We're going to go ahead and get started on time. Again, welcome to the second day of the Citizens Life Science Conference. It's my pleasure to introduce the next presenting company, which is Compass Therapeutics. Presenting for the company, or having our discussion, is Dr. Tom Schuetz, CEO of Compass. Welcome, Tom. Really appreciate your time.
Thanks, Ren. Thanks for inviting us.
So, you know, people in the audience have been hearing me say this throughout the day, but I never know exactly who's in the audience and who knows the Compass story, or who's listening in on the webcast and knows the Compass story. I always like to start off our discussion with maybe a two to four-minute overview of what the company is about, if you can just kind of level set the stage for us.
Sure. Okay, thanks. Compass is located in Boston. We're a monoclonal antibody discovery and development company in oncology. We currently have three drugs in the clinic, and our fourth drug we're driving toward IND submission later this year. We have a mix of monoclonal antibodies and bispecific antibodies in development. Our lead program is a DLL4 VEGFA bispecific antibody that used to be called CTX-009. We actually have our first, very exciting, our first USAN name ever at the company. That drug is now called Tovecimig.
Tovecimig.
With that drug, and we'll talk more about it, about five weeks ago, we announced an analysis of the primary endpoint in a randomized study in patients with advanced biliary tract cancer, and that drug hit the primary endpoint of overall response rate in that study. So we're super excited about that. Our second program is a drug called 471. That is a monoclonal antibody agonist of CD137. It's a next-generation agonist. And we presented data at scientific meetings twice last year. We presented clinical data at ASCO, and we presented an analysis of biopsy specimens at SITC, and we can talk more about that. Our third drug is a really interesting drug that came out of a screening system that we developed at Compass. We have a screening tool that we call StitchMabs, which is a platform that allows us to screen for synergy in bispecific antibodies.
We asked an incredibly simple question, you know, what's the best combination partner for PD-1 blockade? Made a small library, did a screen, and much to our surprise, the answer turned out to be PD-L1 blockade. We have a PD-1, PD-L1 bispecific antibody that is in the midst of dose escalation phase one with clinical data expected later this year. Finally, our last program, obviously the intersection of the Venn diagram between angiogenesis disruption and immuno-oncology is a VEGF PD-1 bispecific. We have a next-generation molecule there that is more potent than drugs that are currently in development. It is a differentiated PD-1 VEGF bispecific. That drug is currently in IND-enabling studies. We announced this morning as part of our business update and our Q1 earnings that we had a successful pre-IND interaction with the FDA in the first quarter.
That drug is on track for IND submission in Q4 of this year.
Excellent. Great to hear it. Huge pipeline to discuss. Let's jump right into 009. Would love to understand a little bit more of the mechanism of action here, DLL4 and VEGFA, you know, why that combination came out of StitchMab, you know, that platform. I guess really just honing in on this data. This is pivotal data that was reported. Can you just give us the highlights?
Sure.
You know, that you reported.
Okay, two-part question there. DLL4 is delta-like ligand 4. That is the cell surface ligand for Notch 1. VEGFA, of course, is the well-known soluble ligand for the VEGF family of receptors. Those are two different angiogenesis signaling pathways. Disruption of those signaling pathways individually has different phenotypic effects on angiogenesis in the tumor microenvironment, and dual blockade has been shown to be important. Possibly synergistic. The way that we believe this drug works is that DLL4 anchors the bispecific in the tumor microenvironment where it can provide local capture of VEGFA. This is truly a next-generation angiogenesis disrupting agent. We completed phase one, and surprisingly in phase one, this drug had monotherapy responses. We had monotherapy responses in patients with gastric cancer and colorectal cancer, and we had an unequivocal dose response. You know, why do I say surprisingly?
Because in general, angiogenesis targeted agents as monotherapies have extremely low response rates. In a very heavily pretreated phase one population, to see monotherapy responses was pretty unexpected. Of course, the real therapeutic potential of angiogenesis disruption, obviously led by Avastin, was uncovered when Avastin was combined with chemotherapy. In general, for the most part, any clinical trial that has the design of chemotherapy versus chemotherapy plus Avastin, the vast majority of those studies have been positive. Avastin is labeled in multiple different tumors, and it really was the first-in-class molecule in my view. One of the things that we did, of course, then was we quickly did a small Phase I B study where we combined Tovecimig with chemotherapy. We picked either irinotecan or paclitaxel because in phase one A, we had responses in patients with colorectal cancer and gastric cancer.
