Thanks, Sherry, and thanks to JPMorgan for inviting us to present at the meeting this year. Just very briefly, I will be making forward-looking statements today, and I would refer you to our regulatory filings if you have any questions about those. Compass Therapeutics, we're based in Boston, Mass. We're a monoclonal antibody discovery and development company in oncology. We have three drugs in the clinic, and I'm going to talk about each of these drugs today because we've developed very important efficacy signals for each one of our drugs that's in the clinic, and this quarter, our fourth drug is going to enter the clinic, and I'll talk briefly about that drug as well. This slide is a snapshot of our pipeline. Our most advanced program is tovecimig. Tovecimig is a DLL4/VEGF-A bispecific antibody. That drug is a next-generation angiogenesis inhibitor.
Recently, we announced that we hit the primary endpoint in a randomized study of tovecimig in patients with advanced biliary tract cancer, and I'll talk more about that study in detail. Last week at ASCO GI, we presented data from a Phase II study in patients with very advanced colorectal cancer, and we have unequivocal monotherapy in patients with colorectal cancer treated in the third- and fourth-line setting. Our next most advanced program is a drug we call CTX-471. That's a next-generation CD137 agonist antibody, and I'll talk about the efficacy data for that drug as well. In the past six months or so, one of the most important developments at the company is with our third program, CTX-8371. That is a novel PD-1/PD-L1 bispecific antibody.
I'll go through the discovery and development of that drug, but most importantly, we now have responses in the post-checkpoint inhibitor patient population, and that's extremely important, and I'm going to spend some time discussing that program in detail. Finally, our fourth program, a PD-1/ VEGF-A bispecific antibody. That drug will go into the clinic later this quarter. Our management team, some folks in the room here with me and others, importantly, last week we announced some additions to our management team: Cyndi Sirard, our Chief Medical Officer, and Arjun Prasad, our Chief Commercial Officer. Arjun comes to us from Servier and Agios, where he participated in the launch of a drug in patients with biliary tract cancer treated in the second-line setting. So, obviously, very on-point experience, and I'll talk more about that in a couple of minutes. Okay, let's talk about tovecimig.
VEGF-A, of course, the well-known soluble ligand for the VEGF family of receptors. DLL4 is delta-like ligand 4. That is the cell surface ligand for Notch-1. These two signaling pathways have different phenotypic effects on angiogenesis in the tumor microenvironment, and dual blockade has been shown to be synergistic. Interestingly, and what could turn out to be ultimately quite important, is that DLL4 upregulation in the tumor microenvironment has been shown to mediate resistance to VEGF-targeted therapies such as bevacizumab. The reason that's important is because we have shown monotherapy activity with this drug in patients with colorectal cancer and gastric cancer who were previously treated with VEGF-targeted agents such as Avastin or Cyramza in colorectal cancer and gastric cancer, respectively. Our Phase I program was completed, and then a small Phase Ib study was done in which tovecimig was combined with either irinotecan or paclitaxel.
Those two drugs were chosen because paclitaxel is part of the standard of care in patients with gastric cancer. Second-line therapy for patients with gastric cancer is Cyramza plus paclitaxel. Irinotecan, of course, is part of the well-known FOLFIRI regimen that is used to treat patients with colorectal cancer, and in this study, there was quite an unusual observation. This is the waterfall plot for all patients treated in the study. On the far right-hand side of this slide, there are two bars, and those are two patients with advanced biliary tract cancer who had very deep and very durable responses to therapy, and those two patients happened to receive tovecimig in combination with paclitaxel. That empiric observation was then confirmed in a small Phase II study that was disease-specific for patients with advanced biliary tract cancer.
This is the waterfall from the Phase II study, and you can see there is a clear treatment effect here with multiple responses in patients with advanced biliary tract cancer treated in the second- and third-line setting. When we shared these data with the FDA, discussions with FDA led to the design of this study, which is currently ongoing in the United States. This study is called COMPANION-002. This is a randomized trial in a two-to-one fashion of tovecimig plus paclitaxel versus paclitaxel alone in patients with advanced biliary tract cancer who have received one prior line of therapy. So this is a pure second-line study. As I mentioned earlier, we have read out the primary endpoint of this study. Tovecimig significantly improved overall response rate in this study as compared with tovecimig. I'll go through those data in some detail.
