Analyst in Moray Waycraft's team, Jefferies. I'm happy to introduce Yuval Cohen, CEO of Corbus. This is a fireside chat for me, so thank you for joining us today, Yuval.
Thank you for having us here.
For those who may be new to the story, can you provide a one-minute intro to Corbus?
Sure. We are a drug discovery company or drug development company. We're based just south of Boston in the town of Norwood, Massachusetts. There are about 33 of us these days, I think. We have a pipeline with three assets. They all happen to be in the clinic. Two of them are in oncology. One of them is in obesity. Each one is unique. Our assets have nothing to do with each other. I'm sure we'll talk a lot about all of them today. The last thing I'll say as an introduction is we had a really not boring year last year. It was really wonderful. We're looking forward to the second part of this year because all three of these clinical assets or experimental drugs we have, all three of them will have clinical readouts in the second half of this year.
We're really excited about it.
Great. Starting with the 701 program, you have the dose optimization ongoing across 2.7 and 3.6 mg per kg doses in MUC, cervical, and head and neck cancer, with results expected in 4Q 2025. Will that be at a medical conference? Can you clarify what should we expect in this update?
Yeah. So we showcased our dose escalation preliminary Western data, U.S. and a little bit of U.K. ASCO GU back in early this year, a handful of patients. We had seven head and neck patients, I think two cervical patients, and four UC patients. Since then, we have been undertaking Project Optimus, so monotherapy, two doses for each one of these tumor types. Later this year, in the second half, at hopefully a large oncology conference, we will showcase that data. Unlike the dose escalation here, we're only focusing on three tumor types. We're not being diluted by other tumor types. Because of that, and because of our enrollment, which has been highly satisfactory, the handful of patients for each one of these tumor types will translate into a, I think, a markedly larger number of patients that we will be presenting.
I think the priority here is important to think about. Bladder, for us as a standalone company, is not a top priority. Bladder, given PADCEV plus KEYTRUDA's dominance of it, is something that needs to be, we think, undertaken by a large pharmaceutical partner. It simply, to us, does not strike us as sensible for us to try and emulate, for example, some of our peers who are trying to go first-line therapy in effect head-to-head with PADCEV plus KEYTRUDA. That's an incredibly ambitious effort, too ambitious for us. Bladder, for us, is a very important positive control, especially because it's the only tumor type where we can juxtapose our safety with PADCEV. For us, the tumor types of interest as a standalone company from a commercial perspective are the other two tumor types, so head and neck and cervical.
In both of them, again, for all three, but especially for those two, we will have a meaningful number of patients at this conference as monotherapy. In both of them, there is a yardstick that we can look to. In other words, there is a bar that we can benchmark against. For head and neck monotherapy, that benchmark is the monotherapy data from Merck. It was 36% ORR. We think it is really only sensible to move forward in head and neck if we look similar. It does not need to be identical, but it needs to be similar. If it does look similar, I think we have a viable and exciting path forward, and we can maybe talk about what that path is. If it does not look as good as that, it probably does not make a lot of sense. For cervical, the benchmark is Tivdak, right?
The only other ADC, second line. It is a standard of care in second line as monotherapy. ORR in the low 20s, I think 24%, if I remember correctly. If we look better than Tivdak as monotherapy in a meaningful number of cervical patients, then again, we have a path forward. We could certainly discuss what those look like. I think that, to us, is a very clarifying go, no go. Of course, what our hope is, is that we come up with both of those as positive outcomes.
Makes sense. Just to push a bit on the meaningful number of patients, do you have a number in mind?
I think it will be we have had very robust enrollment in the U.S. and in Europe. I think we will be providing an update on anticipated patient numbers soon. I suspect it will come as a very welcome number. It is not an incremental number of patients. It is a meaningfully larger number of patients.
Got it. Seems like it's still enrolling, but should we expect early durability data as well?
Yes. Again, within the context of, remember, this is a study that started in effect at the end of last year for Project Optimus. Within reason. Yes, you'll see swim lanes that I think will be really informative.
