Good morning, everyone. Thank you for attending Jefferies London Healthcare Conference. My name is Kelly Shi, one of the senior biotech analysts here. In this first chat session, we are very pleased to have Mr. Andrew Hirsch, Chief Executive Officer, and Dr. Leo Reyno, Chief Medical Officer, join us. Welcome both.
Thanks, Kelly.
Great. Thanks, Kelly. Thanks for having us.
It's always our pleasure. Maybe we can start with high-level questions. C4 can develop both molecular glue and PROTAC. What are the unique features you see for each of these two modalities? Do you see each modality is suitable for a certain group of protein targets?
Yeah, no, thanks. It's a great question and something that we get asked all the time. And we think it's important as a leader in the field to be able to do both what we call the MonoDACs, or molecular glues, or BiDACs, or heterobifunctional degraders. And we think it's important because it's important for the target, right? And so we don't have a preference for either one. We think they're both very suitable and viable. And in fact, we have examples of both in the clinic. It really depends on the target, right? And I think that's how we go about it. There are some targets which aren't really amenable at all to a BiDAC approach. There may not be a site that we can bind to. The surface of the characteristics of the target may be smooth.
That's the case with IKZF1/3, the target of our lead program, cemsidomide. And so we take a glue approach there. Others, where there's a binding site and we can either find ligands or do ligand discovery, we can take a heterobifunctional approach. And they're different, right? The hit-finding approach is different. You can more rationally design a BiDAC. But once you get a hit in either, our platform in terms of optimizing and creating highly catalytically efficient degraders really is the same. I think as a field, there can sometimes, in some cases, people would prefer to do the MonoDACs or the glues. And I think a lot of that is because people are still stuck in trying to follow the Rule of 5. And I think our data has shown across our heterobifunctional degraders that we can have beautiful drug-like properties.
In fact, at this past TPD Summit, we presented data on CFT1946, which is our BRAF V600 degrader, that showed brain penetrance. We had Kp,uu values of 0.35-0.88, depending on the model used, which is as brain penetrant as many brain penetrant kind of traditional small molecules on the market. We don't think it's a problem. We really think what matters is the target and what's the right degrader approach for the specific target.
Super helpful. And maybe we can start with the leading program, cemsidomide, target IKZF1/3. In the ASH abstract, you already give a preview of the ongoing clinical trial in the pretreated multiple myeloma patients. Maybe you can walk through the initial data and also set expectations at the ASH conference. What kind of additional data and a longer follow-up you'll be presenting to investors?
Sure. Sure, I can handle that. I think if you consider our public disclosures in totality for 2024 on cemsidomide, there's lots to be learned already. And so, prior to talking about the abstract, just to refresh everybody's memory, we started 2024 having just cleared the first dose level plus dex 37.5 micrograms. And as we marched through the year operationally, we did the same thing at each dose level. So we would increase a dose level. If it was declared safe, then we would backfill the dose level such that we would get 10 to 12 patients at that dose. So if you take it in totality and as manifested in the ASH abstract, what we've done in 2024 is prosecute at 37.5, 62.5, and 75 microgram- dose levels.
You can anticipate in the ASH presentation that we'll fully characterize the safety and any emerging anti-myeloma effects at each of those dose levels. We've also shared with the street that we have escalated to 100 micrograms, but that is an ongoing early cohort. The 100- microgram cohort will not be shared at ASH. Taken together, what do we know about that? The fact that we were able to escalate through all those dose levels is actually a really important finding in itself because early on in the development of the asset, there was the fear that it had a narrow therapeutic window. In fact, that's not true. We already know from those disclosures that we have a much wider therapeutic window than anticipated.
In terms of what you'll see at ASH vis-à-vis responses at different dose levels, et cetera, obviously we can't disclose that based on the embargo, but at ASH, we'll give a full accounting of, again, safety across the doses, including the characterization of how, in particular, neutropenia manifests at each dose level because that's an expected, albeit manageable, side effect, and we will also report out each dose level vis-à-vis efficacy. That said, really important, these are not individual clinical trials, so it's really important to think about the data as a collective, and I think what we will show is that the data as a collective serves to make the case that this is a best-in-class IKZF1/3 degrader.
Okay.
Yeah, I think I'll just add, I think the other thing to recognize is that the data cutoff for the abstract was in July, and obviously, we have a more contemporaneous data cut for ASH, so you'll have more patients. The lower dose patients will obviously have longer follow-up, just given the course of enrollment, so you'll see it's not a completed phase I, it's ongoing, so you'll see patients with different levels of follow-up across the different doses.
I see. So we will have additional patient data at a higher dose and probably three plus months of additional follow-up?