The key secondary endpoints of progression-free survival, overall survival, are expected to read out later this quarter. I'll talk more about that timing in a minute. So on this slide is a summary of the overall response rate data. So on the top line here, tovecimig more than tripled the overall response rate compared with control, and that was statistically significant. We had a centrally adjudicated complete response in this disease, which is extraordinarily unusual. I should mention that all of these data are responses as assessed by blinded independent central review. So this is an extremely rigorous assessment of response. I'm just going to go forward one slide and then come back to this slide. Here is the waterfall plot from the randomized trial, and you can see clearly that there is a difference between the two treatment arms.
Patients in the control arm on top, the majority of these patients have increases in their linear tumor burden as evidenced by bars above the line. Below the line is good. On the bottom, you can see the vast majority of patients treated with tovecimig have decreases in their linear tumor burden. Let's go back to the previous slide for one minute. There's also something extremely important that came out of the best overall response rate data. In the red box on the bottom, second from the bottom, in the progressive disease line, there's a very significant difference in the incidence of progressive disease between the two arms. Now, because this is best overall response rate, this could occur at any time during the time that a patient is treated on study, except for progressive disease.
If progressive disease is your best overall response, by definition, that means it occurred at the first scan time point, which was week eight. So already at week eight, we see a separation potentially of the progression-free survival curves with 42.1% progression in the control arm, but only 16.2% progression in the combination arm. So we believe that there's a high likelihood of the study being positive on progression-free survival, which is the next endpoint to be analyzed, and that will be followed by an analysis of overall survival. So where are we with the study? So in August, we made what I believe is an extremely important announcement, and that is in the red box on the top here.
In August, we announced that we have fewer deaths in the study than originally projected, said the exact opposite way, more patients are alive and living longer than we had projected. Now, importantly, these are pooled blinded data. However, given what we saw in the overall response rate data, we believe that this is suggesting that we're seeing an effect of tovecimig on overall survival. For reference here, one three-drug regimen that is used in patients with this disease is the three-drug combination FOLFOX. In one randomized trial with FOLFOX, at month 18, there was less than 10% of patients remaining alive. When we made this disclosure in the fall, at 18 months follow-up in our study, there was more than 20% of patients alive, of course, because we had not hit the 80% mortality threshold to trigger the analyses of the secondary endpoints.
So an extremely important update that we made to the study in August and September. We disclosed last week that we believe we're on track to report the analyses of progression-free survival and overall survival later this quarter. Should those studies be positive, should those analyses be positive, we believe that tovecimig has the potential to become second-line standard of care in patients with advanced biliary tract cancer. First-line standard of care is chemotherapy plus a checkpoint inhibitor, gemcitabine and cisplatin plus either durvalumab or pembrolizumab. In the United States, there is no second-line standard of care. FOLFOX has a 6.2-month median overall survival, which is 0.9 months better than best supportive care. That is not a standard of care. So should our PFS and OS analyses be positive, we would prepare to commercialize this drug in the United States. How many patients are out there?
In 2025, we completed some market research, and we've identified approximately 25,000 new patients annually. This is a very significant commercial opportunity. Unfortunately, only about 15% of patients have an actionable mutation in their tumor, making them eligible for a targeted therapy such as FGFR2 or IDH1. Unfortunately, the other 85% of patients literally in the U.S. have no labeled standard of care. We would target those patients, making some adjustments for how many patients might be well enough to make it to second-line therapy. We believe that the treatable patient population is approximately 15,000 patients annually, assuming perhaps six-plus months on drug and making any modern reasonable assumption about drug pricing. This is well over a $1 billion market annually in the United States alone in second-line biliary tract cancer only.
