I chat with Thomas Schuetz, CEO of Compass Therapeutics. Just from a disclosure standpoint, we cover Compass with a buy rating. Thank you, Tom, for joining us.
Thank you.
All right. We've definitely got lots of recent updates and upcoming news flow to cover in our chat today. Just to first orient our audience, could you provide maybe an overview of your core technology, the StitchMabs platform, and your general approach to developing antibody-based therapeutics for cancer treatment?
Thanks again to H.C. Wainwright for inviting us to the conference. Compass Therapeutics, we're located in Boston. We're a biologics discovery and development company in oncology. All of our drug candidates are either monoclonal antibodies or bispecific antibodies. We currently have four drugs in the clinic. Our most advanced program, which we'll talk more about, is a DLL4 VEGF-A bispecific antibody. That's a next-generation angiogenesis disruptor. We have a novel PD-1/PD-L1 bispecific antibody, which is currently in cohort expansions in a phase I study, and we're gonna present data at ASCO in about 10 days' time, including responses in patients treated in the post-checkpoint inhibitor setting. We have a CD137 agonist antibody that we're gonna be talking about more later this year.
Finally, we have a PD-1 VEGF-A bispecific antibody that we developed ourselves, and that has just begun Phase I testing in patients with renal cell cancer, endometrial cell cancer, gastric cancer, and liver cancer. As you mentioned, you know, we still do a lot of discovery research, and a couple of these, our drug candidates, were discovered using a technology that we call StitchMabs that allows us to screen for synergy in bispecific drug candidates.
Great. As you mentioned, your lead asset is tovecimig, so let's start there. Certainly have a lot to discuss, particularly with the recent biliary tract cancer data you reported. Maybe before we get into that, as you discussed, you know, it's a next-generation antibody, maybe why you believe that two-pronged angiogenesis approach could be differentiated from a mechanism-of-action perspective.
The two targets of tovecimig as a bispecific antibody, of course, are VEGF-A, vascular endothelial cell growth factor A, the well-known target of drugs like Avastin. The other target for tovecimig is delta-like ligand 4, DLL4. That is the cell surface ligand for Notch1. Those two signaling pathways, DLL4 and Notch1, and VEGF-A, VEGF receptor, have different phenotypic effects on angiogenesis in the tumor microenvironment. Importantly, dual blockade of those two pathways in preclinical models has been shown to be synergistic. Also, in preclinical models, DLL4 upregulation in the tumor microenvironment has been shown to mediate resistance to VEGF-targeted therapies such as Avastin. We believe that the dual targeting approach will usher in really a whole new generation of angiogenesis inhibition.
Great, you've certainly got a decent amount of data now backing this up. Your lead indication for tovecimig is second-line biliary tract cancer, and you actually recently reported the secondary endpoint data from your pivotal phase II/III COMPANION-002 study. You know, of the results, the PFS result here was striking. You had a hazard ratio of 0.44. Maybe can you just walk us through what that result means in the context of second-line BTC, and maybe why you believe the totality of the data you've amassed so f ar presents a compelling data and treatment option for patients in this setting.
Sure. Very complicated and detailed question, of course. Maybe just briefly on the design of the study. The COMPANION-002 study was a randomized study, a two-to-one randomization of tovecimig plus the chemotherapy agent paclitaxel versus paclitaxel alone in patients with advanced biliary tract cancer who had received one prior line of therapy. It's a pure second-line study. Patients treated in the front-line setting get gemcitabine, cisplatin, and a checkpoint inhibitor. The study was a two-to-one randomization. 111 patients were randomized to the combination arm. 57 patients were randomized to the paclitaxel control arm. Importantly, a crossover from the paclitaxel arm to the combination arm was allowed after central confirmation of progression. It's important, the crossover was a very important part of the study.
When we were first putting the study together, you know, scientific review committees at major academic institutions required us to have a crossover arm in the study. That's a very, very important aspect. One thing that we've skipped over maybe, and I just wanna make this point, you know, last year we announced that the study hit the primary endpoint of overall response rate, and we more than tripled the response rate. All responses were assessed by blinded independent central review. The most rigorous way you can do it. We more than tripled the response rate in the control arm. The study by definition is positive. The first secondary endpoint to be analyzed in the stat plan was progression-free survival.
As you mentioned, it was strikingly positive, with a hazard ratio of 0.44 and a P- value less than .0001. You know, an overwhelmingly positive result. I think importantly, at the end of 2025, a meta-analysis was published in the Journal of Hepatology, showing that there's a correlation in patients with biliary tract cancer of progression-free survival and overall survival. I think part of the reason for that is unlike many other, you know, solid tumors, biliary tract cancer is really associated with a lot of anatomic complications in the biliary tree. Any effect on progression-free survival is very clinically meaningful. Hit on ORR, hit on PFS. The overall survival analyses were confounded by crossover, which is not something complicated or new.
