Great, thanks, everybody, for joining us here the second day of the Piper Sandler Annual Healthcare Conference. I'm Joseph Catanzaro, one of the Piper biotech analysts. It's my great pleasure to kick off this next session with CytomX Therapeutics. Joining us is their CEO, Sean McCarthy. Sean, thanks for joining us. Obviously a bunch of topics I want to try and run through in these next 25 minutes or so, but maybe first I could give you a couple of minutes. You could introduce CytomX, what you guys have been up to, and what we have to look forward to heading into the new year.
Yeah, great. Thanks, Joe. Thanks for having us at the conference. Always a pleasure to be here. So at CytomX, we are developing a multimodality, broad, deep pipeline focused on oncology and all built around our Probody therapeutic platform, which is the leading platform for the localization of potent biologics into tumor tissue. It's a very exciting time for the company. We've had a very busy year of execution this year, looking really over the 2024-2025 time horizon to a lot of developments in the pipeline, beginning with CX-904, our EGFR- CD3 bispecific, which is making good progress in phase I, for which we anticipate initial phase I-A data in the first half of 2024.
The team is working very hard right now to file two new INDs for two wholly owned programs, CX-2051, which is our EpCAM targeting antibody drug conjugate, and CX-801, which is our first move into the clinic with a masked cytokine. This is a masked version of interferon alpha-2b. And then we're also continuing our work with Bristol- Myers Squibb on the CTLA-4 program and looking ahead to potentially data from them in 2024 on what we call 288, which is the masked version of a non-fucosylated version of ipilimumab. So, a lot, a lot going on.
We can talk maybe a bit more about the partnerships, a bit later, but, as we come to the end of 2023, we're certainly setting our sights on 2024 and, and into 2025 as being important, an important time period for the company.
Great. So maybe with that, I can ask you a question that I think I sort of surface to you every year or so. Sort of the notion that CytomX and Probody, I think, remains the masking platform with the biggest clinical data set in its hands. So when you look back at that data set that you've generated with Probody in the clinic across multiple programs, like, what do you pull out as sort of the most validating data points that tell you, "Hey, you know, this masking technology is doing what we say it does."?
Yeah, it's also a very interesting time in that regard for us, in that we're beginning to see data sets from others who have entered this field of masking or conditional activation, you know, pro-drug strategies for biologics. And I would say that, you know, for CytomX, it's really been a story of firsts. So you know, we're the first company to have conceived of and built this strategy, first company to have checked a number of really important technical boxes in terms of the development of the platform in the clinic, showing that masking can remain stable in the circulation of cancer patients. Masks can be removed in tumor tissue to levels that can create at least a tenfold differential between how much intratumoral activation we see relative to plasma activation.
That activation in the tumor has been accompanied by tumor shrinkage across multiple tumor types, which is a function of our multi-selective protease substrate strategy. Also we've seen evidence of decreasing, in some cases, really minimizing systemic toxicities and on-target binding as a function of the masking remaining intact. Now, we've moved multiple programs into the clinic over the years. We've learned a lot from them, including our PD-L1 program, which really was the workhorse that allowed us to check those boxes for the first time. Our CD166 and CD71 ADCs, which we also learned a lot from.
We're now taking the learnings from those more than 500 patients that we were able to evaluate with those drug candidates into the current pipeline, which really are the next generation of programs, the EpCAM program, the EGFR- CD3, and the interferon program, which really incorporate learnings. The way we think about that is that the... You know, you want to always be increasing probability of technical success for every new program that you put into the clinic, and I think that's very much what we've been able to do.
Great. So maybe with that, and maybe before we jump into specific programs, I want to sort of cover a couple of high-level questions on the potential to leverage Probody within the context of T- cell engagers and other bispecifics. And so when you look at this bispecific field, and there's been a couple approvals in BCMA and the liquid tumors and some data generated in solid tumors with unmasked bispecifics, where do you see the challenges with those unmasked T- cell engagers bispecifics that Probody maybe can readily address?
Yeah, we really think we're in the right place at the right time with our technology as we think about the potential for bringing T-cell engagers to solid tumors. So, one of the... The two principal limitations in this area in solid tumors are looking to minimize CRS and also minimize on-target toxicity. And the thing about T-cell engagers, of course, is that they're so potent. You know, we're treating patients with micrograms per kilogram rather than the typical milligram per kilogram with a typical therapeutic antibody. They're so potent in their ability to activate the immune cells, that they're relatively unforgiving in terms of causing damage to normal tissues when the target is present on a normal tissue.
