Good morning, everyone. My name is Eric Schmidt. I'm one of the analysts at Cantor Fitzgerald. Welcome to day two of our healthcare conference. My pleasure to be hosting the next fireside chat with C4 Therapeutics. We're delighted to have with us up on stage, the company's Chief Executive Officer, Andrew Hirsch, as well as the company's Chief Medical Officer, Len Reyno. We've also got a couple of representatives from the company in the audience, CFO and IR, so thanks for the full team for being here today. Andrew, for those a little less familiar, maybe just start with a high-level two-minute overview of the company and state of affairs at C4.
Yeah. Thanks, Eric, and thanks for having us. So, you know, C4 Therapeutics is one of the leading targeted protein degradation companies. We were founded back in twenty fifteen. We came out of Dana-Farber. We were founded by Nathaniel Gray, Jay Bradner, Ken Anderson, and Marc Cohen. Had a foundational collaboration with Roche that kicked off the company, and, you know, we've been pursuing targeted protein degradation in oncology as our focus. We've put in our history three programs into the clinic, hopefully soon to be a fourth. We currently have two ongoing phase I studies. One, we just actually presented preliminary data this past weekend at ESMO.
That's our CFT1946 program, which is a BRAF V600X degrader, which I'm sure we'll talk about. And then the other program, which I know garners a lot of interest, is cemsidomide, our IKZF1/3 degrader, where we have data coming in the fourth quarter of this year. So, everyone's been very busy at work, really driving those programs forward and enrolling patients, and we're looking forward to sharing the next data sets. We also have a number of ongoing collaborations, which I'm sure we'll talk about.
Great. Yeah, absolutely wonderful timing with regard to 1946 and the ESMO presentation just this week, so we'll look forward to spending some time there. But maybe just before we get into the individual programs themselves, Andrew, this field or Len, whoever wants to comment, this field's kind of a protein degradation. It's had its ups and downs. It's sort of finally gotten to the point where, you know, we're seeing some on-target potency and selectivity. What has sort of been, you know, the challenge with some first-generation programs? What are you doing a little bit differently from others in the field to solve those challenges?
Look, I mean, I think the first thing is, I think the biggest challenge the field had is reasonable expectations, right? I'm very excited, we all are, about what targeted protein degradation can do, and it's an amazing tool, and I think that's reflected the fact that the number of companies in the space and the number of large companies that have internal programs. So it's going to be a, as part of the drug development toolkit for years to come, and it really has some key advantages in terms of drugging the undruggable, solving some of the challenges with inhibitors.
And so I think initially there was some unrealistic expectations, because at the end of the day, you're still subject to all the risks and uncertainties in drug development, right? This. There's not a magic bullet. And so I think there were lots of high expectations, and I think, you know, we run into your typical drug development challenges, and I'll give you a good example. Our BRD9 program team made an amazing degrader, highly selective, highly potent, really on-target degradation of an undruggable target, which is amazing for the field. The biology was really compelling. We put it into the clinic, and lo and behold, BRD9 is overexpressed in cardiac tissue. We ran into some on-target toxicity. That's drug development, right?
Those are the risks that you take when you go after novel targets that have not been drugged in the clinic before. And so I think you, you're seeing those challenges. I think where we have focused and thinking about how we make degraders is, it's really important that we have good oral bioavailability. That's a key hallmark of all of our programs. And then our focus is on making the most highly catalytically efficient degraders we can, so the most potent degraders. We think the deepest and fastest you get the protein down is the best. That's where you get some differential biology. And then really being smart about target selection, I think that's the other challenge.
You know, one of the things that you know, I ask the team all the time. I learned this back from my Avila days, which is sort of what problem are we solving? You know, I call that our degrader rationale. You know, why do we need a degrader against this target? How can it differentiate versus other modalities that are out there or that maybe currently address the target?
Okay, so certainly target selection is important. Sounds like you believe the field in terms of the degrader technology has made some good leaps and bounds with regard to the ability to selectively and potently degrade. What are you guys doing, a little bit more specifically, with the cereblon binders, and why is that maybe an ideal approach to degradation itself?