Yes, roughly three plus of additional follow-up. And so if you map back to 37.5 micrograms, that cohort has roughly nine months of follow-up. Each subsequent cohort is sort of missing three months. So you can do nine, six, three based on the dose levels.
Great. And it is nice you can dose higher. Curious, should we expect dose-dependent responses? Because in certain scenarios, degrader molecules are known to have this bell-shaped response. So what should we expect for this molecule?
We're not dosing at the range where we would expect we would encounter the bell-shaped response. That's the first thing. I don't think it's going to get worse with time. I think there's two parts of the question, though. What do we know about the class in general? This is obviously not a first-in-class asset. And so what we know about the class in general is dose or exposure does matter. There tends to be more anti-myeloma effect at higher doses. And prior to our data set, there's also been a pretty tight relationship between that anti-myeloma effect and the safety event of neutropenia associated with blocking maturation of neutrophils. So in terms of that premise, I think we'll address that two ways. One is we'll obviously look at individual dose levels.
But the other issue, of course, to remember is that this is an orally available drug. So population pharmacokinetics is really important. And what is the population exposure relationship to emerging efficacy? So we'll characterize that in the presentations, plural. One point of fact is the ASH presentation is a poster. So there's limited real estate, if you will, in a poster. So at the same day as the poster release, we'll be having an industry event, which will allow us to go a little bit deeper on some of those questions.
Terrific.
I think if
you're referring to the hook effect, I think, right? That's something we typically see preclinically with heterobifunctional degraders, right? Because what happens there is you oversaturate the system. And so you have the degrader attaching to an E3 ligase, the degrader attaching to a target, and they don't come together. This is a MonoDAC. So we don't really typically see that effect at all with a MonoDAC. And actually, clinically, we never really ever get to concentrations where you actually see the hook effect. It's more of a preclinical artifact where you're really pushing exposure to the extreme levels.
Okay, great to know. And cemsidomide achieved stringent CR at 37.5 micrograms. And with the patient initially having very good partial response before converting to the stringent CR, curious, is it common for responses to deepen over time with the CELMoD treatment?
I'm going to anchor on the word common. I think I don't know how common it is, and it has to be defined in each setting and each line of therapy. But in principle, the principle that you can have deepening responses over time is absolutely supported by our own available data, but also data in the class. One of the things that we will show in sort of a swimlane plot is how does that manifest in the broader data set. But in terms of how frequent it really is, I think you have to then look at lines of therapy and whether it's a single agent or combined, et cetera. So the other issue with response in myeloma, of course, is how you determine response with myeloma working group criteria. Actually also maps back to the status of the disease.
So that if patients have light chain disease only, it's much easier in those patients to declare response early because you're just measuring that almost like a pharmacodynamic effect that occurs even in the first cycle. But if you have M protein or extramedullary disease, those actually take longer sometimes to actually manifest because the parameters you have to measure are measured at different points.
Great, and we have seen initial differentiated safety signal, particularly with a low rate of neutropenia. Can you discuss the factors that may contribute to this improved safety compared to the monotherapy cohort?
I think there's two things at play. There's the drug itself, and then there's the dosing schedule of the drug. When you think about how the drug was designed by our preclinical and discovery team, the drug has been designed to have low protein binding, favorable to bone marrow distribution, and having tissue effects. That may have something to do with it. But the equal thing that we were able to show, and actually serendipitously from the original experience when the drug went into clinic, and we tried 21/7 because we anticipated a shorter half-life, obviously, we had neutropenia that was more prolonged and unacceptable. That led to a really terrific opportunity where the team in-house then backed off that and reevaluated and developed a model of what the dosing schedule should be for a 48-hour half-life.
And in fact, with that model, what happens is that you can get IKZF1/3 degradation through the 14 days and beyond, and it will continue because the half-life is longer. But by stopping at day 14, you give the window for the neutrophils to start to recover. And so because that recovery window is introduced somewhat earlier, actually, you have a potential opportunity to aggregate the risk of prolonged neutropenia. So it's really elegant preclinical modeling that led to that. And this is a really exciting use of PK/PD preclinical relationships mapping to clinical use. So it's been a tremendous success that really the whole company should take credit for.
Okay, very insightful. And it's still early days, but I'm curious, could you share your thoughts in terms of future development strategy? And especially when compared to pharma CELMoDs , they actually move forward different molecules for different indications. And how do you see cemsidomide could be differently in terms of development strategy?
Yeah, it's a great question. And it's really exciting to see the changes in the myeloma field as there's an advent of really exciting new therapies, CAR-Ts, bispecifics that have really made stark improvements in patient outcomes. And so as we think about the landscape and we think about kind of what our competitors in the space are doing, our competitors look like from what we can see from their development strategy trying to replace the existing franchise, right? So lenalidomide and pomalidomide are becoming soon to be not already generic. And so it looks like they're positioning mezigdomide and iberdomide to replace those kind of in some of the current treatment landscapes. To some degree, we're free from that constraint. And so as we think about the best place for cemsidomide, we actually really think it's in combination with some of the more novel agents.
