Okay, great. Thanks everybody for joining us here on day two of Piper Sandler's annual healthcare conference. I'm Joe Catanzaro. I'm one of the biotech analysts here at Piper. It's my pleasure to kick off the day here with CytomX and their CEO, Sean McCarthy. Sean, thanks for joining us. Maybe before we jump into Q&A, I could give you a minute or two. You could sort of introduce CytomX, let everybody know what you guys have been up to, what we have to look forward to.
Yeah, thanks, Joe. Pleasure to be here again. And so CytomX, we are really the pioneer in this whole field of what we call conditional activation or antibody masking, which is a strategy to take therapeutic antibodies and localize them into disease tissue. We're focused principally on oncology to improve the therapeutic window of anti-cancer drugs across a wide range of modalities. And I think it's a really exciting time for this field where it is a field that I think we're widely regarded as having initiated, having pioneered, and we are seeing others beginning to make progress as well. And I've always said that we're onto something with this strategy, and I think that is broadly true, that localizing therapeutic antibodies really does have the potential to be the future of biologics. And CytomX is at the absolute forefront of this field.
So what we're really focused on right now as a company are three clinical programs, which continue to really push the boundaries of our platform in a multi-modality setting. Our lead program is CX-904. This is a masked T-cell engager targeting EGFR and CD3. That's partnered in a global co-development relationship with Amgen. Our second program is an antibody-drug conjugate. It's a masked anti-EpCAM ADC. We see EpCAM as an incredibly high potential target. We'll say a lot more about that later on. That's also in phase one. The third program, which we actually entered into the clinic this year, is yet another application of our technology focused on masking the very powerful immune modulatory cytokine interferon alpha-2b. That program entered phase one just a couple of months ago.
We continue to push forward, taking learnings from now more than a decade of work on this platform with these three really exciting clinical programs and building towards what looks like it will be a very exciting 2025 for us with a number of key milestones across that clinical pipeline.
Great. So I think you mentioned external validation of the idea of masking antibodies. I do want to ask about it, but maybe first we could focus on internal validation. We've got a little bit more data this year. When you look back across the sort of broad set of clinical data that Probody has generated, what do you see as sort of the most validating point for the platform?
Yeah, CytomX really has achieved every first in this field. If you go back to some of the very early work we did with our anti-PD-L1 Probody, we call our particular approach to masked antibodies, we call Probody Therapeutics. And just to recap what that exactly is, we take an antibody, we engineer the antibody to have a peptide mask that blocks the ability of the antibody to bind to the antigen until the mask is removed. Mask removal is achieved specifically and selectively in cancer tissue by tumor-associated proteases. We've known for decades that tumor proteases are fundamentally dysregulated across a wide range of different tumor types. And so we rely on that fundamental property of tumor biology to clip off the mask, allowing the antibody to bind target in tumor tissue, not in the periphery, and thereby expanding the therapeutic window.
So what we've been able to do is show with not just the PD-L1 program going back a few years, but also our first-generation ADCs that we put into the clinic, that we can measure that masking is maintained in the blood of cancer patients. Important first box to check. And I can't stress enough how important these first experiments were in really defining this field and actually enabling others to enter and do great things for patients. So masking is maintained in the periphery of cancer patients. Masks are removed. We can see that in on-treatment biopsies. We see, in the case of that program, more than 90% receptor occupancy by the activated drug, showing that we do see sufficient activation to engage target. And the target engages the appropriate pharmacology with robust anti-tumor effects in the clinic across multiple targets.
So that's the foundation that we've laid at CytomX, both in terms of clinical data that we presented. We've published a broad-based intellectual property portfolio that protects this technology. And as I said, we're really excited now to see this field moving forward across multiple programs, really across the industry.
I want to focus maybe first on masked T-cell engagers and maybe T-cell engagers more generally. Maybe ask a two-parter. So one is when you look at the field of T-cell engagers, some of the challenges it faces, what can a masked approach potentially offer and solve for? And then the second part of that is the external validation. We saw a pretty interesting data set earlier this week for a PSMA masked bispecific. When you look at that data set, what sort of points of validation do you see there that support this idea that masking carries tremendous value?