However, there are many other opportunities to expand the indications for this drug, and we're preparing to initiate a basket study in patients who have tumors that are known to be DLL4 positive, and there are many other such indications, including, as I mentioned earlier, colorectal cancer, where we have unequivocal monotherapy activity for this drug in the third- and fourth-line setting, as well as gastric cancer, hepatocellular, ovarian cancer, and others, so following the approval of tovecimig in the United States, we would move rapidly to expand the label into many of these other indications. Ultimately, of course, our goal would be to replace Avastin in all of the Avastin indications, which would make this an enormously commercially successful drug, so with tovecimig, our lead program, here's where we are today. Our randomized study by definition is positive because we achieved the primary endpoint in that randomized trial.
We expect to analyze the secondary endpoints of progression-free survival and overall survival later this quarter. Interestingly, MD Anderson in Houston was the second largest enroller in our randomized trial, and after they had treated six or seven patients or so, they came to us with a proposal to do an investigator-sponsored study where tovecimig is being added to the front-line regimen of gemcitabine, cisplatin, and durvalumab. That study is now ongoing at MD Anderson. It'll be very interesting to read out data from that study potentially later this year, and as I mentioned, the potential to expand the label. Following the readout from our randomized trial, we would expect to have an interaction with the FDA. If that goes well, we could potentially submit a license application later this year.
Because we have fast-track designation in the U.S., we would certainly get a priority review, which would put our PDUFA date inside the first half of 2027. And we're now preparing to commercialize this drug in the United States. Let's spend a few minutes on a couple of our other drugs, starting with CTX-471. 471 is an agonist antibody of CD137. It binds to a unique epitope with a unique affinity. In a Phase I dose escalation study followed by cohort expansions, we had responses in the post-PD-1 patient population in patients with melanoma. We had just under a 30% response rate there. We had one out of four patients with mesothelioma who responded.
And fascinatingly, on the top of this slide, we had a patient with metastatic small cell lung cancer who received front-line therapy with chemotherapy and the PD-L1 blocker atezolizumab, was on that regimen for about four months and then progressed, then got second-line nivolumab, was on that for about two months and then progressed. So this is a third-line metastatic small cell lung cancer patient. The median overall survival of that patient population is approximately nine weeks. Think about that number. This patient started monotherapy with 471, quickly responded, that was confirmed. This patient had a PET-negative complete response and was on therapy for more than three and a half years when he came off therapy with a complete response.
Interestingly, we collected biopsy specimens from this study, and this patient's biopsy, which is shown on this slide, turned out to be very positive for neural cell adhesion molecule, NCAM, also known as CD56. That made us look at other patients' biopsies in the study, and I think we might have made an important discovery here. NCAM might be a biomarker of response to CTX-471 because it could mediate NK cell activation in the tumor microenvironment. So we're now planning to do an NCAM-positive basket study. NCAM is expressed on neuroendocrine tumors, small cell lung cancer, of course, melanoma, and many others. That study should begin in the first half of this year. Let's now spend a couple of the last minutes talking about CTX-8371. So at Compass, we developed a technology that we call StitchMabs.
StitchMabs is a tool that allows us to screen for biological synergy in bispecific drug candidates. We used that to ask an important question, which is, what is the best combination partner for PD-1 blockade in a bispecific antibody? StitchMabs also allows us to create small libraries. We did that. We screened that library, and fascinatingly, the best combination partner for PD-1 blockade turns out to be PD-L1 blockade. A very unexpected discovery. On the top right-hand side of this slide, the combination of dual ligand receptor blockade in a bispecific antibody is between 100- and 1,000-fold more potent than either antibody alone. But it turns out not to be just dual ligand receptor blockade, which it is. But on the right-hand side of this slide, 8371 is unequivocally a T cell engager.
In addition, in the middle right-hand side of this slide, 8371 exposes a cleavage site on PD-1, leading to the cleavage of PD-1 off the surface of effector T cells. This drug actually converts PD-1-positive T cells into PD-1-negative T cells that subsequently cannot be suppressed. So this is a very novel next-generation checkpoint inhibitor. We recently completed the dose escalation portion of a Phase I trial. We enrolled patients with melanoma, non-small cell lung cancer, head and neck cancer, Hodgkin's lymphoma, and triple-negative breast cancer, all who have previously received checkpoint inhibitors. That is an extremely important take-home point. All of these patients are treated in the post-checkpoint setting. We had no dose-limiting toxicities in the study, which is fascinating in and of itself because we believe this drug could be anchored in the tumor microenvironment via PD-L1, where it could provide local high-concentration blockade of PD-1.
