for our audience, if you can just give us an overview of Sensei and what you are focusing on.
Yeah. Thank you, Li. First of all, thank you to you for moderating and Cantor for hosting us for this conference. It's been great to see everybody, and for those of you that aren't familiar with Sensei, we are an innovation-focused, clinical-stage biotech company, focused on conditionally active antibodies. That is to say, antibodies that have specific action within the tumor, but not action outside the tumor. There are a whole host of targets for which those sorts of antibodies are well-suited. Our lead program is now in the Phase 1b portion of a Phase 1/2 study. We recently announced our first set of data at ASCO this year, and we're looking forward to announcing our best next big bolus of data at the end of this year, so it's a big year for Sensei.
Great. Maybe, let's go into your platform a little bit. Tell us a little bit of how it works.
Yeah. So conditionally active antibodies are a set of antibodies that are really designed to engage their targets only under certain conditions. In Sensei's case, we're using the inherent pH sensitivity of the tumor microenvironment to leverage conditional activity, and that has a few key advantages relative to other conditionally active approaches. I think the key one being that pH-sensitive antibodies have activity that's reversible.
That is to say, they bind to their targets within the tumor. If they happen to migrate back out, they no longer have action. If they tend to migrate back in, now they can bind again. And that's a key leverage point relative to other approaches, where there might be some de-masking that occurs within the tumor. If those antibodies migrate back out, they can now bind target thereby potentially creating toxicity. So that's been the approach we've used for all of our programs so far, both monoclonal antibodies and our first bispecific program, which is conditionally active.
Now let's dive into your VISTA program. So talk a little bit about, you know, this target. Obviously, there's some prior work has been done around this target. So I guess for you guys, what is sort of the problem that you're trying to overcome?
Yeah. So our first target, for our, the subject of our lead program is a target called VISTA. VISTA is a highly interesting molecule from a biology standpoint, but also because its history, as you were alluding to there's been some prior work done. First of all, on the biology side, we know that if you knock out VISTA in mice, these mice tend to develop sort of, immune-related disorders.
So that tells you right off the bat that VISTA is an important molecule within the immune system, plays a very important role. Second, within the rubric of oncology, when you combine VISTA inhibition with PD-1 or PD-L1 inhibition you almost always see, additive or synergistic activity, in a non-redundant kind of way. So, there are some advantages to that approach. And then third, very interestingly, VISTA itself has enhanced, immunosuppression under low pH conditions. Again, this ties back to our platform approach. So it really is a perfect first target for what we call our TMAb platform, Tumor Microenvironment Activated Biologics.
With regard to the prior history, there are a couple of programs to point to, which, in a sense, make drug development easier for Sensei because we know what the shortcomings were of the first approaches, and we've specifically designed our molecule to avoid those shortfalls. The first one is, you know, a study that was conducted by J&J in 2016 that was terminated after 12 patients for severe CRS. And CRS, it turns out, as we found out much later, is a class effect of VISTA inhibition.
Had nothing to do with the antibody itself, it was the biology of VISTA, as other antibodies had the same issue. So that's one hurdle we wanted to overcome. The other hurdle is, if you dive into the data, you'll see the half-life of these antibodies was about 24 hours which is not really commercially viable. So we felt that the pH-sensitive approach to blocking VISTA would solve both of those challenges, and that's why we chose VISTA as our first target for our platform.
Now, let's talk a little bit about your phase one study. Walk us through the design and what is sort of the key findings so far?
Yeah. So the design, I think, will not surprise any of your guests here. Fairly traditional Phase 1/2 study. The Phase 1a portion of that was a traditional dose escalation, both for monotherapy and in combination with Regeneron's Libtayo, an approved PD-1 blocking antibody. The Phase 1b portion is a dose expansion study. That's the subject of our enrollment right now. And then there's a Phase 2 design that we'll start talking about a little bit more, as we head into our FDA meeting at the end of this year. You asked about the results so far?
Yes.
So we announced those at ASCO. Essentially, what we were able to show at ASCO is that we overcame the hurdles of the prior antibody, specifically the toxicity associated with VISTA and the poor pharmacological properties. So in terms of toxicity, we showed no dose-limiting toxicities. We did have some Grade 1 CRS in our study, which was actually highly encouraging. Shows we're hitting the target.
But not so severely outside the tumor that we're inducing massive cytokine release.
So we were very encouraged by that. In terms of the PK, the PK of our antibody of the half-life in our body goes out weeks. So we're talking weeks, not days, as we did with J&J. So having overcome those first two hurdles, our focus now turns to the dose expansion and really focusing in on the activity. Despite the very heterogeneous population that one typically gets in dose expansion, we did see some initial signs of in-tumor activity.