Got it. Then coming back to with respect to your partner, CSPC. They're going to present dose optimization data in the second half as well. Can you elaborate on the level of communication and information flow between the two companies and how this improves efficiencies?
It's a really, really positive, constructive, mutually beneficial relationship. We are immensely, immensely impressed with them and with the quality of work that they do. It's one thing to hear about how trials in China enroll well. It's very different to see it in real time. It is astonishing. It's absolutely astonishing the amount of enrollment that can be done in China. In terms of quality, we're delighted with the quality. Remember that because we have ownership of the IND in the U.S., that means we have pharmacovigilance responsibilities in the U.S. and Europe, and that gives us additional insight into data. I think what we're seeing from CSPC, and not just CSPC, with Chinese companies in general, is the way they like to present data is they like to reach critical mass of patients. They like to go for oral late-breaking presentations.
Let's see what and when they do it. They had two late breakers at ASCO for other assets, and that's exactly what you saw. Critical mass, mature data sets, go out there and get to a podium. That's fine. Remember, they are an enormous company in terms of size and market cap. We have very different calculus. We cannot, nor would we be interested in waiting until we have hundreds of patients before disclosing data. That would be nonsensical.
Makes sense. The U.S.-U.K. study had a 16% discontinuation rate and 45% dose interruptions compared to 0% and 3.3% in China. What operational or cultural insights have you gained from these differences? How are they informing the trial design and patient management strategies going forward?
I think it's super interesting. We're seeing patterns. CSPC, we get to see the data from the inside, but I think we see these patterns elsewhere as well. There is a different culture, clinical culture in oncology. In fact, the other day I was commenting on this in social media. I think we see it even in things like obesity that we don't see in the West. In China, the culture, the clinical culture, and as well as the dynamic between a patient and a physician is what I somewhat lighthearted jokingly call a victory at all costs. The priority is efficacy. Stay on the drug, endure whatever challenges you have in terms of quality of life, but stay on this dose. We're pushing to efficacy. In the West, the value placed on quality of life and comfort is much different.
What we saw was really interesting. Chinese data set, you see very few interruptions. You see that not just with CSPC, with data from Mabwell and Hengrui, et cetera. Even in the context, for example, some of our peers of very high background levels or very high levels of treatment emergent adverse events that for a Western eye would be quite surprising. What we saw with our clinician was much more open to, let's dial it down, let's give you a break. I want to keep you on the drug, and I want to keep you on the study. I'm going to modulate the drug as you accommodate or get over an adverse event. The idea is, let's keep you on the study and make sure that you remain comfortable. We had one case in China, I think I remember.
We had one patient with, I think it was a grade two or grade three ocular tox for more than 100 days. That's very unusual for the West. It's not that unusual for Chinese clinical practice. As I said, we saw it the other day. Somebody was tweeting about it. It was an obesity. It was a GLP, an incretin. I think it was a dual incretin analog. There were almost no discontinuations in the study. Whereas the same type of drug in the West, you'd have a healthy amount of discontinuations. It's the same thing. It's just a very different clinical practice culture.
Interesting. Coming back to head and neck, given the strong Western data in head and neck, I think the 57% ORR, should we expect to see more head and neck data from your partner, CSPC, in the future update?
It's funny. It's the only one where we beat them to it, right? We were the first one to do it. We're immensely proud of it. Your assumption is head and neck is an immensely attractive market in China. Our advantage over them lasted for probably about a month and a half, given how they recruit. Nevertheless, for a brief moment, we beat them to it.
Fair. Got it. And then one of the Corbus, one of the CRB-701's key strengths is this Q3W, so every three-week dosing, which is convenient and ideal for combination with pembro. So have you conducted any preclinical work supporting this combo? And are you planning to initiate a human combination study soon?
Yes, for preclinical. We have reached the point in Project Optimus monotherapy where we are ready to switch now to Project Optimus combo. We look forward to updating everyone about that shortly.
