Hello everybody. Welcome to this last session of the Guggenheim Oncology Conference. My name is Paul Jeng. I'm a member of the biotech research team here at Guggenheim. Our next presenting company is Xilio Therapeutics. I'm really pleased to welcome the CEO of Xilio, René Russo. Rene, welcome.
Hi, everybody.
All right.
Thank you, Paul Jeng.
Yeah, I figured we could sort of start, excuse me, broadly on the platform. Maybe you can give an overview of your GPS platform and your approach towards sort of tumor-specific activation of immune modulating biologics, and then we can go from there into programs.
Sure. At a high level, our GPS platform, which stands for geographically precise solutions, really exists to get highly potent therapies at high concentrations into the tumor microenvironment and protect healthy tissue from the effects of those agents. It's sort of a way to give a systemic agent, but yet get intra-tumoral like injection activity locally at the site. We're doing this because we believe it's better for patients, and it allows us to access many therapies, in particular cytokines, which we're very interested in, that we know shrink tumors clinically, but are incredibly toxic and really unattainable as therapies at the doses with which you see responses, and certainly not able to combine these agents because of the toxicity. That's the essence of why this platform exists.
The way we design these molecules is through a few components. We have the active agent. We have a masking domain, so this is a protein engineer blocking domain that turns that active site off when it's in place. This is all held together by linker sequences, and a protease cleavage site. What we do is we utilize what's unique inside the tumor microenvironment, which is a high activity of matrix metalloproteinases, and we utilize that high activity to cleave that protease cleavage site and activate the molecules preferentially inside the tumor.
Got it. Great. Why have you chosen to prioritize masked cytokines?
When you think about it from a patient perspective, where you can really utilize this tumor selectivity is for agents that we know have anti-tumor activity, but are too toxic, essentially, to give at therapeutic doses. Where can we make a drug out of something that currently is not a drug? That's really what we're going after. Cytokines are a great example of this. Highly potent, we know shrink tumors clinically, incredibly toxic, and also, if you think about from an evolutionary perspective, cytokines evolved in the human body to act locally, right? When they're given systemically, they cause, you know, extensive damage. That's not really their intent. We think that's a perfect fit for the GPS platform.
Got it. Okay. Makes perfect sense. Moving on to the programs, starting with XTX202, which is your masked IL-2. Obviously, IL-2's been a very challenging space over the past several years to improve upon aldesleukin, despite all the different approaches that have been attempted. You know, there's been some, you know, program discontinuations across the landscape. Maybe just talk a little bit about what your learnings have been from those programs, and you know, what sort of, you know, you see for the future for this IL-2 landscape before going into a specific program.
Sure, sure. I think taking a step back, you know, why has everybody been trying to crack, you know, IL-2? That's because when you go back to the original aldesleukin data, for the patients who could tolerate this agent, when they had responses, they were long-term durable cures. That is something patients should have and should have access to. The challenge has been that the doses, when you go back and look at the aldesleukin literature that were associated with efficacy, were very high. It's high-dose IL-2 that really led to those deep durable responses, 0.5-1 milligram per kilogram of aldesleukin.
Since that time, I think we've attempted in a number of ways to try to engineer IL-2 to work on, there were different theories around reducing the toxicity through engineering. Ultimately, I think what we've seen is that the systemic IL-2 agents, no matter what the engineering has been, have been really sort of stuck at very, very low doses, right? Nothing like the high-dose Aldesleukin that we saw, you know, where we saw those, um, long-term durable cures. Our approach really that is watching sort of the evolution is that we're bumping up against toxicity. The therapeutic index is still a problem. You can't really solve it systemically.
Our goal is to get back to very high doses, around 1 mg per kg of our XTX202 agent, get it into the tumor microenvironment, and then get those concentrations that were associated with long-term durable cures.
Got it. Your IL-2 is both masked as well as a non-alpha IL-2.
Yes.
Can you talk about the non-alpha component and why that's really important for your efficacy and safety?