Which we found highly encouraging. In fact, we found that despite the fact that 85% of the patients enrolled during dose escalation had what we would define as cold tumors. Cold, cold tumors, in this case, meaning tumors that typically these patients would not respond to a PD-1 blocking antibody. So we felt that was very positive data.
Now, in terms of the doses that you pick, obviously, you were testing a range of doses, right? Just how confident are you, given that you haven't seen a lot of talks, that you're actually at the top dose, and you cannot push the dose potentially higher?
Yeah. So in terms of the dose range in our phase one, we started at 0.3 mg per kg. That's actually the dose level where J&J had to terminate their study at 0.3 mg per kg. That was our starting dose. So we felt we were on pretty good footing there because of the selectivity of our antibody. Nonetheless, the J&J experience also shows that dose can be quite potent. So that was the one sign that we were feeling we were starting at a fairly high dose level.
Then we dosed all the way up to 15 mg per kg, so a dose 50-fold higher than where J&J got. And based on our preclinical models, we felt that was well within a pharmacologically meaningful level. In fact, what we've said publicly is that anything 3 mg per kg or higher is very likely to be a pharmacologically relevant dose, and in fact, now that we're in dose expansion, we've chosen two doses to test. One is three mg per kg, and one is 15 mg per kg, and we feel there's a good chance that we could find three mg per kg as effective as 15 mg per kg.
So you had one responder in endometrial cancer from the combo arm. So what is your take from that patient, given that, you know, it's a combination, and I believe that patient is naive to IO?
Yeah, that's right. There's a couple of examples from Phase 1 I would point to if we're talking about activity. First of all, in terms of RECIST responders, you're right. That patient was a PR. That was a patient that was checkpoint naive, but this was a patient with MSS, microsatellite stable endometrial cancer.
So, according to the literature, patients with microsatellite stable endometrial cancer have response rates to PD-1 that can range anywhere from 0% up to about 10%. So you could argue that the activity we saw was due to Libtayo, but it's unlikely. We'll put it that way. There were other patients I would point to that may be even more encouraging. One of them was a patient with microsatellite stable colon cancer with that experienced a 27% tumor shrinkage. Now, there, if you look at the literature, the response rates to PD-1 are exactly 0%. So I think it's hard to argue that that activity was due to the PD-1 agent, very likely that that was due to SNS-101 alone.
And we also had another patient that had an 18% tumor shrinkage following a fairly serious immune-related adverse event but then stayed on study and experienced a very profound tumor regression. We also had a patient in our monotherapy cohort with renal cell carcinoma that had 17% tumor shrinkage. So again, I think, you know, looking at Phase 1 data is always more of an art form, but given the patient population, we felt very encouraged by where we stood, and we're very keen to go into the dose expansion in a more homogeneous population.
Let's talk about the expansion cohorts that you guys are in. What are some of the tumor types that you guys are looking at? Obviously, you're going with two doses, as you mentioned, right? Maybe just set the expectations for us. What do you hope to see from the expansion cohorts?
Yeah. So we're looking in the expansion cohorts, both at monotherapy and combination therapy with Libtayo. In the monotherapy setting, we look specifically at microsatellite stable colorectal cancer, and the reason to do that is obvious.
Yeah.
I mentioned the one patient that had activity. There is evidence in the literature to suggest that VISTA, anti-VISTA inhibition is orthogonal to PD-1. And so the hypothesis we're testing is in these patients with relatively cold tumors that may have a innate or acquired resistance to PD-1 blockade, can you see continued activity with VISTA monotherapy? So that's one hypothesis we're testing. In the combination setting, it's a different grouping of patients. These are patients with non-small cell lung cancer, melanoma, head and neck, or MSI-high colorectal cancer. So these are what we would call hot tumors, and the reason for that is that in the dose escalation study, 85% of the patients enrolled had cold tumors.
So we had very little opportunity to test the combo regimen in the hot tumors. So that's where we're focused on now. And we're testing actually two doses there, fifteen and the three mg per kg dose.
I guess, can you help us to understand sort of what the bar is, for both the monotherapy and the combination? Obviously, there are different tumor types that you guys are looking at. Obviously, they probably have different settings and prior treatments. So how should we think about the bar for you maybe to moving to the next step for a go or no-go decision?
Yeah. I think there's two totally different bars to consider if you're looking at the mono versus the combo. In the mono, I think the bar is very low.
These are patients that do not respond to monotherapy. I think if you see any activity in that setting, you should be delighted, and we would be, too. In the combo setting, the bar is a little bit higher. Nonetheless, we are focusing on patients that have some form of resistance to PD-1. The main point of that, of course, is to help us separate signal from noise. It's very tricky when you're doing single-arm studies to determine what level of activity is due to your agent versus the co-administered agent.