Yeah. So we've made a big, deep investment in cereblon as our E3 ligase of choice, and that's for a number of reasons. First, it's clinically validated. That's the ligase that's used by the IMiD class of medicines, which are degraders. Of course, we didn't know they were degraders when they were developed, which is why they are sort of poor degraders of those targets. I'm sure we'll talk about that at some point. And so it's clinically validated. The other thing is we actually know what its natural function is. It doesn't have any regulatory function. It actually is a garbage collector. That's its role. That was elucidated by a paper out of Harvard.
And so, you don't worry that you're interfering with some natural regulatory function of a cell by kind of co-opting it for target protein degradation. The third reason why we like cereblon is that it's well-distributed in all tissues and compartments, so we know that it's gonna be there to degrade the target. Now, that creates some liabilities. For instance, if we wanted to go after a pan-essential gene, like a GSPT1, you wouldn't want to use cereblon because you're gonna have lots of on-target toxicities. But again, that's our strategy is not to go after those type of targets, so that's not something that we're concerned about. In fact, targets like that, those are anti-targets in our screen.
and so, so, you know, and, and the reason why we like cereblon too is what we've learned is that very, very subtle changes in the exit vector of the cereblon binding component of a degrader out of the E3 ligase can have a really, really big impact on catalytic activity. And so we've made a big investment in our library. I think of cereblon binders, there's now probably over 10,000 compounds, and that helps us make, you know, highly potent, selective degraders through being able to use that tool. I think if we were to pick another one, we'd wanna make that same investment.
Mm-hmm.
And so far, we also haven't had the need to.
Mm
come up against a target where we haven't been successful with cereblon.
You've got both MonoDAC and BiDAC degraders. Is there preference for one over the other?
No, we can, we can do both. I think that the reason why we like we can do both is, A, part of that library lets us evaluate it for any glue-like activity, and it really gives us wide target latitude. You know, there are some targets, like transcription factors, for example, that don't really have good binding pockets that would lend themselves to a BiDAC approach, and so that's where we would use the monoDAC. And we've had programs where we've looked at one, where we've taken a dual track, where we've tried to do both. Certainly, hit finding is a little more difficult for glues. But once you have a hit, then the tools in our platform are sort of don't really care about which type it is in terms of optimizing the degrader.
Okay. So let's talk about targets. You said earlier, you maybe made a misstep with BRD as a target.
I haven't made a misstep.
Uh-
It wasn't a misstep at all. I think it was a great target.
I think it was pretty important work, actually.
Yeah, yeah. It was. I think it was. It wasn't a misstep at all. I think it was an exciting target. It was a classically undruggable target. The biology was really compelling in terms of the synthetic lethal setup. I think it showed A, the challenges with some of those targets. I think we've seen other degraders against some of the chromatin targets out there. I think it's just they're challenging.
Okay.
It's challenging biology, but that doesn't mean we don't do it. I think it's a question of what level of risk do you take across the portfolio?
Okay.
I think we like to have a blend of, you know, unvalidated targets that are novel that we think could really open up disease care. But I think the key thing is understanding the biology, right? That's you know, the degrader and the chemistry can be fantastic, but at the end of the day, you're gonna run up against biology. And so, you know, it's-
Either way, BRDs is water under the bridge, right? So-
Yeah.
So let's talk about where you are focusing your time, either with BRAF or the IKZF1/3. I mean, why are these, generally speaking, priority targets-
Yeah
... for you?
We'll start with the IKZF1 and 3 target, and that was sort of the first drug we put in, or program we put into the clinic. And that's sort of a natural first step to test a new platform because, you know, as I mentioned earlier, right? We know that there are degraders of those targets, but we know when thalidomide, lenalidomide, and pomalidomide were developed, they weren't. No one knew they were degraders, so they weren't optimized as degraders. So what better way to test a platform than to come up with, you know, a best-in-class, fully optimized degrader of those targets to really drive better activity, better efficacy, better tolerability?