We've seen this hundreds of times before in oncology clinical trials, and we showed, in the subset analyses that the patients who crossed over to receive tovecimig did incredibly well with a median overall survival of 12.8 months, which is about what you would see in the frontline setting. And of course, as you know, I'm not trying to be trite here, but as you know, there are hundreds of regulatory precedents of drugs that are approved with either ORR and PFS, without an OS demonstrable OS benefit. Clearly, you know, in our study, the ITT analysis of OS was affected by the crossover.
You know, we're planning to submit all this data to the FDA, you know, have a meeting with FDA in the summer, and I, you know, feel pretty good about that meeting, and we're continuing our preparation for our BLA submission later this year.
Great. I actually wanted to dive a little further into that survival data. Actually, one of the most intriguing findings that came out of it was that the patients who progressed most rapidly on paclitaxel monotherapy before crossing over actually went on to do quite well once put onto tovecimig. What's your current thinking or hypothesis of what might be happening, and is there any work that you're doing that's underway to kind of explore this particular phenomenon further?
Sure. As you mentioned, you know, one of the things we, a prospective secondary endpoint in the study was something that we called PFS-2, where we took the crossover patients, and we compared their progression-free survival after crossover to their progression-free survival pre-crossover, right? This is a perfectly internally controlled set, right? It's the same 31 patients on paclitaxel alone. They had a median PFS of 1.9 months. After crossover, that improved to 3.5 months and, you know, again, you know, with a similar magnitude hazard ratio. Clearly an effect of the drug, and there's just no question, you know, that the drug clearly works in this patient population. You know, I could probably sum up your question here with one single word, why, right?
Why did those patients who crossed over do so well in the study? One of the things we're working on is, you know, a simple but complex question, of course, which is, does a couple of months of paclitaxel somehow prime the tumor microenvironment for subsequent therapy with tovecimig? That's a very, very challenging question, and we've started some work in the lab, you know, to try to, you know, think about investigating this. It's an important question because, you know, this could inform our future clinical trials, especially in other indications such as gastric cancer. You know, I think one of the things that we really learned from this study is there seems to be something unique about the combination of tovecimig with paclitaxel, and that has made us think about indications like ovarian cancer and gastric cancer.
In gastric cancer, we clearly had a monotherapy signal in our phase I study. Second-line standard of care in patients with gastric cancer is the VEGF receptor-targeted antibody ramucirumab with paclitaxel. You know, we really need to figure out, you know, how we do that study, right? We haven't really decided yet, but a very simple thing to do would be a two-month lead-in with paclitaxel and then add tovecimig to that. We would need to really have strong scientific rationale for doing that, and we don't yet have that, but that would be a very straightforward thing to do in the clinic.
It definitely could be very intriguing. You touched on this a little bit before, but just looking forward as to your next steps for tovecimig and you've guided that you're gonna meet with FDA before filing a BLA later this year. Just walk us through maybe your current, you know, timing and regulatory strategy and, you know, kind of what's your read on the FDA acting, like you said, on the ORR and PFS in this particular setting alone in patients that really don't have another option.
Sure. We're gonna meet with FDA in the late July, early August timeframe. That meeting has not yet been scheduled, it'll be in that approximate timeframe, where we'll present all our data and align on a plan to submit a license application. Yeah, it's a very complicated question, what is my read on FDA? I'm not, I'm not sure. I get asked this question a lot, and I always say the same thing, which sounds silly, but we have not noticed any difference in FDA at all. Our medical reviewer is the same, the responses to our submissions are the same. Like, everything's the same.
I realize that there's lots of turmoil there, for us, we haven't seen or felt that. I don't know what that means.
Hopefully that stays-
Yes.
-being positive for you then. Yeah. No, that's good to hear, and we'll be looking forward to the updates. Now, aside from the second line setting, you also have an investigator, an IST study in the first line.
BTC, with standard of care combination with tovecimig. Maybe could you give us a status update of how that's going, and maybe framing what the potential commercial opportunity could be for tovecimig in the first line setting?
Sure. As you mentioned, at MD Anderson, there's a frontline study ongoing, which is adding tovecimig to the frontline regimen of gemcitabine, cisplatin, and durvalumab, IMFINZI. That study's enrolling patients at MD Anderson. And that study's going well, and I think, you know, we could use that data, you know, to help us, you know, think about tovecimig commercially once we launch the drug in the second line setting. You know, the commercial opportunity in biliary tract cancer in the United States is very significant.
I think very importantly, a couple of weeks ago at the annual Cholangiocarcinoma Foundation meeting, an abstract was presented where they've done a new analysis of the epidemiology data in the U.S., and they found about 25,000 newly diagnosed patients annually in the U.S. It's a very, very significant commercial opportunity. You know, probably, you know, triple-ish, double, triple platinum-resistant ovarian cancer. You know, indications that are, you know, thought to be very big markets, you know, biliary tract cancer is well above, some of these other large markets.
Okay. Now, you know, with the time we have left, I want to switch gears a bit and spend some time discussing your broader pipeline, which as you mentioned, is really going to be coming into focus now going into the second half of the year. First up, you have CTX-8371, your PD-1/ PD-L1 bispecific, for which you mentioned you have at that upcoming ASCO presentation. Maybe, yeah, could you preview what we could expect to see at the conference, yeah, in a few days?