Yeah, we totally see T-cell engagers as one of the killer applications of Probody technology at CytomX. And our work on the EGFR CD3 program has really demonstrated the potential power of this approach. And so why do we need to mask T-cell engagers? Well, there's enormous potential if we can harness the power of this modality in solid tumors. The challenge, of course, is that T-cell engagers, and I'll focus specifically on CD3 T-cell engagers, are extraordinarily potent molecules. I mean, the doses that we're using in the clinic are fractions of the dose that would be typically given of a therapeutic antibody or an ADC. And because of their potency, there's very little room to maneuver with target expression on normal tissue. And that's been a major limitation for the field in moving T-cell engagers into the realm of solid tumors. We have seen progress, of course.
Amgen has achieved the first approval with the DLL3 targeting T-cell engager, tarlatamab. That's a relatively clean target to express at high levels on small cell lung cancer, relatively low in normal tissue. They've demonstrated a therapeutic window. That drug is now approved. That's a big step forward for the field. But there are really very few of those clean targets. So what we're trying to do at CytomX is we're trying to break through by applying our technology to more broadly expressed targets that have powerful anti-tumor potential like EGFR. And earlier this year, we presented our initial clinical results on the first 35 patients treated with CX-904, our EGFR CD3, showing quite honestly a pretty remarkable safety profile given masking on the CD3 side of the equation.
Also, a very attractive safety profile with regards to EGFR engagement, low rates of systemic EGFR toxicities, and clear evidence of single-agent activity in the form of, in that data set, a couple of confirmed responses in patients with pancreatic cancer, which we're really excited about. So this is the type of work that we're doing at CytomX. We're also working across multiple partnerships in the field of T-cell engagers. In fact, most of the work that we're doing right now in our partnerships with Amgen, Astellas, Regeneron, to name just three, is focused in T-cell engagers. And as far as data this week, it's been a pretty big week for the field. It's exciting to see the continued evolution of the PSMA-directed CD3 that we've all heard about over the last couple of days. I think that data is encouraging.
I think it's showing that, again, the power of conditional activation and masking in unlocking new targets or improving therapeutic window where there's room for improvement and doing great things for patients. And so we've never been more excited about the work that we're doing and actually more broadly the impact that we're having with the field broadening as well.
So maybe sticking with CX-904, the recently updated guidance around sort of the strategic update that we had been anticipating and sort of now a 2025 event, maybe just talk us through sort of what drove that sort of slight shift in timelines and where the program sort of currently stands?
We're all learning a lot about the development strategies for T-cell engagers in solid tumors. These are not trivial drugs to develop. It takes time. You have to start at very low initial doses in the clinic. So the short answer is it just takes time to get to your RP2D. And we've been successful in continued dose escalation with 904. And as of the May update, we had not reached MTD. We had not declared RP2D. And that continues to be the case. And so working with our partner Amgen, it just makes a lot of sense to continue to push dose to maximize dose. We do believe that T-cell engagers do show a dose response in the clinic.
That dose response is a little less clear than you would typically see for a conventional antibody or for an antibody-drug conjugate because, of course, we're engaging the immune system and T-cells to actually act as the effector mechanism for the entire therapeutic strategy. So dose response for T-cell engagers is complex, but we do believe there is one, and we just want to make sure we maximize dose before selecting RP2D and moving into phase I-B. So for that reason, we're giving ourselves a little bit more time. And our key goal for that program right now, in addition to driving continued dose escalation to get to that RP2D, is to, of course, continue to engage our global development partner Amgen in dialogue as to what the next steps for the program will be.
We are currently focused in three tumor types: pancreatic, of course, where we have demonstrated initial single-agent activity in the form of two confirmed partial responses in that initial 35-patient data set across multiple tumors. Two out of six efficacy-available patients showed confirmed PRs at that time. We also continue to enroll patients in head and neck and in non-small cell lung. Those are our three areas of focus right now. Our goal in 2025 is to reach RP2D and make a decision on moving phase I-A to phase I-B in one or potentially more of those tumor types.
So maybe sticking with this notion of identifying RP2D, I think you guys noted that you've cleared the 15-mg target dose level, still dose escalating. So what goes into assessing whether you've reached an RP2D? Is it the goal to reach an MTD, or are there other sort of considerations of sort of picking this is the phase I-B dose that we're going to move forward with?
Yeah, great question. And it's really a triangulation of multiple data points. So obviously looking at PK, PD, safety, anti-tumor activity, and rolling that all up into an overall assessment of what is likely going to be more than one dose that we would take forward into phase I-B. Amgen really set the standard for how to develop a T-cell engager in solid tumors. They took two doses into phase II. They selected one of those doses then to do a further expansion. And it's a 10 mg dose that they have registered and is now being utilized commercially. So we will take multiple factors into account as we make that decision. But most likely, I'd be surprised as and when we make phase I-A, phase I-B transition, I'd be surprised if we're not still evaluating more than one dose.