So we think that to really address broadly distributed tumor antigens, which more often than not are also present on normal tissues, our strategy for masking can open therapeutic window in a meaningful way and bring T-cell engagers across multiple tumor types. Now, the broad field of T-cell engagers has made some big advances this year, notably at ESMO. We saw really important datasets from Amgen and Regeneron. Amgen, most notably the DLL3 tarlatamab phase II data, which builds on a multi-year program against the DLL3 target, which I think is the most convincing data so far that T-cell engagers can break through in solid tumors. Now, DLL3 is a relatively selective target, and it's quite specifically expressed in small cell lung cancer.
That's where the activity has been seen. They're now advancing into a registrational study. So I think it shows mechanistically that T-cell engagers can work in solid tumors. The types of targets that we're going after, like EGFR, are more broadly distributed, and that's what our technology, we think, is well suited to addressing. So we, as I said, I think we're in the right place at the right time.
Yeah, maybe sort of looping in your extensive and notable collaboration. So the last three recent collaborations, Amgen, Astellas, and Regeneron, have been around bispecifics and T-cell engagers, and, you know, one thing I always wonder is whether this is the antibody modality where Probody is most well suited. And so I'm wondering how you think about that. Does it come from the fact that something you said, where these engagers are so potent that Probody unmasking does it need to be terribly efficient and you still get some good activity? I guess, how do you think about that in terms of Probody in this context versus antibody and ADCs?
Well, there's no question that this field has attracted substantial partnerships to the company. You mentioned Amgen, Astellas, and most recently, Regeneron. We're absolutely thrilled to have these partners and taken together with the work that we're doing with those three companies and also our own work, we do have a very broad program now in T-cell engagers. We do think this has the potential to be a sweet spot for the technology, but I wouldn't by any means, you know, minimize the potential, you know, the continued potential of the Probody platform in ADCs, and in cytokines. I think that we've learned an enormous amount about ADCs, and we can talk more about the EpCAM program in a moment, perhaps, and how some of those past learnings are integrated into the EpCAM program.
But yeah, I think we're excited about T-cell engagers, but we continue to view the Probody platform as a multimodality opportunity to enable us to have the broadest reach we possibly can across many different cancer types.
Maybe the last high-level question before we dig into CX-904. There were some recent data, maybe the first data for a masked platform outside of Probody, a T-cell engager in solid tumors. Sort of any insights into that data set, what it suggests about sort of the potential benefits, the potential risk around, and challenges around leveraging a masking approach within TCEs?
Yeah, I think the data that we've seen so far from others, you know, beyond CytomX, is encouraging. I think that it's consistent with what we've observed in the clinic, which is that masking, you know, can allow for certain dosing strategies that you might not be able to pursue with naked, unmasked antibodies. I think specifically, though, in the field of T-cell engagers, there's a lot we need to learn here, even for unmasked T-cell engagers, but certainly for the masked strategies, including what is the precise etiology of CRS in the context of solid tumor therapy? We know a lot more about that in the liquid tumor setting, where tumor burden seems to correlate pretty directly with CRS, and tumor lysis syndrome and such like.
In solid tumors, it's not so clear, at least to me, you know, where... From data we've seen publicly, where CRS initiates, is it a function of, you know, as has been postulated, is it a function of, response, so antitumor responses is associated with CRS, or is it something more systemic? I think we have to, you know, take our time to understand that whole, that whole area. And then there's the on-target toxicities. In the case of some of the data that's been put out already, it's principally on PSMA. The PSMA is a fairly restricted target. It's not a target, that has, you know, particularly challenging, normal tissue toxicity.
In our case, we're taking a different strategy at CytomX with EGFR, where, you know, what we'll be looking for in phase I, in addition to managing and mitigating CRS through masking strategies, is looking to manage and mitigate the predicted toxicities with an EGFR- CD3 strategy that would may be expected to include skin toxicity, GI toxicities, and such like. You know, the canonical therapeutic antibody-mediated EGFR toxicities. So there, there's a lot to unravel here, but I would say, you know, to more specifically answer your question, I think we're encouraged by the data that we've seen so far on from others on masking of T-cell engagers.
Maybe now shifting specifically on the EGFR- CD3 bispecific that you're working on in collaboration with Regeneron. Could you maybe sort of talk through the design of the molecule? And I guess I'm most interested in whether there were any learnings from, you know, CX-072 and the CD166, CD71 ADCs that were incorporated, whether it's, you know, the protease sites, where they are, which proteases you try and leverage, things like that, that you sort of plugged in here with CX-904.