Yes. I think the most important component is the tumor selectivity, right? Getting high concentrations in the tumor. Again, that's what we know works. That's what's worked originally with aldesleukin. We chose a beta-gamma design because what happens with wild type IL-2 or native IL-2 is that there is a Treg sink, essentially, CD25 binding that has a very high affinity for the IL-2. So when you give a low dose, it really goes essentially into that Treg sink. To avoid that and to avoid having to overcome that with an even higher dose, we essentially engineered out the alpha binding, so we have a beta-gamma molecule that does not have that Treg stimulation. It goes right to CD8 and NK effector.
Great. The program is currently in a clinical study. Sort of based on updates across the landscape, it really seems that finding the right dose level.
Yes
Dose sort of interval is really important for these next gen IL-2s.
Yes.
Maybe you can update us on sort of, you know, where you've escalated to?
Sure
... the progress in the study, and how you're thinking about enrollment of expansion cohorts at this stage.
Yep. Yeah, we're very pleased with the progress on the program. We are through dose level 4, so that is through the 1 milligram per kilogram dose, which is our original target dose. We are continuing to dose escalate. Our goal is to have an every 3 week dosing interval. We believe that the molecule is behaving in support of an every 3 week dosing interval, which I think is important for patients, and the PK, you know, that is necessary for that convenience. That's where we are now. We are right now opening up the Phase 1B. That's the Phase 1A dose escalation. We are opening up the Phase 1B, it is open, which is the biopsy, mandatory biopsy cohort.
We, because of some of the data we've seen, which we can talk about, in our first patient tumor samples, we're seeing high levels of CD8 increases in the tumor without CD8 increases in the periphery, and that has caused us to go ahead and get ready to open the phase 2, which we'll do the first half of this year.
Got it. Maybe it'd be a good time to just provide some more color on, in any of the clinical data that you've talked about so far.
Yes
... including that biopsy data that you mentioned.
Yes. I think two aspects to IL-2, right? First, on the safety tolerability side, while we haven't shared detailed safety data yet, what I can share is that at the 1 milligram per kilogram dose, we're completely in the outpatient setting. We are not supporting these patients with any other fluids. There is no sign of vascular leak syndrome, which has historically been the major toxicity. We think this is unique and this is differentiated as a profile. In addition to seeing that, we were very fortunate that 2 patients in the Phase 1A gave us biopsy samples.
They signed up for that optional biopsy, and these were at low doses, and we were very pleased with the data, much more than we expected, because these are heavily pretreated, multiple prior IO patients. The biopsies, the tumors have very low TILs to begin with, so it's a very sort of coldish tumor setting. The first patient sample we received had a dose of 0.38 mg/kg, and this was a stage 4 melanoma patient. What we saw between the pretreatment and the post-treatment tumor biopsy was a 3-fold increase in CD8s in the tumor and no change in CD8s in the periphery. It was the first time we demonstrated tumor selective immune activation with the molecule. We're very excited about that. We got another sample.
This was in a renal cell carcinoma, also, you know, multiple pretreatments, late stage disease. This patient was at the 0.5 mg per kg dose, 0.58, and we saw a sevenfold increase in CD8s in the tumor, and nothing in the periphery. We thought very significant. To put that in a little bit of context, we are aiming for around a fourfold increase in CD8s in the tumor microenvironment, and that is based on some recent data that has come out of Memorial Sloan Kettering in December of 2022, where they actually injected IL-2 into melanoma lesions in patients, Each patient was their own internal control, so they had injected lesion and non-injected lesion.
They found that a fourfold increase in CD8s was correlated with the responses in the lesions that responded. That's some of the data we're using, as well as our own preclinical data, to support, you know, a target of fourfold increase.
Okay. Great.
We have platform validation now, I think, in these first two patients, which is very exciting.
Sure. You're heading towards a pretty important readout in the third quarter of this year, which will include Part One data in solid tumors, and I believe some Part Two dose expansion data.