So because this is a 3 plus 3 design, no DLTs, five dose levels, we've treated 15 patients, right? Five times three. And we already have three responses. We have a response in a patient with non-small cell lung cancer. This is a patient post-Ipi-Nivo. In the blue circle here, you can see complete disappearance of a metastatic tumor. Our most important patient in this study is shown on this slide. This is a patient with triple-negative breast cancer who relapsed while receiving adjuvant KEYTRUDA, then got TRODELVY, followed by two different chemotherapy regimens. This patient had 9 cm of tumor burden at baseline. And in the two blue circles here, a lung metastasis on the top and a pericardial heart metastasis on the bottom that was 5.2 cm large. Those have completely disappeared.
This patient is a near-complete response in the post-TRODELVY setting, now durable at month eight, we recently announced. Finally, we announced last week that we have another response in a patient with Hodgkin's lymphoma. We now have solid tumor responses and our first hematologic malignancy response. The triple-negative breast cancer patient is one of two patients in the study with triple-negative breast cancer who were evaluable. The Hodgkin's lymphoma patient is the only patient in the study with Hodgkin's lymphoma, and that patient has had a confirmed metabolic partial response. We've now opened cohort expansions in patients with non-small cell lung cancer and triple-negative breast cancer, and we're currently evaluating the next steps for patients with Hodgkin's lymphoma. Finally, in the last minute, our fourth drug, CTX-10726, is a novel PD-1/VEGF-A bispecific, 2x2 valency, N-terminal anti-VEGF, C-terminal PD-1 blockade.
We showed some preclinical data at SITC in the fall demonstrating that this drug has superior PD-1 blockade to other drugs in the class. And interestingly, it has better tumor control in mice than other drugs in the class. I'm sure you all well know this is a new class of drugs, PD-1/VEGF-A bispecific antibodies, and our drug is superior to other drugs in the class in both syngeneic mouse models and in xenografts where we're testing this against human tumors. That drug is going to go into the clinic this quarter in patients with renal cell cancer, endometrial cell cancer, gastric cancer, and hepatocellular cancer. So this is my last slide. 2026 is going to be a very busy year for us at the company, starting with the PFS and OS readout from our tovecimig clinical trial in patients with biliary tract cancer.
As I mentioned, if those data are positive, we would then have an interaction with the FDA. If that goes well, that would be followed by a biologics license application, which again could put us in position to launch this drug in the United States in the first half of 2027. Our DLL4 positive basket study to begin following that readout, our NCAM positive basket study for 471. And again, for 8371, our cohort expansions are now open in patients with non-small cell lung cancer and triple-negative breast cancer. Just to speculate here, should we see additional responses in patients with triple-negative breast cancer? I think we would have an accelerated conversation with the FDA about moving that drug directly into an approval trial in the post-TRODELVY patient population. Stay tuned for that. That's going to be one of the more important developments at the company in the coming year.
And then finally, our fourth drug in the clinic, 10726, we look to report data in the second half of the year from that Phase I study. Thank you very much. Happy to take some questions, and Sherry and I are maybe going to have a conversation as well. Thank you.
I can start here. So starting off with your tovecimig asset, you're currently waiting for secondary endpoints, and this could be transformational for your company. In April, you achieved statistical significance results in your primary endpoints of ORR. What gives you confidence in the upcoming PFS and OS readouts?
Sure. Thanks for that question. I think that's an extremely important question, and I went back to Slide 11 in the presentation to help address this question, so in the red box on the bottom, you can see the difference in progressive disease in the study, so with 42% radiographic progression already at week eight, and unfortunately, this disease has tremendous early mortality. Unfortunately, some of these patients in the control arm may have already died by week eight with FOLFOX. Think about this number, by the way. The two-month mortality with FOLFOX is about 10%-15%, so 10%-15% of patients with this disease don't live two months, so we believe that the median progression-free survival in this study is going to be approximately two months. That's what these data would suggest.
