Good morning. Let's kick off our next session. It's my pleasure to be hosting Compass Therapeutics. With me is Tom-- of the company. Welcome. It's the first time for me to host you at the GS conference, welcome and look forward to a very exciting discussion. Before we begin, I'm going to kick it off to you for an opening remark.
Sure. Thank you very much. Thank you for inviting us to the conference. Really excited to be here. Compass Therapeutics is located in Boston. We're a monoclonal antibody discovery and development company in oncology. We currently have four drugs in the clinic, which hopefully we'll talk about today. Our lead asset is a drug called tovecimig. That's a DLL4 VEGF-A bispecific antibody, we just read out secondary endpoint results from a randomized trial, and we're going to approach the FDA with those results later this summer.
Our next most advanced program is a next generation CD137 agonist antibody called CTX-471. That drug has shown responses in the post-checkpoint inhibitor patient population. Our third drug, we presented data last weekend at ASCO, is a novel PD-1/PD-L1 bispecific antibody, which sounds counterintuitive, but that drug came out of a screen using a technology that we developed.
It came out of a screen for synergy with PD-1 blockade, fascinatingly, PD-L1 turned out to be the best combination partner for PD-1 blockade. At ASCO last weekend, we presented data from a phase I dose escalation study in which we have responses in, again, the post-checkpoint patient population. Finally, we also have a PD-1 VEGF-A bispecific antibody, which obviously is a new class of drugs, that drug is now in phase I testing, moving through the first dosing cohort. Happy to talk about those, again, thanks for inviting us. Happy to be here.
Great. Yeah, very exciting. A lot to talk about. Can you give us a quick view on what's in the catalyst for the next 12 months? What's that path look like?
Sure. I think we're very excited about the next 12 months because we've got multiple catalysts for each of our four drugs. Beginning with tovecimig, we'll have a conversation with the FDA later this summer. If that conversation goes well, we would then plan on submitting a biologics license application, and about a year from now, we would potentially be in label negotiations. That would be incredibly exciting. We've got a fast track designation for that drug, so we almost certainly would get a priority review, which would put our PDUFA date in the second half of 2027. At this conference next year, it'd be super cool to be talking about our PDUFA date.
Those are the big things for tovecimig. Also, interestingly, we believe that we've uncovered some rather unique combinatorial synergy between tovecimig and the chemotherapy drug paclitaxel. We've been thinking about which indications we might combine tovecimig with, and we're looking at potentially ovarian cancer and gastric cancer, among others. That will also be some interesting information coming up for tovecimig.
For CTX-471, fascinatingly, we identified a potential biomarker of activity, which was based on analysis of biopsy specimens from a phase I study, and we discovered that NCAM, neural cell adhesion molecule, also known as CD56, is correlated with responses to CTX-471 in a phase I study. In the second half of this year, we're going to be initiating an NCAM positive phase II basket study. CTX-8371, our PD-1/PD-L1 bispecific, in the phase I dose escalation study, we identified responses in patients with triple-negative breast cancer, Hodgkin lymphoma, and non-small cell lung cancer. Again, all post-checkpoint inhibition.
Those were monotherapy responses in the post-checkpoint population, and based on that, earlier this year, we initiated cohort expansions in the phase I study in patients with those three diseases, triple-negative breast, Hodgkin lymphoma, non-small cell lung cancer. Later this year, by the way, the cohort expansion portion of that study has been enrolling extremely well, and I think later this year, we'd be in a position to report some data from the cohort expansion, which could inform pathways to further development for that drug. Then finally, for our PD-1 VEGF-A bispecific, that drug we call CTX-10726. By the end of this year, we should get dose escalation data from the phase I study. In the next 12 months, I think we have really meaningful catalysts for all of our four drugs.
Again, very exciting. What does your cash runway look like? Also, what does it cover and not cover?
Ended Q1 with $195 million, which is cash runway into 2028, covering all of those clinical trials that I have just described. We have begun some commercial preparation for launch, but our cash runway probably doesn't fully fund us to profitability. It probably doesn't cover all of the launch.
I see. Okay. Let's now move our discussion into tovecimig. That's a very exciting product. I think you mentioned that potential BLA filing. Soon in the future. With respect to this mechanism, with the DLL4, can you walk us through this mechanism in terms of how it can be used broadly across patients post that IO front line setting?