Irinotecan and paclitaxel are part of chemotherapy regimens that are used in patients with colorectal cancer and gastric cancer respectively. We did a phase one B study, typical all-comers phase one B, and we saw something very unexpected, which is we had two patients, just to be fair, it was only two, with very advanced biliary tract cancer who had deep and durable responses when Tovecimig was combined with paclitaxel. That was the initial empirical observation. We then confirmed that observation in a Phase II study. That Phase II study was discussed with FDA and led to the design of a randomized trial. That trial is currently ongoing. That study is a two-to-one randomization of Tovecimig plus paclitaxel versus paclitaxel alone in patients with advanced biliary tract cancer who had received one prior line of therapy. It is a pure second-line study.
In the front-line setting, where standard of care is chemotherapy plus a checkpoint inhibitor, response rates are in the range of 25%-26%. Okay? In our second-line randomized trial, and we disclosed the top-line data from the study about five weeks ago, we had a 17.1% response rate in the Tovecimig plus paclitaxel arm versus 5.3% in the paclitaxel-only arm. We had more than triple the overall response rate, and that difference was statistically significant with P equals 0.031. The study hit the primary endpoint. We are now waiting for the PFS and OS analyses, which are going to occur later this year.
The market reaction was kind of mixed on this data. Like normally, you would see, you know, a positive P value and that's it. Like it's off to the races. What do you think has led to kind of this, call it confusion in the market? What, if anything, has the market gotten wrong?
Anything else going on in the world?
We haven't gotten to the regulatory question yet.
No, you're correct. You know, I think, you know, perhaps investors were hoping for a little higher response rate. When you think about it, you know, with front-line response rates in the 25% range, you know, we're one of the, we're the first randomized trial, you know, that has shown a statistically significant difference in response rates. We, you know, tripled response rates that are seen with other chemotherapy regimens in this disease. You know, the study is unequivocally positive. I think maybe people were hoping for a higher response rate that could lead us to maybe think about accelerated approval right now. You know, I think we have always said, you know, that we believe that we would require the full data set, you know, from the study.
You know, the other thing that emerged from the data that we disclosed is there's a very large difference in the number of patients with progressive disease. So 42.1% in the paclitaxel-only arm versus 16.2% in the Tovecimig plus paclitaxel arm. So we strongly believe that there's a chance that the progression-free survival analysis could be positive to very positive. You know, that's very exciting for us. The other thing that we disclosed was the overall survival analysis is taking longer than we had thought. This is extremely important. When we originally modeled the overall survival for the study, we assumed 6.2-month median OS in the control arm and 10.9-month median OS in the combination arm. Today, we have more than 14 months of median follow-up, and we're not close to where we thought we would be.
We recently announced that the OS analysis would not occur probably till the beginning of Q4, which would be 20 months of median follow-up. There is, it's quite likely that we're seeing a treatment effect just based on cumulative mortality. I think that's extremely important. You know, we're looking forward to those analyses later this year.
As we think about the endpoints you've reported, the ones that we're expecting, and then your comments that we want the totality, the evidence of the data as we file. I believe you've also publicly said you're going to be talking with the FDA, right, in regards to, you know, BLA submission. What more is there, what is there to discuss? Will you be thinking about accelerated approval strategies with just the data you have? You know, what can come out of this?
Sure. That's a little bit of a moving target right now, just to be fair. You know, I don't think that we'll be discussing accelerated approval with them, you know, imminently. I don't think we're going to do that. I don't think we have enough data in hand yet, you know, to be able to have that kind of a conversation. We are looking to interact with FDA, you know, in the very near future. Again, I said, I'm not trying to duck your question. It's a little bit of a moving target. You know, we're still working on what that might look like.
Got it. I know that from that call that we listened to that you used RECIST 1.1, there was some discussion about whether that requires mandatory, you know, confirmatory scans. Can you maybe just clarify for us, like, you know, the importance of, A, the confirmatory scans, but then also given that it is RECIST 1.1, the correlation between ORR and PFS, is it more or less correlated?
Sure, sure. So you're raising an important point. Obviously, you know, what you described is, of course, accurate. In RECIST 1.1, you know, let's just talk about the concept of a confirmatory scan. You have a scan that shows a response. Then you get a second scan in order to confirm, in air quotes, that the first scan was correct. That is only required in single-arm studies. The purpose of that is not to prove that a response is meaningful. The purpose, which is explicitly stated in RECIST 1.1, is to control for measurement error. That's it. In randomized trials, the randomization takes care of that, and a second scan is not required in a randomized study. RECIST 1.1 is unbelievably clear about this. What it does suggest is it's more important to have the scans reviewed by blinded independent central readers.