Yeah. And I want to ask about CRS. So far, the program has been remarkably clean on cytokine release. Are there any expectations as you get to maybe higher target doses that you should eventually CRS as these sort of immune responses mature, become maybe a little bit more systemic? Just maybe some overall thoughts on how you sort of see that bigger picture.
Yeah, the successful masking of CD3 that we've achieved so far, which has enabled us to deliver this pretty remarkable safety profile in doses evaluated to date, actually gives us a lot of room to maneuver in terms of pushing dose. And so that's another reason that we're continuing to escalate. We know that, obviously, if you look at the cell therapy field, there's enormous tolerance for toxicity in the way that those drugs are used. That's probably going to be similar for T-cell engagers because at the end of the day, it's all about driving robust and durable anti-tumor responses. And as with any immunotherapy, I think we would all anticipate that over time, we just get better at managing complex safety profiles.
So we think that the foundation that we've laid with that first 35 patients of data, particularly on the safety side, and not just on CD3, also on EGFR, where we have a very manageable EGFR safety profile, gives us that room to maneuver and continue to push the dose to maximize clinical benefit for patients in the form of durable responses. So that's kind of holistically how we think about it.
Last question, and I do want to move to CX-2051. Any updated thinking on why pancreatic cancer has thus far been the most responsive tumor type?
Not really. I think that we continue to believe that the data that we generated is consistent with observations that were made actually around that same time by others, including data on the Claudin 18.2 T-cell engager, which has shown also a pretty interesting response rate in PDAC. We're never afraid to challenge dogma at CytomX. And if the dogma is that pancreatic doesn't respond to immunotherapy, well, okay, let's do more work and try to understand that. We're strong believers in doing the pharmacology and then drawing conclusions from that.
Great. So let's move to your EpCAM ADC program CX-2051. Maybe first, just sort of talk through how the design of this ADC is maybe different than your first-generation approaches with CD166 and CD71.
Absolutely. So really excited about CX-2051. So just to briefly recap the design of the molecule. So it's an anti-EpCAM antibody that we have masked using our Probody therapeutic technology, so a peptide mask, protease cleavable linker. Important to state that the protease cleavable linker is a precedented one that we've shown previously is activated in tumor biopsies. So we feel very confident about the protease cleavable linker being able to be activated in tumors. The payload is a topo-1 inhibitor. We call it CAMP59. We licensed it from ImmunoGen. Think of it as functionally equivalent to DXd. The drug-antibody ratio is eight. And we have shown in our preclinical results that the efficacy of the masked ADC is equivalent to the unmasked in colorectal cancer models. I'll come back to CRC in a moment.
We've also shown that the payload, CAMP59, is equivalent in potency in the ADC context to DXd. So we think we've designed a really interesting first-in-class antibody-drug conjugate here against a really interesting target. So EpCAM is actually one of the first tumor antigens actually that was ever described and cloned. And it was first identified on colorectal cancer. It's expressed on CRC at levels that are equivalent to amplified HER2 in breast cancer. So there are millions of copies per cell on CRC of EpCAM. Given the unmet medical need in CRC, given the responsiveness of CRC to topo-1 chemo in the first and/or second-line setting, we strongly believe that the right first experiment to do with this molecule, and in fact, the reason it was designed this way, is in colorectal cancer.
So we're escalating specifically in CRC, in late-line metastatic CRC, and really excited to see what this drug can do for patients over time.
How do you think about the receptivity of EpCAM to a masked ADC approach given its expression profile relative to priors with CD166 and CD71?
Yeah, so the first targets, CD166 and CD71, which were the targets of our first-generation ADCs, we learned a lot from those programs. I think, again, it's hard to kind of overstate the importance of learning in a field like this. If you're really blazing a trail with a novel technology like we have, you learn from those first programs. And in the case of both of those programs, we actually achieved a lot. I mean, we did open a therapeutic window for both of these broadly expressed tumor antigens. We didn't see exactly what we wanted in the late phase I or early phase II setting to move into registrational studies, but sometimes you've got to take steps back to step forward. But those first two programs really encouraged us that masking can open a therapeutic window for very broadly expressed targets.