Yeah. tovecimig, as you mentioned, is a DLL4 VEGF bispecific antibody. DLL4, delta-like ligand 4, is the cell surface ligand for Notch1. VEGF-A, of course, is the well-known soluble ligand for the VEGF family of receptors. VEGF-A, VEGF receptor signaling, and DLL4 Notch1 signaling have different phenotypic effects on angiogenesis in the tumor microenvironment. This drug really is a next-gen angiogenesis inhibitor. A couple fascinating things, dual blockade of those two pathways has been shown to be synergistic in preclinical models, and what could turn out to be even more important, potentially, is that DLL4 upregulation in the tumor microenvironment has been shown to mediate resistance to VEGF-targeted therapies such as Avastin.
We believe that this can be a next-gen angiogenesis inhibitor, which would mean just like bevacizumab or VEGF kinase inhibitors, these could be used in patients with multiple different malignancies and combined with multiple different chemotherapy regimens. We could be on a pathway to, in our dreams, replacing Avastin in all of the Avastin labeled indications, including renal, gastric, colorectal, et c. That's where we would ultimately imagine going with that drug.
I see. Okay. Very ambitious. I think recently you guys presented some survival data with the tovecimig and that paclitaxel combination t he phase II/III COMPANION-002 study, in the second-line BTC. What were the key highlights there?
Sure. Let's briefly on the study design. The study design was 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's a pure second-line study. Crossover was allowed from the control arm, paclitaxel, to the combination arm after a centrally confirmed progression event. This is very, very important because about 54% of patients in the study crossed over to receive active drug. Because it's a two-to-one randomization, that means about 85% of patients in the study received tovecimig at one point in time, either at randomization or at crossover.
Let's go through the efficacy endpoints. The primary endpoint was overall response rate. We hit the primary endpoint. We more than tripled the response rate with paclitaxel, and that was statistically significant, and all responses were assessed by blinded independent central review. A very, very rigorous process. The key secondary endpoint was progression-free survival, and we hit that endpoint, and that was highly significant. The difference between tovecimig and paclitaxel had a hazard ratio of 0.44 with a P value less than 0.0001.
A highly significant effect on PFS, which frankly is very clinically meaningful in this disease because patients with biliary tract cancer suffer from anatomic complications, obstruction of the biliary tract, for example, and any delay in progression has a very clinically meaningful impact in this disease. The next endpoint was overall survival, as you mentioned. Because of the crossover, the intent to treat analysis, which of course, the intent to treat analysis makes no sense, right?
We're comparing chemotherapy plus tovecimig versus chemotherapy plus tovecimig. It's like, well, what are we talking about here, right? The statistical word there is confounded. The ITT analysis was confounded by the crossover. We did some subset analyses, which showed that patients who crossed over to receive tovecimig versus patients who had paclitaxel alone more than doubled their overall survival. We also had a very key secondary endpoint, which is something we called PFS2, which was an analysis of the crossover patients before crossover and after.
We saw a significant improvement in PFS after patients crossed over to drug. I think the most important analysis is if you look at all the patients in the study who got tovecimig at any time, their median overall survival was 9.9 months. Patients who got paclitaxel alone, their median overall survival was 6.05 months. I think a clear effect on overall survival, we also have to put those data into some historical context.
In second-line biliary tract cancer, believe it or not, there are no labeled therapies in the U.S. The only second-line therapies available are for patients with actionable mutations FGFR2, IDH1, HER2, etc. That's a relatively small component of the patient population, maybe 20% total for all of those actionable mutations. The other 80% of patients get a mix of chemotherapy regimens, the three drug combinations, FOLFOX or FOLFIRI. In multiple randomized studies, the median overall survival with those regimens is in the five point seven to six point three month range. Spot on for what we saw with paclitaxel, 6.05, and 50% longer if you got tovecimig in the study. We're very excited about the overall survival results.
I see. Okay. Can you walk me back a little bit, maybe, was that an oversight when you guys first designed the study to allow the crossover and then did the analysis at ITT? What was the rationale behind that-
Yeah.
...design initially-
Sure.
...which became very much apparent that the result didn't matter?
The study was simply undoable without a crossover. The patient advocacy groups would not support the study, and I think even more importantly, scientific review committees at a couple of academic centers in the U.S. declined the study because it didn't have a crossover. The study was simply undoable without a crossover.
I see.
I think it's absolutely the right thing to do for patients. I think, obviously, you know this full well, the RAS inhibitor data that were presented at ASCO last weekend, a P value on overall survival of something like 10 to the - 11th power, right? Do we really need a P value that low?