Yeah. So, with all of our programs, we've continued to take a tuning approach to the design and optimization of our clinical candidates. And so we think of the substrate, the protease cleavable substrate and the mask, and then associated linkers as a pro domain in its entirety. And so one way to think about this is optimization of the pro domain, which incorporates optimizing the masking affinity as a function of the protease cleavability. And as time has gone on, we've continued to build and build and build our repertoire of protease cleavable linkers to broaden their activity across multiple tumor types. This has been a core strategy of ours from the beginning, and so, you know, CX-904 was optimized with these strategies in mind. So we mask on both CD3 and EGFR.
We think that's important. We think the masking on CD3 is an important strategy for minimizing CRS. We think masking on EGFR is critical for minimizing these, as I said, these predicted EGFR-mediated toxicities. And then the protease cleavable linkers have been selected based on preclinical optimization to get the broadest window in the context of the broadest antitumor activity. So we feel, you know, that we've put the right program in the clinic, and that's the strategy that we'll continue to pursue across all of our programs.
So I know you guys are guiding towards initial data disclosure in the first half of 2024. I guess currently, what's the status of the trial? Where are you in terms of potentially identifying a dose and schedule? You know, what's the early sort of indications you have there?
Yeah. So the goal for the program, working with our partner Amgen in 2024, will be to assess our phase I-A dose escalation data in collaboration with Amgen, and work to defining our phase I-B strategy, with phase I-B being tumor-specific expansions in EGFR-positive tumor types. And so, you know, no decision made there yet on what those tumor types will be. We'll make that decision in conjunction with our partner. So leading up to that, what data are we generating in phase I-A? Well, as you said, we're dose escalating, we're exploring dose, exploring schedule. We are laser focused on safety in this early part of the clinical program.
And if you look at, again, the roadmap, the Amgen data from ESMO is very helpful in particular because it gives you a sense of the roadmap as to how you develop a T-cell engager, and how important phase I is to really assess safety before you make those decisions about what your dose and, and likely more than one dose would be that you would take into phase I-B. Because we also, of course, have to take FDA's Project Optimus into account now as we develop assets like this. So, what we're particularly focused on in phase I-A is escalating, assessing safety, assessing the CRS profile with this drug, assessing the overall EGFR-mediated toxicities. Obviously, as we get to doses in the predicted therapeutic range, beginning to look for signs of tumor stabilization, potentially signs of tumor shrinkage.
I would emphasize that in the phase I-A setting, in contrast to where Amgen and Regeneron have gone with their DLL3, STEAP1 and MUC16 programs, where they were tumor selective, even in phase I, we're taking a broad approach in an EGFR all-comer setting in phase I-A. So, really, you know, to get initial experience with the drug candidate, it'll be specific tumor types in phase I-B that we expand into. But again, what we're looking for in phase I, as I said, is safety, any evidence of initial tumor stabilization or shrinkage that we think would be very encouraging, pointing the way to potential phase I-B.
So I wanna follow up in a bit on the sort of potential strategy in the phase I- B, but maybe first sticking with the phase I- A. I think you guys have noted that you've started backfilling cohorts.
Yeah.
What was the sort of trigger to do that? Is it sort of PK/PD driven? Is it safety driven? Sort of what sort of gives you, you know, the notion that, "Hey, we should enroll more patients at this dose."?
Yeah, a combination of factors. Obviously, safety first. We are, as is typical for these programs, following a wide range of, you know, pharmacodynamic markers to get a sense of the molecular performance of the drug candidate, looking closely at PK, as well. And so all of that has gone into... together with our QSP modeling of, where the computational modeling predicts one would be in a predicted potential therapeutically active range. And so those factors all interdigitate into that decision to just start backfilling, which we have now started to do, in select cohorts.
So as we look to the disclosure in first half 2024, I know you've emphasized it's primarily gonna be a lot of emphasis on, on safety. I guess one thing I'm wondering is whether we have a good reference point for, like, CRS. So, like, what would be a good indication that like: Oh, yeah, the Probody TCE is mitigating CRS risk. I'm not sure if we have, like, a good comp to look for, of what, like, an unmasked EGFR CD3 would look like.
Yeah, there is no, there really is no solid comp to point to. I think that, you know, we're all learning in this space. I think the simple thing to say is we're looking to minimize CRS to the maximum of our ability. So, but, yeah, I think we'll cross that bridge when we come to it, when we present and discuss the data.
When you sort of discussed the phase I-A with Amgen, what are some things that need to be discussed? What are some key questions? I know we've talked about sort of phase I-B strategy. Are there other important sort of factors that are gonna be on the table on how to take this program forward?
Well, I think one of them will be what cancer types to move forward in. Obviously, you know, Amgen has its own, commercial strategy and preferences, and, that will be an important conversation that will integrate everything that we've learned from the early clinical program together with, you know, the evolution of the competitive landscape and what we're seeing from others. So I think that's probably one of the biggest things.