Yes
Maybe you could help set some expectations for that readout later this year, you know, maybe how many patients or dose levels we could see, you know, what kind of efficacy data we could expect, and anything else in that readout?
Sure. Sure. This readout in Q3 is certainly not the final phase two readout. It's an early readout. What we'll be doing is collecting a basket of patients from the phase 1A, 1B, and any of the phase two patients that are available that we've enrolled from the time we've opened it until that Q3 mark, and anyone who's been treated at one milligram per kilogram or higher, and we're gonna put those together in a basket. We have our own sort of internal bar that in this, you know, post PD-1, all-comer, mixed solid tumor population, we'd like to see about a 10% response rate. We wanna see monotherapy activity with this molecule before continuing on into the longer term phase two.
Got it. Then in the melanoma and RCC patients specifically, is the bar just the aldesleukin data, you know, pretty straightforward, or is there something else that you're trying to look for?
I think in this particular slice of data, there will be some melanoma, there will be some renal, but it is.
Sure
... tumor population, and that's sort of why, we think 10% is the right number. It's higher than, other IL-2s have seen in this setting. Ultimately, when we go into the phase 2 and we're looking at powered, you know, kind of controlled, homogeneous patient populations in RCC and melanoma, we will look for higher responses.
Great. Looking forward to the part 2C PD-1 combo cohort, you know, what are some of the gating factors to initiating that portion of the study, and advancing, or for advancing it later on into a registration enabling study?
Yeah. Yeah. Our philosophy has been, and I think we've all learned this with IL-2s, we need to see monotherapy activity before we move into PD-1 combo. What we don't wanna do is jump into combinations and be unclear on what this molecule really contributing, and is it a standalone drug that's meaningful for patients? We wanna see that monotherapy activity first. If we see it and we feel convinced that this is, you know, that the monotherapy activity is proven, then we will move, you know, in a very disciplined way into the PD-1 combo. From there, potentially into other therapeutic areas like in non-small cell lung cancer in combination with PD-1.
All right. Great. You have another very interesting cytokine program that I'd like to talk about, the XTX301, which is an IL-12 mask cytokine. This is also a space where there's been some, you know, historical experience, but also some severe toxicities and some safety issues. You know, similar type of question, what have you learned from past historical studies?
Yeah
... with IL-12, that were sort of applied to the molecule design?
Yeah. Yeah. We're fortunate because our head of R&D, president and head of R&D, worked on one of the original IL-12 programs at Roche. It was given sub-Q. This is back in the '90s. He's quite familiar with the activity. It did shrink tumors clinically, but also the toxicity, which is, you know, really can be a fulminant hepatotoxicity. I think for an IL-12, you really wanna be very cautious about the masking. You wanna make sure you have this molecule under tight control. It's incredibly potent, and it has the ability really to, you know, really stimulate innate and adaptive immunity. Very potent.
What we've learned along the way, for all of our molecules is that preclinically, you have to establish a very clear therapeutic index, and that means that your HNSTD in non-human primates has to be significantly higher than the concentrations you needed for efficacy in mouse models. You know, that's not often the case with IL-12s, right? It's very hard to get to that without tumor selectivity or an intratumoral approach like we've seen some people do. For us, it's really about getting that therapeutic index right with the molecule design, testing multiple molecule constructs in non-human primates, really understanding the molecule before you move forward.
As an example, for us, the HNSTD was around 2 milligrams per kilogram in NHPs, and then the dose needed for activity in mice around 0.04 milligrams per kilogram. We think that is what we've learned. You gotta establish that first, and until you have that molecule, you're not ready.
Got it. Okay. One of the unique characteristics of IL-12 that we've heard discussed a lot is its ability to sort of remodel the tumor microenvironment and maybe turn some cold tumors hot.
Yes.
you know, what, you know, how does that factor into your thinking about?