With only 16.2% progression in the combination arm, there's a suggestion here that the progression-free survival curves are already separating and potentially separating in a very meaningful way. This study is powered to detect a hazard ratio of approximately 0.6. A hazard ratio of 0.6, of course, would be clinically meaningful by any definition and would actually be an extraordinary result in this indication. In terms of the OS readout, simply the amount of time that has gone by gives us confidence that we're seeing an effect on overall survival because the study enrolled 168 patients, as you can see on this slide, in a 2 to 1 randomization, 111 in the combination arm, 57 in the control arm, 168. The 84th patient was enrolled the first week of March in 2024, so we are coming up on two years out.
Every patient in the study is more than 16 months out from joining the study because the study was fully enrolled in August of 2024. So simply the amount of time that has passed suggests to us that we're seeing something very, very important and meaningful here.
And now.
Go ahead.
Can I ask the indication selection for tovecimig versus you have a PD-1/VEGF going into clinic as well? Can you talk about selection between?
Sure. Sure. That is a fantastic question. Thank you. So with tovecimig, we're mostly focused on DLL4 positive malignancies. So there's a whole set of malignancies there that I mentioned. So ovarian cancer, for example, is well known to be DLL4 positive in addition to hepatocellular, gastric, and colorectal cancer. But you're correct. There is some overlap between our development program with 10726, our VEGF-A /PD-1 bispecific antibody. The overlap would be patients with renal cell and gastric cancer. So I think we have very good mechanistic rationale to go into both of those indications. Ultimately, I think where we'd like to end up is with combination therapy. And I think for our PD-1/VEGF bispecific, I think ideally it would be great to see a signal in patients with hepatocellular cancer, for example, where frontline therapy is the combination of VEGF blockade and PD-L1 blockade.
So second line hepatocellular would be a very interesting place to go into. In terms of tovecimig, I think ultimately we'd like to move in a direction where we're combining tovecimig with 8371. And I think there we would probably focus on patients with gastric cancer and probably renal cell cancer with tovecimig. So it's a very interesting question. Honestly, that'll be a very good problem to have, right, if we see efficacy with both drugs in those indications.
Now, I guess still focusing on tovecimig, you mentioned your most recent thoughts on your in terms of commercial readiness within the context of your new commercial officer. Will you be commercializing this on your own, or are you speaking to potential partners?
We are preparing to commercialize this drug in the United States on our own. I think interestingly, I'm just going to go back one slide here. On the far left-hand side of this slide is a diagram of the liver and the biliary tree. The reason that I'm focusing on that is to make a very simple point. Patients with biliary tract cancer have anatomic complications of their disease, so they end up with obstruction in their biliary tree, which requires endoscopic stent insertion and replacement. That type of procedure is not generally done at community hospitals. The vast majority of these patients are seen at academic referral centers. We believe with a focused commercial team, we can commercialize this drug ourselves in the United States, and we are preparing to do so.
We have said publicly that we're entertaining conversations with strategics, but I don't know where those conversations will go. Certainly, the rest of the world, Japan, and the European Union, we would need to address commercially in the future, but we are going to launch this drug ourselves in the United States.
Thank you. Now turning to your 8371 asset, your PD-1 and PD-L1 bispecific, you've shown responses in three separate tumor types within the first 15 patients that were dosed, all of whom are post-checkpoint inhibitor. What about this asset is driving this efficacy?
Sure. So I'm just going to go back to the mechanism of action slide here because this drug is clearly different from a mechanism of action point of view than any known checkpoint inhibitor. So either PD-1 blockers or PD-L1 blockers do something relatively simple, which is on the far left-hand side of this slide. Those drugs will either block the interaction of PD-1 with its ligands or block PD-L1's interaction with PD-1. That's what they do. 8371 does. Not only does it do those things, it does some other things as well. And I think the activity of this drug as a T cell engager is quite fascinating. And then the fact that it converts PD-1 positive T cells into PD-1 negative T cells is an extraordinarily unique mechanism of action. So this observation has now been confirmed in the clinic with responses in the post-checkpoint patient population.