And so I think we've developed a highly selective and potent degrader of those targets, and I think you know the data's gonna play that out as we share out in the fourth quarter.
BRAF? Why is that a good target?
BRAF is a great target because it's a validated target, right? And there's a number of BRAF inhibitors. But there's a challenge with BRAF inhibitors. There are resistance mechanisms that occur as a result of inhibition, and, you know, paradoxical activation is one. So, you know, you can inhibit the monomer and stop signaling through the MAPK pathway, but it's well known that that inhibited monomer can dimerize and lead to activity through the MAPK pathway. In addition, there's on-target toxicities, and you see skin toxicities with that. And so we think a degrader is a really elegant solution to solve that problem. Simply by removing the protein, you both stop monomer signaling, and there's nothing to dimerize.
And so we think it's a great solution for some of the challenges we see, and I think some of that started to play out in the data that we just presented at ESMO.
Any unknown biology around degrading BRAF that we should be aware of?
Um-
Any risks?
Don't know yet. You know, that's something we're gonna learn. You know, we have a pretty robust translational effort as part of this program. Haven't shared that data yet. Whereas that data is still coming in, it's a bit delayed from the dataset that we shared this past Friday, but still collecting that and digesting that. But so far, when we get into the clinical data-
Yeah
... you know, we'd really, the drug looks to be incredibly well-tolerated, well, safe. We don't really see any signs of cutaneous adverse events kinda that are hallmarks of BRAF inhibition.
Len, you just got back from ESMO. Thanks for being able to join us here today. What are the key learnings from your presentation late last week?
So, we had the opportunity at ESMO to do a preliminary presentation of the ongoing phase I program, and I'll start by saying I'm super proud of what C4T's done in this program 'cause it's extremely ambitious in terms of the number of questions we're trying to answer. The ESMO presentation was the first unveiling of the preliminary data set. And on that regard, I think you know it's a really elegant proof of mechanism and proof of concept for the degrader concept as it relates to BRAF. So what did we present at ESMO? As I said, it's a complicated study, but the key features of the study is we're taking patients who are BRAF-experienced from the key clinical areas where you'd expect to have BRAF-treated patients.
So that's melanoma, colorectal cancer, and non-small cell lung cancer, as well as a fourth group, and the fourth group is other malignancies not just described, that are driven by a BRAF. So we've taken those patients, and they've had a BRAF inhibitor at some point in their course of therapy, as well as other standard of care regimens. And the first part of the study, which we've unveiled at ESMO, is the monotherapy dose escalation.
... as well as, the first pharmacodynamic tissue-based studies, looking at PD endpoints. So in terms of dose escalation, what we've been able to show is that the drug is exceptionally safe. We've dose escalated through five dose levels, starting at twenty milligrams BID and culminating at six hundred and forty milligrams BID. And what we've shown across those dose levels is that we've had increases in exposure associated with dose, which is the first drug-like thing we must get. But really importantly, the drug's been exceptionally safe. Across all those five dose levels, we've had no dose-limiting toxicities. But the story is better than that, actually, because the other promise of the degrader that Andrew just alluded to is that the degrader will spare the wild-type inhibition, and then therefore spare patients the wild-type toxicities.
If you go back and look at the early development of the inhibitors, you'll see that, in fact, the dose-limiting toxicities of monotherapy inhibitors is skin toxicity, including cutaneous rashes, but also the early induction of squamous cell carcinoma. If you look at our data set that we shared at ESMO, in fact, we have only very sporadic skin toxicity terms at all, nothing greater than grade two, and there's nothing emerging in the data set at all to suggest we have wild-type toxicity. In fact, in conversations with both our investigators as well as clinicians who saw the data for the first time at ESMO, their uniform response is: "Wow, this drug is super well-tolerated and going to be a great combination partner." What else did we see at ESMO?
