All right. Good afternoon, everyone, and welcome to this next session of Guggenheim Oncology Conference. My name is Yige Guo, and I'm part of the biotech research team here at Guggenheim. Our next presenting company is Surface Oncology, and with us today on behalf of the company is Rob Ross, Chief Executive Officer. Thanks for joining us, Rob. Yige, it's great to be here. Thanks for having us. All right. We'll start with the lead program of the company, which is SRF388. It's monoclonal antibody targeting IL-27. What makes IL-27 an interesting cancer target? It's a great question. At Surface, we are a full-on next generation immuno-oncology company, right? Very focused on what the next great targets are in immuno-oncology to increase the anti-cancer immune response. We have a very strong Scientific Advisory Board that really from the beginning has worked with us on what we think the new best targets are. IL-27 happened to be one of the very first targets that as a company we started looking at in 2014. It's, if you'll forgive the expression, it's a little bit of an unusual food group, right? IL-27 is a cytokine. Often when we think about cytokines in oncology, we think about cytokines that stimulate the immune system, but cytokines can have all sorts of effects on the immune system. IL-27 in particular is an immune suppressive cytokine. It prevents the immune system from having an overwhelming response when you get infected with a bacteria or a virus or a parasite. Our scientists, along with members of our SAB, noted that interestingly enough, in some tumors, not all tumors, but in some tumors, there was a high concentration of IL-27 and evidence that IL-27 was having an immune suppressive effect. That can't be good, right? You want an immune system stimulated to fight the tumor. That led us on a chase, right, that ended up with SRF388. What SRF388 does is it basically binds the cytokine, right, and neutralizes it. Prevents it from having the downstream effect on T cells in particular that makes them less inflammatory. Importantly, we don't think this will work in every tumor type, but in tumor types where you see higher levels of IL-27, we think it's an important part of the suppressive local microenvironment of a tumor, and if we can get rid of it, then we can increase anti-tumor immunity. That's what led us to particularly look at tumors like lung cancer and liver cancer, where you see high levels of IL-27. Great. That actually answers my next question. I guess, you obviously have a ongoing phase I multi-cohort evaluating SRF388. Yeah. Can you, I guess, start by reminding us the design of the [trial] before we dive into, you know, the key findings? Yeah. We've been, so we started, so we're the first people ever to have an anti-IL-27 program in the clinic in any disease, which is always incredibly exciting. We've been in the clinic now for over two years. We have progressed through the phase I as a monotherapy. We've reached a dose that we want to move forward into phase II and have now opened phase II expansions as a monotherapy and in combination in lung cancer, and also an additional phase II in combination in first-line liver cancer. We were very gratified to see that in the dose escalation of the phase I, we were able to identify monotherapy activity, in particular, a patient with non-small cell lung cancer who was refractory to multiple prior treatments, including anti-PD-1, who had a nice, sustained partial response in treatment with just SRF388 by itself. We have subsequently shown and reported monotherapy activity in kidney cancer and another patient with lung cancer who was treated with SRF388 alone. This patient was also refractory to anti-PD-1 treatment and also had multiple prior treatments with chemotherapy who had a nice confirmed partial response to SRF388 alone. Right now we're moving forward in monotherapy in lung cancer, in combination in lung cancer, and also in first-line liver cancer in combination. Great. That's super interesting. I mean, late stage, heavily pretreated patients, you've seen single drug activity. Yeah. Super interesting. Okay. 100%. Let's start with non-small cell lung. Yeah. You mentioned you've seen responses. Can you just, dive a little bit deeper, you know, talking about those responders and, was there a correlation with, you know, potential PD-L1 or any biomarkers that could explain the response? Yeah. Yige, it's a really good and important question. As I mentioned, we've seen two responses, both confirmed in non-small cell lung cancer. The patients had several similarities, right? Both patients were refractory to anti-PD-1 treatment, they'd been treated with anti-PD-1 and progressed during or very soon after that treatment. They both had multiple rounds of chemotherapy, both patients had squamous cell histology. When we think about non-small cell lung cancer, we think about two primary histologies, adenocarcinoma and squamous cell histology. Both of these patients had squamous cell histology. Interestingly, both patients were PD-L1 low. We think about PD-L1 as a marker for the inflammatory state of the tumor. We think about it as a potentially a marker for being more likely to respond to anti-PD-1. What all of that tells us in a nutshell is that these were both highly refractory patients who are very difficult to treat. The fact that we saw monotherapy activity was super interesting. You also mentioned pharmacodynamic effects. I want to think about two buckets. One bucket is the drug hitting the target? Do we have evidence that the drug is doing what it's supposed to be doing, what it says on the label, right? The answer to that is yes, and we've shared some of those data already, that we've shown that when we treat patients with the drug, we get complete blockage of the pathway. We can detect that in whole blood and we also see activation of NK cells, which