All of the scans that we had from the study were confirmed by blinded independent central review. It is about as rigorous as you can get.
Got it. As we think about other gating items, I mean, to me, like right off the bat, PFS and OS are the key clinical kind of gating items. Are there any other gating items for manufacturing or, you know, anything else that we should be aware of before a filing takes place?
Sure. You're correct, of course. You know, there's a lot that goes into a filing. We've been working very hard on the CMC side. We're doing, you know, sort of BLA-ready CMC work now. We're expecting, you know, a lot of that work to be completed toward the end of this year. I think we have our CMC and clinical timelines pretty well synced. Also, I'll just, you know, we've got a U.S.-based contract manufacturer, for whatever that means for the future. The manufacturing and the clinical timelines, I think, are pretty well aligned.
In two weeks, it's going to be good news. If we think about, you know, let's fast forward a year, right? It's been filed. You've gotten what I would imagine would be a priority review, I would assume.
Correct.
This is now on the market. What does this market look like? We always kind of hear different numbers, BTC, it's too small, this and that. Kind of, you know, how do you view the market? And are you commercializing it by yourself? Are you bringing a partner on board? How do you pull it off?
Okay. Let's talk about the market size first. In January, the U.S. SEER data, now the Surveillance Epidemiology and End Results data, were just published again, 23,000 patients with BTC estimated in the United States in 2025. 23,000. It's an incredibly significant market. Komodo Health, the claims aggregator, did an analysis of U.S. claims data, and they found that there were 23,000 patients in the United States per year in the 2021 and 2022 time period. They presented that data at the Cholangiocarcinoma Foundation meeting. The third data point on that is, and take this for what it's worth, but when AstraZeneca disclosed that Imfinzi was labeled in the front-line setting, they said the work that they have done identified 23,000 patients annually. That's a rock-solid number. How many patients get to second-line therapy? That requires some estimates.
Based on data from Topaz One and Keynote 966, which were the front-line studies of Imfinzi and Keytruda respectively, we've estimated that about two-thirds of patients will make it to second-line therapy. So about 15,000 patients annually. The fully penetrated market for this is a multi-billion dollar annual market. It's probably three-four times larger than the platinum resistant ovarian cancer market, something like that. In terms of commercializing ourselves, you know, we're going to, we have said publicly that we're, you know, having various strategic conversations. You know, obviously we'll do what's best for shareholders, but we're preparing to commercialize this drug ourselves.
Can you just remind us from a competitive landscape perspective? I thought there might have been a recent BTC approval, or maybe there's, you know, another drug in the pipeline. How are you looking at the competitive landscape and where, you know, this might fit in?
Yeah, this is super important. There is no competitive landscape in the U.S. There is no labeled second-line therapy in the U.S. Period. The only drugs that are available are targeted therapies. FGFR2 inhibitors, IDH1 inhibitors, and the drug that you're alluding to is a HER2 targeted agent. The total in biliary tract cancer of all three of those, in our view, is about 15%. The other 85% of the patient population has literally no labeled therapy in the U.S.
Okay. Perfect. As we think about, let's say, the rest of the pipeline now, I was going to jump right into kind of the PD-L1, PD-1 bispecific. During your prepared remarks, you mentioned your second molecule. Can you just take us through that and kind of where we are?
Sure. You know, we've got a next generation CD137 agonist antibody that was both epitope and affinity optimized for activating CD137. So many, many drugs in development targeting CD137, as you well know. That's an extremely important target. But you know, we developed a novel molecule. And last year, we presented data at ASCO. We've had five responses in the post-PD-1 patient population as a monotherapy. Okay? So I heard one of your questions at the end of the last talk, right, about monotherapy data, right? I think monotherapy data is critical. I really do. So we had three out of 11 patients with melanoma, again, post-checkpoint inhibitor patients for a 28% response rate. We had one out of four patients with mesothelioma, and we had one out of three patients with small cell lung cancer.
The small cell lung cancer patient actually had a complete response after having received two prior checkpoint inhibitor regimens. That was a really incredible clinical response that we saw, presented that at ASCO. We got on-study biopsies, and we discovered that the patients who responded generally had tumors that were positive for NCAM, which is CD56. Why is that important? CD56 positive NK cells are a major class of effector NK cells, and it interacts with itself. We believe we may have discovered a new IO phenotype in the tumor microenvironment. CD56 negative tumors cannot attract NK cells, whereas CD56 positive tumors can attract NK cells where they can be activated by 471. Presented all that data at SITC, now planning for an NCAM positive basket study, which will begin later this year.