So the other thing that we saw at ESMO is across the range of doses tested, we've been able to show that, in fact, the target is present in fresh biopsies, which was done in select patients, and we're getting tissue degradation when we measure in core needle biopsies at day fifteen. That would be an interesting endpoint, but really the icing on the cake in the story is what does that map to clinically? And in the clinical data set, in terms of antitumor activity, what you'll see is we presented both at ESMO and in the webcast waterfall plots, showing that we have antitumor activity as monotherapy, really emerging across the range of doses tested, including objective confirmed partial responses in a patient with melanoma, as well as quite excitingly, a patient with refractory pancreas cancer.
So where we're positioned at the end of the study that was presented is the study's not finished. So we're continuing to optimize the monotherapy dosing. So we've disclosed previously that we have enrolled a 320 milligram dose cohort in melanoma, and that's to look at first clues of monotherapy efficacy. That will be subject to presentation sometime in 2025, and we've also now explored 640 monotherapy in melanoma as well. To come is more tissue-based studies on the pharmacodynamics. This is just a teaser set, if you will, that we shared at ESMO, but we'll be sharing more pharmacodynamic and translational endpoints.
We've also started dosing the drug in colorectal cancer with cetuximab, because we think that's a really important combination in view of the potential need for EGFR blockade in colorectal, and at some point, we'll study MEK inhibition. Exciting snapshot of an ongoing study, but that ticks a whole lot of really important boxes in early development.
Okay, lot to unpack there. Obviously, a very interesting target and some very encouraging early clinical activity. On the biomarker standpoint, you mentioned the tissue staining.
Yeah.
It was sort of a binary-
Yeah
... panel of information.
Yeah.
What else is coming? How do you know that you're hitting the target with good potency?
Let's talk about what we showed in the data set. It's important to put that in context. Just so people are familiar with what happens in the clinical pharmacodynamic research space is, once a dose level was declared safe on the study, we opened that to enrollment for what we call pharmacodynamic backfill patients. And those patients consent to have a core needle biopsy done before they get any study drug, and as well, after fifteen days or fourteen days of therapy on day fifteen, to get a follow-up core needle biopsy. Important to note, core needle biopsies are distinct as opposed to excisional biopsies. You get a modest amount of tissue, so one of the challenges is how much can you do with the tissue you get from a needle biopsy as opposed to a true excisional biopsy?
But let's talk about what we did get and why we got it. So first of all, why day 15? So we think two weeks of exposure is enough exposure to start seeing a tissue-based effect, but I want to emphasize, that's not because we know that's the gold standard time point to get a tissue-based effect. It's an opportunistic time point, and you want to do that biopsy before you know the RECIST criteria of response. The second thing is when we're limited in terms of how we can study that tissue. So we have an antibody probe that will stain for V600E from Roche, which we've used, and then we've used H- score as a surrogate quantitative method to quantify protein expression. So an H- score of 300 is the highest score possible, 0, obviously the lowest.
What we showed is that the patients that enrolled in those cohorts came to clinic with V600E staining, with H- score of around 200, before they were ever dosed. We also showed on day 15 that the H- score was now 100, consistent with quantitative degradation. What's to come is we have a wide biomarker panel of both tissue-based as well as liquid biomarkers, and we alluded to some of that in the pancreas vignette where we showed circulating tumor DNA. We'll continue to build out that data set and as well, eventually start to map outcomes to the pharmacodynamic tissue. It's an early data set, but within the limitations of how to collect it, I can tell you, very reassuring.
Many clinicians actually came up to us after the presentation at ESMO and saying that there's two slides in the deck, the one where you show the tissue degradation, and the next slide where you show the waterfall plot with tumor shrinkage. That's the companion two-slide deck. You want to add a third slide is, this is a slide that has no DLTs. It's emerging data, a lot more work to be done, which we are committed to doing and are doing, and look forward to sharing more comprehensive data in 2025.
Maybe a question on dosing, not yet knowing where you are in that pharmacodynamic range. You mentioned that you're safe up to no DLTs up to 600 milligrams, but the melanoma cohort's 325.