Right.
I think it's better to offer patients crossover-
Right.
...for drugs that clearly have signals of efficacy.
Basically, the design, I guess the complication related to design is something that you guys already foreseen.
Oh, yeah.
Right. From the beginning.
Oh, for sure.
The results were something that you were expecting?
Oh, yeah.
When you looked at the OS and it did not meet stat sig.
Yeah. It's almost impossible for it to have met stat sig.
Yeah. Okay. Got it. What about the combination safety compared to paclitaxel? Have you heard any concerns about from doctors regarding the grade III AEs?
Yeah, so we have not. The AEs in the study were generally manageable. I think the most common on-mechanism adverse events, so for tovecimig is hypertension, so hypertension is on mechanism for VEGF blockade. There are now published algorithms for managing the hypertension associated with VEGF blockade. We have GI oncologists have more than 20 years of experience with Avastin, so managing Avastin-induced hypertension is just not an issue for docs. On the paclitaxel side, neutropenia, anemia, these are the expected AEs associated with chemotherapy.
I see. Okay, got it. How large is that second-line setting? Also, when you look at this data, how translatable is it to other solid tumor indications?
Thanks. Yeah, those are two very important questions. Last month, at the beginning of May, the Cholangiocarcinoma Foundation, which is the main patient advocacy group, the Cholangiocarcinoma Foundation commissioned a new epidemiology study of biliary tract cancer in this country, and they discovered that there are more than 25,000 new patients annually in the U.S. That's an enormous market. From our work that we've done, we believe approximately 2/3 of those patients will move on to receive second-line therapy, so that's about 15,000 patients annually in second-line BTC in the U.S. alone. It is clearly a multi $1 billion annual commercial opportunity.
The second half of your question is also incredibly important. We believe that the data that we've reported in patients with biliary tract cancer really suggest that this drug could be broadly used in patients with other solid tumors. I think we're going to start with indications where paclitaxel is part of the standard of care. The two I mentioned, gastric and ovarian, so post platinum, patients with ovarian commonly get some form of paclitaxel Avastin.
With gastric cancer, second-line standard of care is paclitaxel plus the VEGF receptor-targeted antibody ramucirumab. Those are two very straightforward places for us to go. I think we're also considering, and we'll come to this, I hope, but we're also considering adding our own PD-1, PD-L1 blocker to that regimen. Tovecimig, paclitaxel 8371 as really a next gen IO angiogenesis combination regimen.
I see. Okay. Got it. You mentioned that you're expecting the BLA filing the second half of this year. Given the complication with the OS, I know you guys are meeting with the FDA, I think in early August. Do you anticipate any complications with the FDA, or do you believe that this is something that the FDA has also been expecting-
Oh, yeah.
...and would overlook-
Yes.
...the OS?
Yes. Look, we hit the primary endpoint of overall response rate. We have a massive effect on PFS with a hazard ratio of 0.44. We've got the lowest hazard ratio in this indication, I think, that's ever been reported outside of targeted therapies. We have an overwhelming treatment effect and hitting ORR, hitting PFS, I just described 80%, approximately, of oncology drug approvals in this country. FDA, among all of our constituencies, I think FDA and KOLs are the most familiar with the confounding effects of crossover.
I see. Okay, that makes sense. You guys also have a 1L study going.
Yes.
Can you walk us through the design for that study, and then how long do you expect that study to enroll and run for?
Sure. That's a great question. The investigators at MD Anderson in Houston were the third-largest enroller in our second-line study. After treating some patients in the second-line study, they approached us with a proposal to add tovecimig to the frontline regimen of gemcitabine, cisplatin, and durvalumab. Now, when we got that proposal, at the time, we were still blinded to all of the data in the study. We thought it was a really nice symbol and hopefully, the folks at MD Anderson had seen something. Now, having talked to the folks at MD Anderson after the study read out, we now know that the folks at MD Anderson had seen some dramatic anti-tumor responses in the second-line setting.
That study is a single-arm study adding tovecimig to the frontline regimen of gemcitabine, cisplatin, and durvalumab. That study's enrolling. Because this was a new chemotherapy regimen, we had a small safety run-in, which I think we're through that now. We just met with the lead investigator at ASCO, and we're going to expand that study out to add some additional sites to speed up enrollment a little bit. Your question about how long it would take, I'm not sure. I think it's probably maybe another year of enrollment, and then maybe late next year we could sort of get a read on the efficacy signal there. We're excited about some of the things that we're seeing so far from that study.