Oh, terrific. That basket study starts, is there an easy way to detect NCAM positive tumors?
Yes. Yes. So that, and that can be done, you know, either locally or commercially.
Okay. And then data from that, from this study sometime in 2020.
Correct. Correct.
Okay. In the remaining time, we have switching gears to the PD-1, PD-L1 bispecific. What can investors expect from that program, you know, sometime this year or the next?
Sure. That is the first drug that came out of our StitchMabs platform, our screening platform. That's, again, it's a very simple question that we asked. You know, what's the best partner for PD-1 blockade, right? Because there are a whole ton of PD-1-based bispecifics that are in development today. We think we have potentially a first-in-class drug here because PD-L1 emerged as the best combination partner for PD-1 blockade. That drug, and we published all this data early last year, that drug's superior to PD-1 blockers or PD-L1 blockers in preclinical studies. It's between 100-1,000-fold more potent at activating effector T cells in vitro. It's a really well-differentiated drug. That drug is in the middle of dose escalation phase one. The patient population is post-PD-1 melanoma, non-small cell lung, head and neck, Hodgkin's lymphoma, and triple negative breast.
Those are the five indications we're testing. We're at the fourth dosing cohort now. We've seen no dose-limiting toxicities so far. We should finish enrollment into the fifth dose level probably in the coming month or two, which would put us in position to share the Phase I data set, hopefully at a scientific meeting in the second half of this year. That's a drug we're very excited about. You know, I don't, it remains to be defined what next generation checkpoint inhibition is, you know, after, you know, first line, you know, PD-1 blockade. I think we have the potential to have the leading next gen checkpoint inhibitor.
Let me ask you from a basic science perspective, why in a post-PD-1 checkpoint setting would now dual inhibition of, you know, PD-1 and PD-L1 actually work as opposed to going, I do not know, call it into naive tumors, but it would not work in cold tumors either, but into a more, you know, typical checkpoint responsive tumor?
This drug has a couple of different mechanisms of action. First of all, it is unequivocally a cell engager. It provides that advantage right off the bat. Secondly, something that's really fascinating is the bispecific actually induces cleavage of PD-1 off the surface of effector T cells. T cells that are exhausted and potentially resistant to PD-1 blockade, this drug eliminates PD-1 from the surface of the cells, rendering those cells now more effective at killing the tumor. It is a really, really unique mechanism of action that I think, you know, could provide some basis for testing patients in the post-PD-1 setting.
In the last sort of 30 seconds that we have, your newest molecule is in a space now where I'd say the vast majority of other compounds are from China, and this is the PD-1 VEGF bispecifics. The data to date looks spectacular, right? We'll see how this summer turns out with the Summit data that everyone is expecting. Obviously if it works, it's a, you know, rising tide lifts all boats sort of situation.
Yes.
Just very quickly, what is your kind of differentiating aspect of your molecule versus, let's say, the Summit's Ivo molecule?
Yeah. Comparing our drug to that drug, we are about five times more potent at blocking PD-1. We have a more potent and more effective PD-1 blocking component in our bispecific. I think that's going to turn out to be quite important. You know, as you know, the doses of these drugs that are being tested are enormously high. I'm not quite sure I fully understand that, you know, 20mg-30 mg per kg. We really focused on developing, you know, a drug that first and foremost was an effective potent inhibitor of PD-1.
When do you think this will be in the, an IND will be filed and in the clinic?
We just had our pre-IND interaction with FDA in Q1. Manufacturing talks are underway, and we're driving toward IND submission in Q4.
In Q4. Okay. Excellent. Key drivers for investors to be looking out for for the remainder of 2025. You just reported your numbers today. What's the current cash position?
Yeah. Finished Q1 with about $113 million in cash, which is runway executing on all of these programs into 2027.
Okay. And then just key milestones. Key.
Key milestones this year for Tovecimig are our PFS and OS readout from our randomized biliary tract cancer study.
Third and fourth quarter or both fourth quarter?
Probably fourth quarter based on what we're seeing from the overall survival numbers. Probably early fourth quarter. Initiation of our basket study for 471, Phase one data readout for 8371, and our IND filing for the drug we're calling 10726, our PD-1 VEGFA bispecific.
Excellent. Tom, I want to thank you very much for your time.
Sure. Thank you.
Excellent.
Thanks so much.