Yeah, let's go. I'll go back a step. One of the things that we want to do as a company, and really any company is doing these days, is to start to factor in questions that will help engage the agency vis-à-vis Project Optimist type discussions. As we were going up in the dose escalation, once we got to 320 milligrams BID, from a pharmacokinetic exposure point of view, we were satisfied we were getting into relevant exposures, but we also started to see evidence of antitumor activity. Because melanoma represents a really great histological subtype to test the degrader rationale, we made the decision to explore that as a monotherapy dose exploration cohort.
Not because we think that's where we're going to end up necessarily, but we thought it would be an important starting data set. Likewise, now that we've cleared 640 and we've enrolled 10 patients on melanoma who we're now following, at 640, what happened is, obviously, it's safe. We have this very compelling monotherapy response in a patient with pancreatic cancer, and so again, from a Project Optimist point of view of evaluating multiple dose levels, it makes perfect sense now to do a melanoma cohort at 640. We'll take that data together and make decisions as to what that leads to next, but that's a data-driven decision that will occur, you know, beyond into 2025.
Above 640, are you still going higher and higher?
So we haven't made a final decision there yet. One of the things that you know we're conscious of is whether or not there becomes a point where just from an oral absorption point of view you get less than because of the complexity of the oral absorption. We only have four patients with PK at 640 in the original data set, so we're populating that data set with more patients right now with tissue-based studies. So we'll look at the PK that we get over the next few months and then make a decision as to whether it seems rational to go higher. What I will say, and it's an important thing, is the clinicians are absolutely wanting to go higher if we can demonstrate that we can practically do it.
But the other thing about the drug, I think that's important to note, and we've really got ample evidence in this regard right now, is that this is unlikely to be a drug that you dose to exceed an MTD. And so we need to be prepared to make final dosing decisions on doses that are less than the MTD, and hence the importance of doing these exploratory cohorts, as well as getting more tissue-based studies.
But I do assume you would like to get as much degradation as possible. That's a goal.
So we would like to get as many responses as possible. And I’m not being facetious there-
Yeah.
because I think one of the challenges that we have is the ability to measure degradation in clinical tissue specimens. It's not the same as, for example, in our myeloma program, where we can measure IKZF1/3 degradation in circulating plasma. So you're limited by the technology. So, you know, we're first and foremost want to drive patients to clinical response. We certainly want to see if we can determine if one dose is better than another in those windows. But the final arbiter of it, does the drug work, like all drugs, is gonna have to be in antitumor responses.
Okay, so lots of pharmacodynamic exploration, lots of dose exploration, lots of melanoma cohort exploration. What's next in terms of what you might release to the investment world or the medical world?
Yeah, I mean, so it was a preliminary data, right? So we're gonna hopefully have a more complete data set for the phase I dose escalation, right? There's the monotherapy melanoma cohorts, right? That we just talked about. We expect to have those sometime in twenty twenty-five. And as Len mentioned, we've also started the cetuximab combination safety escalation. And we should have. You know, that data set should also be available in twenty twenty-five. We've not yet kicked off, but it is in the protocol, a combination with a MEK inhibitor-
Yeah.
That depending on when we start that, that may or may not be next year. I think we want to get a read on the monotherapy activity before we sort of kick that off. So, lots of additional data sets coming from this program. This was just sort of the introduction.
Yeah.
There's a lot more chapters to come.
In the interest of time, I guess we better shift over to cemsidomide. But, how do you, Andrew, think about allocating resources to these two programs? Is either, you know, getting full resource support right now, or are you-
Yeah.
constrained in any way, or?
No, I mean, they're both. We have, you know, as we ended Q2 with just under $300 million, which probably does run right into 2027, and both programs are fully resourced to execute the current studies that we're executing, which would, you know, will take us into 2025. You know, we're starting to plan, you know, for both programs, what the next set of studies would look like, kind of drafting protocol synopses and thinking what those might look like based on the dataset. I'm not ready to share what those are now. You know, obviously for the next phase of development, you know, pushing into pivotal studies, we're not funded for that, obviously. That's something-
You've talked about-
And we shouldn't be, right? So-
You've talked about wanting a partner, at least for cemsidomide
Yeah. I mean, ultimately,
I think that's the right place to go. You know, it's a very complicated landscape, multiple myeloma. It's changing rapidly, and it is a field for lots of large pharma players. There's multiple combination regimen treatments. I don't want to say treatment paradigm because as Ken Anderson, our board member, says, it's not. They're not really treatment paradigms, it's more treatment menus, and people sort of pick and choose how they like to practice.