Importantly, there's really two mechanisms that degrading IKZF1/3 have. One is direct autonomous cell kill, right? That is one mechanism. But the other is you see activating T cells. And so we think both of those are important when you think about some of the new T cell therapies where they have really great responses, but sometimes suffer from T- cell exhaustion. And so we think this can be a fundamental backbone therapy really across all the different lines, but really helping to improve the duration of the really fantastic responses you see from the newer agents. So while we're not ready to share details of our development plan, we'll share a little bit more at the upcoming ASH meeting. It's likely to be in combination with some of the more novel agents, which we think are going to start to move into earlier and earlier lines of therapy.
Okay, fantastic. And so we could expect more details in terms of a combo development strategy at ASH?
Yeah, I mean, I think we're not going to lay out specific trial designs, powering, that sort of thing. But we'll certainly articulate what we think a path forward is to develop the drug.
I think in articulating that path forward, I think one of the really important issues is how do you articulate a path forward where you can get data relatively early in that path to say that you're on the right track and essentially de-risk the premise? I think if you look at the effects, for example, of BCMAs or CAR-Ts, where the effect size as it relates to response rate is already very high, that's super. They have longest PFS. What can you do in this? How could you improve the quality of the response? That gives you an early endpoint that you can look at in screening studies.
So for example, if you did a study in combination with a BiTE and you have a baseline response rate of X, can you increase the quality of that response in terms of patients who achieve CR versus patients who achieve a stringent CR or a CR with minimal residual disease? And so I think that's an important principle of development to bring forward in myeloma because the drugs have been so successful that they have created long tails, if you will, on endpoints of PFS and survival. But that doesn't actually preclude doing early development. I think the other issue that we've seen and can expect in the development space is some of these more novel mechanisms will likely migrate into earlier lines of therapy. And two things will happen. Obviously, it will be a greater proportion of patients eligible to receive them.
I think we shouldn't lose sight of the fact that there may well also be a significant public health and therefore investment opportunity in late-line multirefractory, which, in current estimates, is probably around 12,000 patients a year in the U.S. I would posit that, in fact, as these effective therapies come online, that population is also going to be increased. So there's a lot of opportunities to take what we think is emerging as this best-in-class drug and fit it into a contemporaneous development strategy where you can get data quickly enough to know you're on the right track.
Great. And one more question before we move to a BRAF degrader. So multiple myeloma is a quickly evolving landscape. So where do you see cemsidomide actually fitting? Maybe first, let's talk about the last line. Is it like a post-BCMA agents? And how do you think about the current available therapies and help us to set a bar for the data set?
Setting the bar for the data set is the hardest question, okay? So let's start at the end. There is no contemporaneously relevant comparator group to the patients that we're currently interrogating. And so notwithstanding that our median lines of therapy of six in the abstract actually maps exactly to what mezigdomide's phase I data set was in terms of six, the six lines of therapy they received is different. And so what we will see is that there is no contemporaneous benchmark. The important thing in terms of the data is that our data confirms that, notwithstanding all the advances in myeloma MOAs, that degrading IKZF1/3 remains a fundamental therapeutic tool. So I'll leave it at that for the moment. In terms of what we alluded to in the development space, clearly you can't do everything at once.
We would think that really that key sweet spot is likely combined with a BCMA BiTE and/or also interested in, and that would be cemsidomide alone, not with dexamethasone, but also combination with CD38, which would include dexamethasone. Those are two just sort of MOA strategies we're particularly interested in.
Okay, makes sense. So the message here is it's a heavily pretreated multiple myeloma patients and probably experienced all the possible therapies.
Yes.
Okay, great. Good to know.
Yeah, just so you know the abstract, I think we said 66% of patients had seen CAR-Ts or BCMA-directed therapy. So very different from what you see in the mezigdomide phase I experience.
Yeah, very important message. And so moving to BRAF degrader, maybe can you first walk us through the initial data from the phase I trial you presented at ESMO? And what do you say are the advantages of degraders over inhibitors tackling this BRAF inhibiting pathway?
Let's start with the second part first.
Yeah, yeah. So look, I think the challenge we've seen with inhibitors, right, is the fact that you can inhibit the monomer and that stops signaling, but the inhibited monomer can form a dimer, and then that leads to aberrant signaling. And so you see really initial great responses, but that durability with the inhibitors is not quite long because of this, we call it paradoxical activation. In addition, when you inhibit wild- type, you get that activation results in a range of skin toxicities. And so the way we've handled that as a field is we've added a MEK inhibitor to help kind of clean up any of that paradoxical activation that brings its own complications and toxicities.