What is your bar for success in terms of both efficacy and safety-
Sure.
...in that first-line setting?
Sure. I think the combination needs to be well-tolerated, which I think we now have evidence that that's the case. gem/cis/durvalumab in the TOPAZ-1 study had about approximately 26% overall response rate, median progression-free survival in the six and a half month range, median overall survival 12.8 months. Those are the bars for the gem/cis/durvalumab regimen. I think we need to see something better than that. The investigators are looking at six-month PFS, as a marker for how we think about moving this drug forward.
I see. Okay. For the rest of the conversation, I want to start moving into the other exciting assets that you guys have. Let's kick it off with CTX-471. You guys are expected to enter a phase II basket study using a biomarker by the NCAM and CD56 positive tumors by mid-year. What is the status for the study, and how many patients are relevant with this biomarker?
In the phase I study of CTX-471, we collected pre and on-study biopsies, and in that study, again, as a monotherapy in the post-checkpoint patient population, we had responses in patients with small cell lung cancer, melanoma, and mesothelioma. We did an analysis of biopsy specimens, and we discovered that the patients who had clinical benefit from the drug had tumors that were NCAM positive. Our most dramatic responder in that study was a patient with metastatic small cell lung cancer who received frontline chemotherapy plus atezolizumab, then second-line therapy with nivolumab, and was treated in the third-line setting, where realistically, that patient's expected median overall survival would unfortunately be measured in single-digit weeks.
Maybe nine, 10 weeks maximum. That patient had a deep response, was on drug for more than three years when the patient had a PET-negative CR. We looked at that patient's tumor, and it was very highly positive for NCAM. That made us look at other patients' tumors. We presented this data at SITC about 18 months ago. We showed that there's a correlation between response and NCAM positivity. That has sent us down this road.
The study design is finalized, submitted to FDA. We should be beginning that study in the coming quarter. In terms of your question about how many patients are out there's a significant patient population. If you add up neuroendocrine tumors, neuroendocrine carcinomas, melanoma, and small cell lung cancer, it's probably in the range of 60,000 patients annually in the U.S. It's an enormous potential market. That basket study is going to enroll NCAM-positive patients.
I see. What shows the selection of the tumor types in this basket trial? Of all these different indications that you're studying, which one do you think that you have the most conviction on?
Sure. The tumor types came out of some validation work that we did looking at staining tumors for NCAM. That's how we selected the indications, and also selecting indications where we saw efficacy signals in the phase I study. I think that's a very interesting question the way you phrased it, for sure, that obviously we're super excited about small cell lung cancer just based on what we saw in the phase I study. If we could show in neuroendocrine tumors and neuroendocrine carcinomas that we had an efficacy signal there, that is a massive unmet need, and I'm really excited about going into a novel tumor type there.
I see. Okay. Got it. Any guidance on when data would be available for the trial?
Sure. Study's going to get up and running in the next quarter. I think probably mid to late next year, I would think.
I see. Okay. Will you go straight into a registration trial at that point?
Well, if the data supported it, yes, for sure.
I see. Okay. You would pick a lead tumor to go into, or is it one of those that a registrational program would basically be a tumor, a biomarker lead?
I would love it to be biomarker lead. I would love it. Right. That way we could get a very broad tumor agnostic label, which I think would be very interesting if we could do that.
I see. Okay, great. Let's move on to the next asset, which I think is very interesting, the CTX-8371, which is the dual PD-1, PD-L1. Let's talk about the mechanism. How does it work? Why does dual PD-1, PD-L1?
Again, this drug came out of some very comprehensive work that we did in research and asking the question, what is the best combination partner for PD-1 blockade? Honestly, I think we were a little surprised that PD-L1 emerged as the best combination partner for PD-1 because we spent a long time investigating the mechanism of action, and we actually published all that data a couple of years ago. Clearly, the drug does block both the ligand and the receptor. There's no question about that. It does more than that. It's definitely a cell engager. It connects PD-1 positive cells to PD-L1 positive cells.
I think the most fascinating thing we discovered is that 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. It's an incredibly unique mechanism of action, and it's why we had confidence going into the post-checkpoint patient population.
Right. Which leads to my second question. In the post-checkpoint setting, right, patients already seeing checkpoint inhibitors. You basically administered the drug, which is another checkpoint inhibitor.
Yep.
You saw activity there.
Yes.
How reliable is that data? I think you guys presented some data at ASCO. Some imaging data as well.