we predicted non-clinically, and we see elevations of interferon gamma systemically, which we also see pre-clinically. We think we have very good data that show that when we give the drug at this dose, it is doing what our scientist tells it's supposed to do. The second critical question around PD is can we select patients to treat based on what we know about their tumor before enrollment, right? We are hard at work at that question now. We have developed an assay to measure levels of IL-27 within the tumor. We haven't shared any of those data yet from our patients that we treated in the clinic, but we've shared a bunch of those data from historic samples showing that non-small cell lung cancer in particular and squamous non-small cell lung cancer in particular has high levels of IL-27. One of the things we hope to show in the middle of the year when we present more data publicly is whether or not patients, the patients on our trial who had higher levels of IL-27 in their tumor were more likely to respond. If that's the case, if we can draw a line between those two or three dots, then we have, my grandfather would say we're cooking with hot grease, right? We'd be in very good shape. Great. Great. It's just so interesting that the two responders happen to be squamous. How many squamous patients do you have in the trial? And could squamous be a sort of like a, you know, predictive biomarker to reach responders? Yeah. I love where you're going. Right now, we've only treated, as of the last time we publicly reported, we treated only two patients with squamous non-small cell lung cancer, and they both responded. It's a 100% response rate and we're in great shape. No. Clearly we need to treat more of those patients. What's interesting though is that we have studied historic samples from patients with squamous and adenocarcinoma and have seen that in both tumor types, particularly squamous, there's a very high level of IL-27 within the tumor, okay? Something like 85% of the upwards of 100 or so patient samples we've looked at with squamous non-small cell lung cancer have elevated levels of IL-27. Moreover, when we've looked across other tumor types, the other tumor type that scores the highest when it comes to IL-27 levels is head and neck cancer. Head and neck cancer, as I'm sure you know, is almost all squamous histology. We think that's a really interesting link. Clearly, there's more work to do, right? We're feverishly working at that now. Great. That's interesting. All right. You also mentioned obviously you've reached a phase II dose. Just, talk about how, you know, you select the RP2D. You know, what is the rationale there? Yeah. I mentioned before some of the PD markers we looked at. In general, you look at three buckets of things when you select a recommended phase 2 dose. You look at safety, and that's probably most important. You look at efficacy and you look at PD markers. From a safety perspective, the drug's been very well tolerated. We have not identified any dose-limiting toxicity, which is honestly what we predicted pre-clinically based on sort of how IL-27 works in normal folks who don't have cancer. That's what we've seen in the clinic, so that's great. From an efficacy perspective, all the responses we've seen have been at this dose level, so that tells us something, right? That tells us that we're in pretty good shape. From the PD standpoint and from modeling based on our non-clinical data, we believe that at doses of even less than 10 mgs per kilo, we've completely saturated the pathway. We have dosed up to 20 mgs per kilo, and from a safety perspective, that was fine, but you are so supersaturated at that point that you're basically wasting a lot of drug. We are right now treating 10 mgs per kilo dosed once every four weeks or once every three weeks. Got it. All right. Let's talk about your combo strategy. You mentioned you have a cohort evaluating pembro combo. Yep. I guess, you know, what is your development plan there, and what level of efficacy do you need to see in order to move forward? Yeah. Non-clinically, we have a lot of data looking at the combination with anti-PD-1, particularly in on human T cells. We've been able to show non-clinically that if I activate T cells with anti-PD-1 but there's IL-27 around, you don't get the inflammatory response you'd expect. You don't get that inflammatory release of cytokines that you would expect from anti-PD-1. If you neutralize that IL-27, then lo and behold, you recreate that inflammatory response. We're very interested in combining with anti-PD-1. We're doing that in two different ways in the clinic. In lung cancer, we're combining with anti-PD-1 in the sort of relapse refractory setting. These are in patients who've seen anti-PD-1 before in the first line setting. We are re-exposing them to that as part of our collaboration with Merck, who's supplying us, supplying us the drug, and then studying that combination. A response rate of 25% or greater in that population would be incredibly exciting to move that forward into a registration trial. Got it. We are also looking at the combination with anti-PD-L1 in the first line setting in liver cancer. There we're looking at the combination with atezolizumab and the anti-PD-L1 from Genentech as well as bevacizumab, the anti-VEGF antibody, as part of our collaboration with Roche and Genentech. Got it. That makes sense. For the lung cohort, are you focusing on PD-L1 positive? Yeah, so it's interesting. We've gone around and around about that. The answer right now is no, right? We're looking at, we're looking at basically all comers, as long as they've been previously exposed to anti-PD-1. One of the reasons for that is that in the two patients who responded before, one of them was PD-L1 low, and one of them was PD-L1 undetectable. Given that, we thought it would make sense to open it