Sure
... you know, tumor types in the study? You know, how persistent is that remodeling? Is it something that, you know, you'll see for a period of time after discontinuing dosing, or do you have to continue dosing to get that effect?
Yeah. Yeah. Yeah. one of the things we're most interested in, I think, and compelled by is that there are so many patients that are not responding to IO, right? There's so many cold tumors, yet, this should be available to all patients, right, with all different tumor types. How do we get IO, you know, to work in cold tumors? That's sort of what the holy grail of IL-12, right, has been, where we're trying to, you know, get that initial activation of the immune system.
For example, in some of those patients we talked about from our IL-2 study that had, you know, a very low TILs, 5% TILs, a very cold type tumor environment, you stimulate that with an IL-12, then you add on an IL-2 or PD-1, you know, you've really improved the immune sort of profile of that tumor so that it can then respond, you know, to that IO therapy. We're very excited about that and the combination potentially of our IL-2 and our IL-12 together, which would otherwise, you know, not be feasible.
Great. Yeah, maybe expand a little bit about that potential combination. You know, what are the sort of mechanisms, rationale of, you know, underpinning that idea? You know, are there any safety concerns with these two very potent cytokines?
Yeah. Yep. Yep. I think I probably didn't answer your question about tumor types either, so I can go back to that. So, mechanistically, it really is that innate and adaptive immune stimulation that we think IL-12... You know, we've seen it. When you inject it, you know, intratumorally, you can really stimulate the immune system in an otherwise cold tumor. I don't know the answer to whether you have to continue to do that over a long period of time, or that once you do it, and then you add an agent like an IL-2, that that, you know, that process then continues sort of on its own. That's a good question. I don't know the answer to that. I think we'll hopefully try to figure that out.
In terms of, like, the tumor types we would think about, you know, we would want to, with our IL-12, simultaneously look at traditionally IO-sensitive tumors, but in parallel, look at some cold tumor types early on. You know, there would be a number of different tumor types. Could be pancreas, could be ovarian, MSS colorectal. You know, we do wanna study those patients in parallel with hot tumors. You asked about, you know, how do you combine these very toxic agents. You know, I think for us, getting our IL-2 to a dose of 1 mg per kg without any signs of VLS, again, in the outpatient setting, we're continuing to dose escalate. I think we feel confident that that is an agent that can be combined with an IL-12.
We'll need to see how our IL-12 behaves. Can we get it to a safe dose that we think is reasonable? Then start that combination study. You know, we'll start with low doses and combine, and move up slowly and carefully. We think with this mechanism of masking, it's quite possible.
Okay. On the phase 1, maybe provide an update on the initiation progress, and expectations for initial data that could come...
Okay
I think end of this year.
Yes. The IL-12 study is ready to enroll. We will be enrolling the first patient this quarter. We're starting at a dose of 5 micrograms per kilogram, which is about 10-fold higher than an IL-12, you know, has been given in the past. We're gonna be careful about escalating. We expect that by the end of the year, we'll be through dose level 3, which we expect to be around 50 micrograms per kilogram. That data, if we get through those dose levels as we plan, we will present all of the safety data and PK data we have on those 3 dose levels by the end of the year.
Okay. Great. Yeah, maybe a topical question. There's been some, you know, recent competitive landscape reshuffling of the pipeline for IL-12. You know, obviously, some different modalities there.
Yes
... yours. Maybe just review on that and whether there's any read-through to your program.
Yep. Yeah. There's always a lot of turbulence in the cytokine landscape the last few years. I think what we take away, you know, from where we sit and what we're seeing on some of the movement in the various IL-12 programs is the systemic IL-12s are sort of falling by the wayside. You know, potentially for strategic reasons in companies, potentially, you know, due to the profile, it's unclear. We have seen in the example of AZ that they are continuing their intratumoral program. I think there is a movement toward understanding that, you know, this is such a potent molecule, can you have a systemic version that really, you know, gets you enough concentration and efficacy without severe toxicity?