Yeah.
Maybe just walk us through the data.
Sure. I think how reliable are the data? We had confirmed responses in patients with triple-negative breast cancer, Hodgkin's lymphoma, and non-small cell lung cancer. I think the scans that we showed in the ASCO poster, which is available on our website, those scans really speak for themselves, I think, right? The patient with triple-negative breast cancer who relapsed while getting adjuvant KEYTRUDA, then got TRODELVY +2 other chemotherapy regimens, that patient had almost nine centimeters of linear tumor burden, including a five-centimeter metastasis to her pericardium lining the heart. All of that tumor burden has disappeared.
The patient is now in a durable response for over a year in the post-TRODELVY setting. Similarly, the patient with Hodgkin's lymphoma, who happened also to be post-transplant. We showed PET scans for the first time at ASCO last week. That patient had a deep metabolic PR post-checkpoint, the non-small cell lung cancer patient similarly had dramatic decline in their linear tumor burden. You didn't ask this, you asked me how reliable the data are, but the real question is why are we seeing this, right? We believe the answer to that's incredibly simple. CTX-8371 is a better checkpoint inhibitor, period.
We have, I know this sounds like I'm trash talking, so stay with me, but we have the best phase I checkpoint inhibitor data ever. There's no other checkpoint inhibitor, which by the way, includes KEYTRUDA, that has better phase I data than we do. We have more responses in fewer patients, and our data are in the post-checkpoint population. All of the early checkpoint inhibitor phase I data were in checkpoint-naive patients. We simply believe we have a better checkpoint inhibitor.
I see. Okay. Makes sense. What is that potential of CTX-8371 in combination with your other two agents that we just talked about? I think you mentioned that.
Yeah. Love it.
You're exploring-
Yep.
...development with these different combination. Maybe just walk us through what are you guys thinking there?
Sure. Love it. I think in the past maybe five years, I think one of the more important advances in oncology is the combination of checkpoint inhibitors with angiogenesis inhibitors. From my point of view, the first study that showed this was the Roche study, IMbrave150, which was bevacizumab versus sorafenib in frontline hepatocellular cancer. That showed that the combination of VEGF targeting and checkpoint inhibition was synergistic, and that now spawned a whole set of studies, combination of checkpoint inhibitors with VEGF kinase inhibitors.
Really, from my point of view, people aren't going to love this, but from my point of view, the bevacizumab study is what got the PD-1 VEGF bispecific field going. We would love to have a sort of super combination, which would be tovecimig, a dual angiogenesis inhibitor, and CTX-8371, a dual checkpoint inhibitor. Put those two together, that could be the next generation of angiogenesis checkpoint inhibition combination. We're going to test that. We've got some designs on the table right now in gastric cancer and biliary tract cancer. Just thinking about sort of where we could go with that would be a very wide field for us to go after.
Fantastic. We only have about a minute left, I want to talk about the last asset-
Sure.
...the fourth one, which is your PD-1 VEGF bispecific. This is a very crowded field.
Yes.
We saw data at ASCO from Summit, and then there's many other companies-
Yep.
...developing something similar. How do you think about your asset and how do you position it among all the other companies?
Sure. We developed that drug ourselves. We know that drug has better PD-1 blockade than ivonescimab. We did head-to-head preclinical studies, our drug, CTX-10726, is superior to ivonescimab in some preclinical studies. We believe we have a novel, better drug. We're going to test that. We selected our phase I indications carefully. We wanted to go into indications where both checkpoint inhibition and angiogenesis inhibition have already had demonstrable efficacy signals.
We're going to go into patients with hepatocellular, renal cell, gastric, and endometrial cell cancer. Each of those four indications has either PD-1 or PD-L1 blockade or VEGF targeting or VEGF receptor targeting as approved therapies. Hepatocellular, bevacizumab, renal cell, CABOMETYX, frontline regimens of the combination. Those are our four phase I indications, and that's how we think we can differentiate development. Phase I is underway. First dosing cohort fully enrolled. Should be moving to the next second dosing cohort shortly.
Fantastic. A very exciting time for Compass Therapeutics indeed. It's been a pleasure hosting you. I'll pass it back to you for any final remarks.
Well, thank you very much again. Thank you for inviting us. I'll go back to the very first question you asked, which is, I think the next 12 months are going to be incredibly exciting for us with multiple catalysts for each of our four drugs in the clinic. Looking forward to talking to you next year.
Great. Thank you.
Thank you.