So by selecting patients that have some level of resistance to PD-1, we hope to gain more confidence that, going into a Phase II study, we will see positive results, and that's the thought there. Typically, in the tumors we're looking at, if you see a response rate of about 10% or higher, you feel pretty good about it, coming up, and, you know, from a PD-1 resistance standpoint.
So are we gonna see data later this year from the expansion cohort?
Yeah. We've guided to initial data in the expansion cohort Q4.
Okay. Just curious, is there a biomarker strategy here? Maybe look at, you know, VISTA expression, PD-L1 expression. Is that something on the table?
Very much so. I think probably all of your listeners are very aware that, you know, biomarkers are a major challenge in immuno-oncology. Patient selection strategies are at a premium, so we've invested, I think wisely, in our biomarker strategy. Of course, we're looking at VISTA and PD-L1 status, but we're looking beyond that as well.
We're looking at taking tumor biopsies to understand the role the tumor microenvironment is playing in resistance to these immunotherapies. We're looking at flow analyses from blood, circulating tumor DNA, and you know, back in June, we also announced some very interesting results from our flow data that showed there was a shift in the phenotypic T cell population that was dose proportional, so we're following up on that, and you know, the goal there is to use that biomarker data to help us further refine our patient selection strategy in Phase II.
The Phase I meeting with the FDA, what is the primary objective there?
Main objective is to gain further clarity on the Phase II design that we intend to pursue and start early next year. That's the main goal of the Phase II meeting. Of course, every time you go to the FDA, you also provide a data update, so you get feedback on that as well.
So what are some of the, I guess, different options for the Phase II design that you guys are thinking about?
I think traditionally there have been two main strategies.
I'll tell you which one, you know, we favor at the moment. You know, people have pursued both single-arm Phase II studies and randomized Phase II studies. I think for an agent like SNS-101, there's a couple of factors to consider. One is the fact that you're treating in combination with another agent.
You may, but don't necessarily have a patient selection strategy, although, as I mentioned, we are selecting patients with hot tumors. One could consider that a patient selection strategy. So those are two very important considerations, and for those reasons, I think we're leaning a little bit more heavily towards a randomized study design for Phase II that can really separate signal from noise, and help us come out of Phase II with a really clear answer about the level of activity with 101 versus PD-1 alone.
So you also have some preclinical programs. Just tell us a little bit about the work that you're doing there.
Yeah, we have some really interesting preclinical programs that we temporarily paused IND-enabling studies for in order to extend cash runway. But nonetheless, these are programs we're actively working on mechanistically and trying to understand the biology of the target. All of these programs are either monoclonal antibodies or bispecific antibodies that are conditionally active, just like our 101 program. We're going after various targets. One of them is called VSIG4, which is a highly novel target with myeloid biology that we understand very, very well. Another one is targeting the adenosine pathway, which has been the subject of prior attempts to treat with monoclonal antibodies, but met with limited success. Again, another candidate for our platform. And the bispecific program is leveraging a pH-sensitive way to co-stimulate T cells via CD28.
So that's very exciting to us as well, and I think all of these programs are actually very difficult for us to rank order because we're excited about all three.
So maybe comment on your thoughts on some strategic, you know, partnership. Obviously, you have a nice platform. You generated some, you know, early clinical data that looks quite encouraging. And it sounds like you have some, you know, really interesting preclinical programs here. So how are you thinking about, you know, potentially leveraging some partnership?
Partnerships are at the top of our list when we think about strategic moves that Sensei could make. I think there are a number of strategies one could employ. You know, I think the challenge right now for all immuno-oncology is getting pharma's attention.
Yeah.
I think we're in a period right now where we're coming off of several clinical data readouts that were negative due to sort of an approach that people took in the early days of after Keytruda's approval, which is, "Let's just combine this with whatever we can." So no surprise, many of these studies failed, and in the intervening years, biotech companies got better at developing antibody drug conjugates, radio, targeted drugs, which I think have really captured pharma's imagination right now. So yes, we talk to pharma companies about our programs, but when you think about conditional activation, it can actually be applied to many different approaches. Antibody drug conjugates, CAR Ts, radiopharmaceuticals.
There's reason to believe conditionally active approach, and specifically pH-sensitive conditional activation, could work, and I think that's been very interesting to talk through with potential partners, to find ways to leverage the interesting mechanisms we've been pursuing at Sensei.
Maybe just lastly, talk about your cash and cash runway, and what are some of the financing options that you guys might be looking at?
Yeah, happy to. So, we ended Q2 with $52 million in cash. We've kept our burn rate very reasonable and manageable, and so we're well-financed into the fourth quarter of 2025. We could extend that further if we chose to. And we've done all that despite the fact that we chose not to raise any money, even since our IPO in 2021.
So we've been fairly conservative about our cash spend and plan to continue to do so.
Okay, great. I want to see if there's any questions from the audience. If not, thank you very much, John.
Thank you, Li.