Great. So maybe in these last five minutes or so that we have here, we could touch on the rest of the pipeline and the two INDs that are coming before the end of the year and start with CX-2051, the EpCAM program. Maybe first question, I wanted to get your thoughts on whether EpCAM is now a more ideal sort of target to leverage within the context of ADC relative to CD71 and CD166. I think those programs established proof of concept. They didn't seem to get where they needed to be in terms of efficacy. Is there reasons to think that EpCAM is the one that's gonna sort of finally realize the potential there?
We certainly think it's got all the properties to do that, and let me just discuss that. So first of all, EpCAM is a validated target in that we know that anti-EpCAM strategies can shrink tumors in patients. That has been documented by others, and so, yeah, that's very consistent with the preclinical results from us and others, that anti-EpCAM can be a very potent way to treat tumors in multiple modalities, actually. We've also really taken off to optimize the payload for this program. So, of course, topo one inhibitors are in the news at the moment with the success we've seen with Enhertu, Trodelvy, and others. I mean, the remarkable success with that payload.
So we have a camptothecin derivative as the payload, and we think that is ideally suited to a number of the tumor types or EpCAM highly expressed, particularly CRC. So as you know, irinotecan type regimens are standard of care for first and/or second-line CRC. By the time you get to the third line, patients have very few options. There's been very little innovation actually in third-line CRC now for a couple of decades. EpCAM is very highly expressed in CRC. We have strong evidence for tumor shrinkage in preclinical models. We've also shown that CX-2051 can be effective in irinotecan-resistant tumor xenografts.
So we see a lot of opportunity for CX-2051 in CRC, and our phase I strategy will incorporate, will certainly bias towards enrollment of CRC, you know, to look for early signals. Not to the complete exclusion of other EpCAM-positive tumor types. The target's so broadly expressed, we believe that there could be, you know, upwards of 300,000 patients potentially treatable with an EpCAM-directed ADCs. This is an enormous market opportunity, but we think that the combination of the target, the payload, and the optimization of the pro domain for this program really does optimize the potential, the potential for technical success.
So when you look back at CD71, CD166 experience, and maybe the ADC field in general, it's gonna be more payload-driven toxicity or more target-driven toxicity? And I guess what does your preclinical models sort of tell you?
Yeah, I think we've learned that, yeah, we don't mask the payload on Probody drug conjugates, and so we would expect to see payload toxicities, over time as we dose escalate. We think from our preclinical work that the window, just given how abundant the target is, should be broader, in the clinic. But we've got to do the experiment as we dose escalate and find those doses to expand.
Maybe a couple of quick ones on CX-801. Maybe speak to sort of the strategy on selecting interferon alpha as sort of the lead cytokine maybe versus IL-2, which maybe turned out to be a wise decision.
Well, IL-2 is pretty crowded, and, you know, we've even seen others in the conditional activation field present early data at SITC a few weeks ago. So, you know, we've purposely not entered IL-2, and we've, in contrast, purposely selected interferon because we think it's been a bit overlooked. I mean, interferon was the first immunotherapy to be approved. We know that interferon can shrink tumors, melanoma, renal cancer, has clear activity in bladder cancer, to mention just three. But it's got significant systemic toxicities, which manifest themselves in large part as, you know, pretty severe systemic flu-like symptoms, together with other challenges like CNS toxicities. But the power of interferon to promote antigen presentation within the tumor microenvironment is profound.
That's how interferon was first described, of course, as a driving antigen presentation for viral infections. So we see interferon as a great partner for checkpoint inhibition moving forward. So our clinical strategy will be to move into quickly into tumor types where we know interferon has shown single-agent activity, modest, but we know it's there, and to advance as quickly as we can into combinations with checkpoint inhibition, to try to unlock potential in the huge number of patients either that don't respond to checkpoint inhibitors or become refractory or resistant to them.
Maybe one last quick one along the lines of checkpoint expectations for what we might see from Bristol in 2024 and their CTLA-4 Probody?
Yeah, so our work with Bristol continues. We continue to be very excited about that. So it's focused on the clinical work on 288, which is the masked version of the non-fucosylated ipilimumab. They are in a CRC study and a non-small cell lung cancer study. They are guiding to having data in 2024, and you know, we're obviously really looking forward to seeing that. And that could be very important for the company and for their franchise, because right now, BMS-986288 is the leading edge of the BMS long-term strategy on CTLA-4. So you know, we're proud to be doing that work with them, and we're very much looking forward to seeing some results.
Perfect. With that, we're just about out of time, so I wanna thank you, Sean, for your time and thoughts. Thanks, everybody, for tuning in. Take care and enjoy the rest of your day.
Thanks.