We achieved meaningful response rates in very difficult-to-treat tumor types. We just didn't quite get the window that we wanted to make those $50 million investments in late-stage development, $50 million +. We decided to go, like I said, take a step back to step forward and work on EpCAM, so why EpCAM? Well, as I said, not only is it very highly expressed in CRC, it's also been a target that a number of groups have tried to drug over time and got close. We know that anti-EpCAM tox in fusions can shrink tumors in patients. They need to be given locally. You cannot give an unmasked anti-EpCAM antibody systemically. Why not? Because EpCAM is present in many, as the name suggests, epithelial cell adhesion molecule. The targets are expressed in most epithelial structures.
Interestingly, the toxicities that have been seen previously with anti-EpCAM strategies are pancreatitis and GI toxicity and liver toxicities. We're very clear about what we're looking for in this phase I study. Again, focused in CRC with dose escalating, we're looking to mitigate and minimize those previously described classic EpCAM toxicities like pancreatitis, liver tox, and diarrhea. We do expect to see payload tox. We're not masking the payload. We know all ADCs ultimately hit payload toxicities at MTD. I expect that to be the case with CX-2051. Those toxicities typically for a topo-1 inhibitor are upper GI, nausea, vomiting. That's in the label for Enhertu. We kind of anticipate seeing that. We anticipate seeing cytopenias at the higher doses as we push to MTD.
Once we get to MTD, of course, that will provide the framework and the foundation for selection of RP2D. We've made a lot of progress already in the clinic. As of our last update on our last earnings call, we only began the study in Q2 this year. We'd already reached our fifth dose level. We predict that the dose level that we're at is already a dose level that would not be tolerated if it was a conventional ADC based on modeling of everything that's gone before and our preclinical modeling of ADC versus masked antibody drug conjugate. We're encouraged by our findings so far. This is a wholly owned program, and it has a lot of potential to really drive a ton of value in CytomX in the relatively near term.
We're guiding towards an initial phase I dataset in the first half of 2025.
Yeah, so maybe on that last point there, maybe you could sort of frame for that initial update your sort of hope of what you're going to be able to show, what you're going to be able to say about the program, and what the next steps are thereafter.
Yeah, I mean, it'll be a first step in man. It'll likely be a relatively early dataset, just to manage expectations. But we are obviously laser-focused on characterizing the safety profile of CX-2051 as it relates to, as I just said, our goal of mitigating these classic EpCAM on-target toxicities, our goal of assessing the anticipated payload toxicities, and of course, looking for early evidence of anti-tumor activity, which is in what is most likely going to be the fourth line or later CRC setting, where the bar is very low. So in that setting, in the third line setting for CRC, current standard of care, single-digit response rates, PFS in the two- to five-month range. So there's a lot of room to deliver value here with CX-2051 in this CRC setting.
We're looking forward to presenting, again, that initial first look at this very exciting molecule in the first half of next year.
So you mentioned CRC. What should our expectations be in terms of how enriched that initial dataset will be for CRC? Is it near 100% of patients or close to it? And then are there any other tumor types where EpCAM could have some potential utility?
Yeah, I'll take the second question first. One of the reasons that we love EpCAM as a target, and one of the reasons that people have been trying to drug it for such a long time, is it really is a pan-tumor target. EpCAM is expressed in most solid tumors. It's particularly highly expressed in CRC, but it's also expressed in many, many other solid tumors. And so in the long run, CX-2051 has pipeline and a product potential. In terms of the dataset, I do expect it to be almost entirely CRC. So we've been quite deliberately enrolling in that subset in that patient population. It's where the drug is designed to really show us what it can do. And the other advantage of CRC is more than 90% of patients express EpCAM at high levels.
And so there's no requirement, there's no need to prospectively select patients, which is one of the reasons that we've been able to make such rapid progress in the clinic. We will retrospectively have data on target expression, of course, in this patient population, but it hasn't been necessary to select. And that's really allowing us to move swiftly to get an initial handle on this drug candidate in the clinic.
So last question. So I know Lilly has started to incorporate this camptothecin payload, I believe, into their ADCs.
That's right.
But your CX-2051 will probably be the first clinical dataset.
Correct.
That will generate for this payload.
That's right.
Perfect. Well, with that, Sean, thanks so much for your time. Always enjoy our discussions. Thanks, everybody, for tuning in. Enjoy the rest of your day here. Take care.