I see.
And so it creates quite a complex environment. And ultimately, if we think about the rest of our pipeline, it's in solid tumors, not in liquid tumors. And so we think of for a commercialization partner, you know, we'll need a partner for that. And so, you know, certainly helping with pivotal studies will be important.
Okay, well, with this cemsidomide program, the goal is a better mousetrap, right? A better IMiD. First, let's just start there. Why does the world need a better IMiD?
I think, I mean, IMiDs have done really well and have helped lots of myeloma patients, but I think they certainly have their liabilities. Patients continue to progress on them, and so to the extent you can, with a more potent IMiD, you can drive better, more durable responses, right? That's good for patients. That's good for the world. And so, you know, that's the goal is let's have a best-in-class drug in what's a $20 billion market today.
Okay. And maybe before we get too far into the cemsidomide data, obviously, Bristol, previously Celgene, has a candidate that they believe, I guess, mezigdomide, that they think might also be a more potent IMiD. What are your thoughts there on the competition?
Yeah. So obviously, we're following mezigdomide closely. You know, they certainly have done a nice job of development there, although I think that we have questions around the toxicity profile with that drug. As you know, most people know, you have on-target neutropenia, that's one of the consequences of degrading IKZF1 and 3, and that's why all of these, this class of medicines are dosed with a dosing holiday to allow neutrophil recovery. Importantly, what the drugs aren't doing is killing neutrophils in the periphery. What they're doing is they're blocking maturation in the bone marrow. And so the neutropenia you see is the natural life cycle of the neutrophils in the periphery dying off and not being replenished. So you need that break to, you know, that...
in treatment to let that replenish. You know, and so when you look at the class of medicines, you know, you really see kind of across the grade 3-4 neutropenia in the 40% range, with 3%-5% rates of febrile neutropenia. But when you look at mezigdomide, that number is significantly higher, about 75%, along with 15%. So we think we don't think it's optimally dosed or designed. Our drug, one of the key feature in addition to potency, is PK. We have a long half-life, 48 hours, and in addition, we have higher tissue residence, so it stays longer in the tissue, which we think is important to drive myeloma cell kill, and so we think that creates an advantage.
We are dosing it differently because of that. We have a fourteen-day on, fourteen-day off schedule, which we think optimizes the properties of the drug in terms of maximizing myeloma cell kill, but also giving a break in treatment to let the neutrophils recover. I think, you know, the initial data we shared in the plus dex cohort in December kind of played that out, and we've continued to execute through two additional dose levels beyond that.
Len, maybe you wanna give us a little bit more of a snapshot of those data and what's next.
Sure.
for clinical trials here.
Again, super excited, and you know, just pointing out that C4T is a small company. We have two drugs well into the clinic that I would argue are both post-POC. That in and of itself is a good benchmark. In terms of you know, where we're at with cemsidomide, and we will be sharing data by the end of the fourth quarter of this year. We've made exciting progress in further delineating the therapeutic index, if you will, for the drug. We've gone through now two additional dose levels since the disclosure in December. But importantly, what we've done is we've added additional patients to each dose level, such that by the time we get to the end of Q4 disclosure, we anticipate having approximately 30 patients at multiple dose levels plus dex.
That will really give us a better handle on both the safety index of the drug, which we now know is quite safe because we've dose escalated twice since we last shared data, but also how the efficacy signal is emerging. So stay tuned. I think it will be an exciting and important year-end data set.
Okay, terrific updates. Thanks, team. I know we're out of time, so please join me in thanking C4 for the presentation today.
Great. Thanks, Eric.
After 15 years, you didn't disappoint.
Yeah.