more broadly, particularly because that's such a difficult patient population to treat. If there's a way forward there, that would be incredibly exciting both for patients and for our sort of regulatory strategy. That makes a lot of sense. Given that, you know, you've seen single drug activity, you know, especially in PD-L1 low, do you think there's a possibility for, you know, there's a registration, sort of strategy or opportunity as a monotherapy for SRF388? Yeah. We need to get more data and see. What I'd say is I think in relapse refractory non-small cell lung cancer, the medical need is very high. If you can show a reproducible response rate of, say, 30% or greater, then that's a very reasonable conversation to have with regulatory agencies. You don't have to get there in all comers, right? If in squamous, and that's the population going after, you see a response rate of 30%-35%, that's an exciting conversation to have. Moreover, if you can select patients based on IL-27 expression with that same response rate, again, that would give you a path forward, we think. Great. That makes sense. Then you've guided a potential clinical update in the first half of this year. Yeah. Maybe give some color on, you know, the upcoming update. What shall investors expect from this dataset? As I said before, we have three ongoing cohorts. First-line liver combo in lung cancer in the relapse setting and monotherapy in lung cancer. We are really focused in the middle of the year update with the monotherapy lung cancer data. That's what you can expect. In terms of number of patients, that trial enrolls up to 40 patients, and we're gonna have as many as we possibly can. I can't give you an exact number, but hopefully as close to 40 as we can get, both safety, efficacy, and translational data to hopefully support a patient selection approach. For the pembro combo, we may have data in the middle of the year. We don't know yet. For the liver cancer data, that's probably in the second half of the year. Great. I guess, last question in lung before we switch over to liver. Post-IO non-small cell lung cancer is a relatively crowded space. Yep. You know, many other agents being investigated. Yep. TIGIT, small molecules. In the longer term, how do you envision, you know, the landscape, especially in the context of SRF388 as a potential player. It's one of these spaces that's in an interesting place in that there is a lot of competition, but to date yet there's not a lot of, sort of, there are not a lot of approvals. In fact, it still remains the case that the, sort of, the only approved second-line drug in lung cancer, is Taxotere, right? As a, sort of, as a single agent. I think what we get excited about with IL-27 is the fact that we've seen monotherapy activity with a very strong safety profile. That lends it to all sorts of possible combinations, whether it's combination with PD-1, potentially combination with some of the ADCs that are that are exciting now in second line or later lung, or potentially broader combinations with more than one agent. I do think that we're gonna pretty quickly go from monotherapy to doublets to triplets and beyond in second-line or later lung cancer, and I easily see IL-27 being an important component of that. Great. let's move into, you know, the liver cancer. I guess, can you start by reminding us the development strategy and, you know, upcoming data disclosure you just mentioned a little bit on the liver cohort. Liver cancer, we've had a lot of translational data support moving forward in liver cancer for quite some time. We wanted to be in a space that was PD-1 naive, that led us to first-line liver cancer. We designed a trial that has a 30-patient lead-in looking at atezo bevacizumab along with SRF388. If the level of activity there is interesting, then it immediately becomes a randomized phase II, in which we randomize approximately 100 patients to either atezo-bev plus placebo or atezo-bev plus SRF388. We'll make a decision about kicking off that randomized trial at some point this year. Okay. What would you consider as sort of like interesting signal in that portion? Lots of different ways you can slice this, but atezo-bev alone has a response rate of a little less than 30%. Response rates that approach 40% I think are quite interesting, particularly if there is some correlation with a patient selection biomarker. Okay. Makes sense. All right. I'm gonna switch gears to SRF114. Your second clinical stage program which targets CCR8. Yep. Can you first talk about the CCR biology and the mechanistic rationale for it to be a cancer immunotherapy target? Yeah. This is a great space, and there is a huge amount of excitement recently, as I know, I know you guys have been following very closely. We have been interested Within the field of immuno-oncology, we've been interested in getting rid of T regulatory cells that are within the tumor for a long time. It is well understood that T regulatory cells create local immune suppression. It's also well understood that some tumors have a lot of Tregs within the tumor microenvironment. Those Tregs almost certainly contribute to local immune suppression. The problem with getting rid of Tregs is that you also the targets that you use to get rid of Tregs are also on normal T cells that we wanna fight the cancer, right? If I get rid of a Treg, but also get rid of a normal T cell, then, you know, I have, I haven't changed the seesaw in any real way, and in fact, it could have made things worse, right? The other thing is, if I get rid of all my Tregs right now, if I just get rid of all of them, right, then I would be horribly sick from autoimmunity, right? Tregs are really important to prevent my body from attacking my colon and to prevent my body from attacking my skin, right? You need to find a way to deplete only Tregs and not normal T cells, and only