The degrader rationale here is that by simply degrading the protein and removing it, you get the benefit of blocking its monomer activity, but then there's no protein left to paradoxically activate. In addition, one of the qualities of degraders that we love is the unique selectivity, and I think, as you saw in the data set, we only degrade the mutant- type BRAF V600. We do not degrade wild type at all, and I think that was clear in the safety presentation, so we think that's a huge advantage over inhibiting the target, and I'll kind of review some of the data that we helped, we think kind of starts to show proof of mechanism and support that.
I'm going to go fast because I'm aware of the time. We're really excited about what we presented at ESMO and have an ambitious clinical program. The first part that we presented is the monotherapy dose escalation, where we interrogated five dose levels starting as low as 20 up to as high as 640 BID using a Bayesian logistic regression model. Each dose level for safety has three to six patients other than the very first one. We've also done pharmacodynamic backfill in that we've done biopsies and select group of patients pre and post to look at tissue effects. What did we learn from that monotherapy experience we presented? One is that when you give more drug, you get more exposure. Important observation, it's a bifunctional molecule and you have dose-related increases exposure across the dose levels. Two, it's safe.
What we showed really beautifully in that we in fact had no dose-limiting toxicities across all dose levels. And importantly, what Andrew just alluded to, we don't have significant wild type inhibitor toxicities. So it looks like a great drug for both single agent, but also a combination strategy. Three, it doesn't matter if it's safe if you don't get any meaningful effects. And what we showed in tissue biopsies that we can demonstrate that across the dose levels where we have tissue, in fact, we're getting degradation of the target at day 15, which may or may not be the optimal place to measure it. And then three, four, that doesn't matter if it does not associate with any tumor effects.
While we had two objective PRs at that data cut, we also had anti-tumor effects across a wide range of patients, especially in the melanoma patients where we expect the monotherapy hypothesis to be easiest to interrogate vis-à-vis a response rate. As it relates to colorectal cancer in the monotherapy, we had no responders, but we had some evidence of tumor shrinkage. We went into the clinic recognizing that in the colorectal setting, you likely need concomitant administration of EGFR blockade. That phase one monotherapy ticked a lot of important drug development boxes in a 36-patient data set.
Very comprehensive. And when could we expect the next data update? And maybe also you can provide more specifics on whether we're going to have a higher dose level and also how much data we're going to expect out of the dose expansion cohorts next year.
Yeah. So next year is going to be a data-rich event. And as the year unfolds, we'll give guidance as to specifically what and where to expect, but just to sort of map what we already have disclosed. We have recruited a monotherapy melanoma cohort at 320 and are recruiting another one at 640. We'll start to get some monotherapy melanoma data vis-à-vis a potential response rate. We've also started interrogating with cetuximab and that we're doing safety cohorts with cetuximab. We anticipate by year end, we'll be able to share cetuximab plus the 1946 data. In addition, we continue to build our pharmacodynamic database. We will have more comprehensive data on tissue studies in terms of degradation and ancillary effects. In terms of trametinib, we will be starting dosing.
We likely won't have the timing of that data to be seen, but we should finish the monotherapy. In terms of your specific question, will we escalate above 640? We certainly have the go-ahead from a safety perspective that there's no reason not to escalate. But before we want to make that decision, we want to see more full PK at 640 as well as additional PD because at some point we may max out absorption of the drug.
Lastly, if we could quickly touch EGFR degrader and what is the status right now, and what do we expect the next milestone?
Yeah, so we're really excited about this program. Obviously, it's something that's gone a little bit under the radar since we've partnered it in Greater China with Betta Pharmaceuticals. We announced earlier this month that we started dosing the phase one, or our partner started dosing of the phase I study. They're going to be the ones to guide to sort of when we have data. But part of the rationale for doing the partnership was we think they can enroll quite quickly. So more to come, but it's a really exciting program because we think while osimertinib is a great drug, it certainly underperforms in the LR population. And this drug is specific to the L858R mutation. And we know that we think that's because this is consistent with other first-in-class inhibitors of the target. There's other kind of non-classical mutations that go along with it.
We think because our degrader binds to an allosteric site that is specifically created by the LR mutation, we don't really care what other mutations happen around the orthosteric site. We think those other mutations aren't going to affect it. So we're really excited to see that data set. Can't tell you when that's going to be, but as we get the trial underway and Betta starts enrolling more, we'll be able to give you some more guidance then.
Fantastic. We're going to wrap up here, and thanks for a fantastic discussion, and thanks everyone for attending.
Thanks, Kelly.
Thanks, Kelly.