Sure. CTX-8371 is a drug that, you know, came out of our StitchMabs platform where we, you know, ask the question, in a bispecific antibody, what's the best combination partner for PD-1? We use that technology to make a small library and discovered some very unique synergy between PD-1 blockade and PD-L1 blockade. In a bispecific antibody, the combination of dual ligand and receptor blockade is between 100 and 1,000 fold more potent than either alone or the combination of the two. It was a very unexpected and very interesting discovery that we made. We spent quite some time investigating that drug, and we found that it has a very unique mechanism of action. It's unequivocally a cell engager in addition to blocking PD-1 and PD-L1.
Most interestingly, the bispecific exposes a metalloproteinase 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. We went into the clinic, we did a dose escalation phase I study, a standard 3 + 3 design, five different dose levels, and we went into the post-checkpoint patient population. That's very important, right? All these patients had previously received checkpoint blockers, so five different dose levels. We had no dose limiting toxicities, so 5 x 3 is 15. 15 patients enrolled at 5five different dose levels, which had a 100-fold dose range from 0.1 mg/kg to 10 mg/kg. At the higher dose levels, we had a couple of very impressive antitumor responses.
We have a patient with triple negative breast cancer who was relapsed while receiving KEYTRUDA, then received TRODELVY. That patient has had nearly complete disappearance of almost 9 cm of linear tumor burden. That patient continues on drug and is durable for more than one year, we'll present new scans and new durability data for that patient. We also have a patient with non-small cell lung cancer who relapsed post-ipi/nivo. That patient also had a near CR after initial pseudoprogression, a very, very interesting result. Finally, we have one patient out of the 15 with Hodgkin's lymphoma, again post-nivo, that patient had a partial metabolic response, almost a complete response. We're gonna show all of the PET scans for that patient, report durability. Based on those observations, we initiated cohort expansions.
We added cohort expansions to the study, and we're doing cohort expansions in patients with triple negative breast cancer, non-small cell lung cancer, and Hodgkin lymphoma. Those cohort expansions opened in maybe around February, you know, in Q1. That study is enrolling extremely well. I think we have more than 10 patients enrolled already, so we'll have some early safety data, you know, from the cohort expansions. I think maybe later this year, you know, we'd be able to talk about some additional efficacy data, hopefully, you know, from the cohort expansions. In Post-checkpoint Hodgkin lymphoma and triple-negative breast cancer, you know, I would think there could be an extremely rapid development path to approval in those two indications.
You know, stay tuned, you know, for an update on the development program for that drug later this year.
All right. Well, looking forward to it. You're also developing CTX-10726, your PD-1 VEGF bispecific, which recently entered phase I studies in solid tumors. Could you maybe frame the opportunity here as well, and anything you'd like to tease us with-
Sure.
-regarding to a first look at data?
Sure. We developed our own novel PD-1 VEGF bispecific antibody, and we showed some data at SITC, maybe a year, a year and a half ago, suggesting that it has better PD-1 blockade than other drugs in the class. We've started a phase I trial, again, a 3 + 3 design, four cohorts. It's not all solid tumors. We selected patients. We selected indications where there's evidence of both checkpoint inhibitor activity and angiogenesis blocking activity. The patients we're enrolling have gastric cancer, hepatocellular cancer, renal cell, and endometrial cancer. Those are the four cancers that we're studying. We had very, very good preclinical safety data, so we're able to start at a little bit higher dose.
We should be getting to, you know, potentially efficacious doses fairly soon, and I'd love to, you know, sort of be able to have, you know, an update from that study later this year as well.
Great. Rounding out your pipeline, you have CTX-471, which is targeting CD137, and I believe you're planning to initiate a Phase II study in tumors expressing NCAM or CD56 later this year. Maybe could you briefly walk us through the strategy behind why you believe the CD56-expressing tumors are the right patient selection strategy?
Sure. That study has been submitted to FDA. That study is getting ready to go. You know, in our phase I study, again, in the post-checkpoint inhibitor patient population, we had responses in patients with melanoma, small cell lung cancer, and mesothelioma. Fascinatingly, when we looked at biopsy specimens that we collected in the study, the patients who responded to the drug had tumors, not infiltrated in lymphoid cells, tumors that were CD56- positive. CD56 has something called a homotypic interaction, which means that CD56 is its own ligand and receptor. CD56 is a marker for a major subset of natural killer cells. CD56 is also known as NCAM, neural cell adhesion molecule. Melanoma and small cell lung cancer are neural crest-derived tumors. They express NCAM.
We believe we've discovered a novel immunophenotype in the tumor microenvironment whereby NCAM-positive tumors can attract CD56-positive NK cells, where we then can activate them with a CD137 agonist antibody. That NCAM-positive basket study is gonna start in the second half of this year.
All right. Sounds good. Lots of exciting updates to look forward to. Thank you, everyone, and thank you for joining us today, Tom.
Thank you so much.