Tregs in the tumor and not Tregs in other places in the body. This has been an active area of research for many, many years. One of the chair of our SAB or one of the chairs of our SAB, Sasha Rudensky, in his lab, was one of the scientists who discovered that a G protein coupled, a GPCR, okay, protein that's expressed on the cell surface of Tregs upregulated in the context of antigen sensitivity. It's upregulated in the tumor microenvironment, but it's not seen on circulating peripheral Tregs, and it's also not seen on normal cytotoxic T cells. This was a really interesting target to go after to deplete intratumoral Tregs, and that's CCR8. We and, you know, many others have developed basically depletion strategies. These are antibodies that are afucosylated and focused on depleting those Tregs that are expressing CCR8. That's what got us and many others interested, and we spent the last couple of years developing the best of those antibodies, which we think will be SRF114. Great. Yeah. I mean, totally makes sense, because of, you know, the growing body of knowledge on CCR8, I guess more and more companies started pursuing it as a therapeutic target. That leads to my next question. Can you talk about the landscape and why are you particularly excited about SRF114? Yeah. So, you know, to be honest, we were working on this target for a while. It took us a really long time to develop an antibody. We finally developed our antibody, and I'll tell you why it took a long time in a second. We were super excited 'cause we thought it was took so long and was so hard that we'd be the first. Then over the next six months, like five other companies announced they had one, right? We thought we were so close to being first, but we ended up not being first, and multiple folks entered the clinic. The leader, the first team in pole position right now is BMS. They entered the clinic about a year and a half before we did. We entered the clinic a few months ago. BMS, then AbbVie is in the clinic. Bayer is in the clinic. Gilead is in the clinic. Gilead in-licensed their program from Jounce and just recently bought all the rights from Jounce, right? There are also several smaller Chinese companies with antibodies that are in or close to the clinic. We're in that first group, in the first group of, say, five or six that are now clinical stage. Importantly, no one has presented clinical data on any of their programs, but BMS just recently updated their ClinicalTrials.gov listing, where their phase I/II went from approximately 150 patients to 650 patients. The number of indications they're going after went from, I believe, three to four to, like, 11. Clearly, they have decided the program is worthy of more investment because that's what they're doing, and we hope to see data soon. Another company that's in the clinic, excuse me, I shouldn't have forgotten them, is Shionogi. Shionogi actually did some of the original work in this space as well. Shionogi has also said that they are expecting to have regulatory interactions about an end of phase I meeting at some point this year. A fair amount of competition. Getting back to why we think our antibody could be best, why we're excited about our antibody, it gets back to this question of how do you develop a CCR-specific antibody? It took us a long time to find one. We found that when we identified antibodies that also cross-reacted with monkey CCR8, which we like to have from a non-clinical perspective because it helps with toxicology, that they then also bound other human proteins that were not CCR8. These were non-specific proteins. Really, we had to wait until we had an antibody that didn't cross-react with monkey CCR8 to identify a protein that was specific only for human CCR8. Does that make sense? When we did that, we then developed SRF114, which is really clean and only binds, as far as we can tell, only human CCR8. When we looked at many of these competitors, many of them report to bind monkey CCR8 as well. That made us think, "Well, maybe they're not human-specific," right? We then built those antibodies, not all of our competitors, but many of the competitors. We built those antibodies from the patent literature. We had to take a guess at which one was right, and, you know, we did our best we could to identify what we think is the clinical candidate. We noted that in the lead four competitors, they all bound another protein besides just human CCR8. They were binding some other cell surface or secreted protein. That is almost certainly not good. How bad it is sort of depends on what happens in the clinic. Our antibody, compared to those antibodies, was the only one that was pristinely specific just for human CCR8. That's what makes us think that we may have a best-in-class antibody. It's important as we read through the safety profiles of the competitors, some of the safety or toxicity they see may not be due to CCR8 binding at all, but due to the binding of one of these non-specific human proteins. Great. Now, you know, with the potentially differentiated CCR8, you recently announced initiation of the phase I trial. Yeah. I guess in the last, one minute or two, last question is if you can talk about the design of the study, and then when might we expect the, you know, data readout? Yeah. We haven't said much about the design other than what's on ClinicalTrials.gov. It is a pretty standard dose escalation, followed by an expansion right now. What we've let on is that we're gonna expand into head and neck cancer. There'll be other tumor types that we go after and other combinations, but we haven't been public about that yet. The dose escalation is ongoing. I expect us to have clinical data to share at some point next year. Great. All right. We're right on time. Thanks for joining us again, Rob. Yeah. It's been a pleasure doing it. Thank you so much. Thank you. Take care.