So these are extremely difficult patients to treat. In the post-checkpoint patient population, there's really very, very few options for these patients. And for us to have monotherapy activity in this patient population is an extremely important signal for us. And the diseases where we've seen activity, triple-negative breast cancer, my God, post-TRODELVY in the triple-negative breast cancer patient population, there's absolutely no therapy for these patients. So this would be a wide open field for us to not only develop this drug, but ultimately to get this drug approved in that indication. And the most recent update that we gave, Hodgkin's lymphoma is a very important observation because, as you may know, Hodgkin's lymphoma is a PD-L1-driven malignancy. So seeing a response in a patient with Hodgkin's lymphoma is directly on mechanism for this drug.
This patient with Hodgkin's lymphoma, I can tell you, had prior brentuximab, prior nivolumab, and a transplant. We're seeing a monotherapy response with 8371. This is a drug we're extremely excited about. Something I don't commonly talk about on the bottom of this slide in the gray bar, the potential for proprietary combination regimens. I think one of the most important advances in oncology in the past 10 years has been the combination of angiogenesis inhibitors with checkpoint inhibitors. The very first observation there was the study IMbrave150 that was done by Roche, where the combination of bevacizumab and atezolizumab in frontline hepatocellular cancer was superior to standard of care sorafenib. That was the very first observation that was published in the New England Journal of Medicine in 2020. That study has opened the floodgates for testing these hypotheses.
The PD-1/VEGF-A bispecific field is a completely new class of drugs targeting both angiogenesis and checkpoint inhibition. Where we really want to go with this drug is to combine tovecimig with 8371 as a super next-generation angiogenesis plus checkpoint inhibitor regimen.
On the question of commercialization, do you have the infrastructure in place today to do that, or how do you plan to support that?
Sure. Very important question. So we announced last week that we hired Arjun, our Chief Commercial Officer, and we're now building that infrastructure and will do so over the coming 12-18 months as we prepare for the launch of this drug.
Thank you. Now moving to your last asset that you've talked about or didn't have a chance to mention. Your CTX-10726, your PD-1/VEGF bispecific. You presented your preclinical data, yes, against the most leading asset in this space, which is ivonescimab. Can you share us about what stands out specifically in the terms of differentiation?
Sure. So when we were developing this drug, we spent some time investigating whether or not we should block PD-1, block PD-L1. Ultimately, we landed on a very similar construct to ivonescimab that you mentioned. But very quickly, we realized that our PD-1 end of this molecule is a better binder of PD-1, and it is a more potent inhibitor of PD-1 signaling. So that was the first little hint that we got that this drug could be different. We then did head-to-head testing in preclinical models. And in selected preclinical models, we have shown that 10726 is actually superior to ivonescimab. Now, take that for what it is. I just want to be fair here. We're comparing these drugs in mice, so we have to be fair about that. But we clearly see a differentiating signal.
And I think recently, I apologize because this is going to be a little bit of a tease, so stay with me here. But recently, scientists at Compass have made what I believe is potentially an extremely important observation about the mechanism of action of this drug that suggests it could well be a very differentiated asset. Stay tuned for that. We would love to potentially discuss that later this year. But we believe that we have some clear differentiation of this drug. Ultimately, of course, we'll need to differentiate this drug in the clinic.
And then just after that tease, when do we first expect the clinical data? And are you expecting to go this alone or a partner in this very increasingly competitive space?
Sure. So Phase I initiation this quarter, which would put us in a position to have some clinical data in the second half of this year. The second half of your question, of course, is fascinating, as you obviously know full well. There are now multiple drugs in this class. It is going to become a competitive area. And strategic partners are quite interested in this class of drugs. And specifically, we have strategic interest in this asset. Of course, we will do the best thing for our shareholders. We're currently going to initiate the Phase I study this quarter, but we'll see where that goes in our conversations with potential strategic partners.
Awesome. Thank you so much, Tom. That concludes our presentation. And please give another round of applause for Tom. Thank you.
Thank you.