I think that's really difficult to do without some type of intratumoral or tumor-selective approach.
Okay. Great. All right, let's move on to your non-cytokine program, which is XTX101, your masked CTLA-4. The phase 1 study is ongoing. You know, can you tell us about sort of, you know, the dose level that you're currently at, you know, how, and how that, you know, corresponds to what ipilimumab has shown in those studies, you know, how you decide on the dosing interval?
Yeah. Yeah. So we did complete the Phase 1A dose escalation. We've enrolled the final dose cohort, and now we're just following those patients through. We'll present this final Phase 1A data package in Q2 of this year. That'll be safety, PK, PD, and any antitumor data we have for the Phase 1A. We did land in the last dose cohort at 150 milligrams Q6, so every six weeks. This is sort of stemming from, you know, I think an emerging view of what I would call newer generation anti-CTLA-4s that have Fc-enhanced design because the concept here is that you're really driving Treg depletion. These sort of have almost a different mechanism than the original ipilimumab type CTLA-4 with that Fc-enhancement.
In order to really maximize that Treg depletion, what we've learned and we've seen from data is that a high dose, higher dose less frequently, does that very well. We've learned that, and we've also learned that from AZ's recent data. We've been sort of following this, and of course, ipilimumab is in every six-week interval as well. That's where we landed, and we're happy to see that the PK supported going to that longer dose interval and still getting us the exposures we think are necessary for strong activity.
Got it. Okay. Ahead of the, you know, the data this year, how are you thinking about sort of what would be encouraging data from a safety, PK, or even efficacy perspective?
For CTLA-4, you know, what we're looking for is convenience of dosing, strong PK. We'd like to see a great safety profile, right? That's with that Fc-enhanced highly potent molecule. Our anti-CTLA-4 has a potency about tenfold ipilimumab and has that Fc enhancement, so incredibly potent. If we can get a reasonable safety profile with that level of potency, delivered with consistent PK, we think that's very attractive. Ultimately, this will be developed, I think, in combination with a PD-1. That's where that real benefit will be. We will certainly look for and share antitumor activity. That is in the plan, and that will be also part of the Q2 readout.
I think if there is early activity, I think that would be, you know, really nice to see for CTLA-4 in this population.
Okay.
Right? Again, it's post PD-1-
Yep
all come or all solid tumor, mixed population, and not many at, you know, the high doses, so.
Okay. You've disclosed, you know, only planning to advance the program if you find a partner.
That's right.
you know, has there been sort of any headway in discussions, you know, will that be sort of after the data's-?
Yeah
... disclosed? you know, is this gonna be potentially out-licensing or just a collaboration for development?
Yep. We are in active discussions with a number of partners on, you know, on this molecule. Certainly, seeing the entire final data package and how these patient, you know, the final dose cohort wraps up is important, I think, for those to move forward. The types of partners I think that are most interested have a PD-1, believe in this the T reg depletion mechanism and see a tumor type where they think this combination's gonna be helpful to them. I think having that complete data package will be helpful, for those to move forward. You know, we would be open to licensing or a collaboration. If, you know, if we could, we would develop it.
We would take it through phase 2, probably an MSS colorectal, given some of the compelling data with other Fc-enhanced CTLA-4s there. It's a time to prioritize. We're prioritizing our cytokines, right now.
Great. All right. I guess with that, maybe we can just wrap up with, reminding us of your cash runway and maybe sort of the news flow that you would expect over the next, several quarters.
Sure. Sure. Yes. Our current cash runway is into the second quarter of 2024. Our major milestones, I think, coming up for this year, you know, are throughout the year, but I think Q3 being an important one for that IL-2, you know, antitumor activity. We have cash runway to get us well through these data readouts and, you know, into Q2 of 2024.
All right. Fantastic. I think, with that, we'll wrap it up here. thanks so much to René-
Thank you.
... for joining us, and thanks for the audience for tuning in.
Thank you.