That would mean two doses at least for.
It's the same logic.
Same logic.
Yeah.
Got it. Okay. Switching to 913. So you have the CB1 inverse agonist. You have the SAD/MAD data is expected from healthy volunteers in 3Q. What type of data should we expect in the update? How will it de-risk the next steps in your obesity development plan?
We got to forget all about oncology for a second and go back to other indications where phase one is done in healthy volunteers. This is a classic SAD/MAD study done in a phase one unit in the U.S. under an IND. We are very obsessed about generating U.S. data. It has to do with a bias that we have against ex-U.S. data and obesity. We really want our obesity clinical data to reflect the population that the FDA is thinking about. The SAD is a SAD. You just go up a ladder with a single dose. The way we do this is this is not a study where you give the drug until the volunteer, sorry, says, "Look, I can't tolerate it." That's not the idea here.
The idea here is we go up the ladder and test for circulating drug levels and that we translate from the preclinical efficacy model, right? Once we've done that, we switch to MAD, which is a week long, and we do the same ladder again. That will give us steady state, et cetera, et cetera, et cetera. Always looking at peripheral levels. That then opens the door to conducting an efficacy study. I mean, technically, it will be a 1B, but really, it's an efficacy study, which will be a 90-day study in obese non-diabetics, again, all in the U.S., nothing unique about them, the classic patients, where in 90 days, we want to know how much weight do they lose. We fully anticipate them to lose weight. This is such a clinically validated mechanism of action that it's highly unlikely that they won't lose weight.
We expect them to lose weight. We will test multiple doses, and we expect there to be a dose response because every CB1 inverse agonist has always shown a dose response curve. We still are unclear about the monlunabant press release. It is disappointing that Novo did not release the actual data for their higher doses. It would be very surprising if there was no dose response. I think the most important thing for both sell side and buy side, and frankly, all of our audiences, is not does it work or not. That is a pretty easy hypothesis, apologies. It is a pretty easy hypothesis. Is it clean? The clean here is around neuropsychiatric adverse events. Otherwise, CB1 inverse agonists are, a reminder to our audience, CB1 inverse agonists have two advantages. They are not sarcopenic, and they are pretty mild on the GI. They do not affect gastric motility.
It is all about the neuropsychiatric adverse events. Going back to your question, what does the SAD/MAD tell us? And then what would the 90 days tell us? The SAD/MAD, if let's say it is a week in healthy volunteers, we do not expect there to be any adverse signal, right? The one advantage, though, that makes it a little bit more interesting, or two advantages, are you are going up to doses that are super high multiples of what you would chronically treat because it is a SAD/MAD . You are pushing the limits. The second thing is, like all obesity studies, all drugs, we look at C-SSRS and PHQ-9. In a SAD/MAD, you get to see them every single day. The granularity of interrogation is extreme. Between those two, I think it gives a little bit more comfort if the data is clean.
The 90-day study will give, I think, markedly more comfort neuropsych adverse events in general in pharmacology, but in CB1, from we see this in, for example, in taranabant, if they happen, they happen very early. Taranabant's Kaplan-Meier curves are really pretty. Everything happens the first two weeks. After that, the curves are absolutely parallel. We are comfortable with the notion that if we see if the 90 days is clean for neuropsych, for us, that's a very important de-risking event. It's very, very, very important to understand that we don't have to prove a negative. You cannot prove a negative. Think of it as a completely novel drug. This is not like the first class of drugs. So far, the FDA hasn't treated it as though it was the first class of drugs.
We're very encouraged that the feedback we've received from FDA fits within the guidelines of obesity drugs. Nothing exotic.
Interesting. Are there any internal benchmarks or gating criteria that you're using to decide whether to accelerate or pivot the program ahead of phase one dose ranging study start?
For obesity?
Yeah.
We couldn't. I mean, technically, you couldn't. You need the one to get to the other. So it's sequential. The one opens the door to the other. Phase one, remember, is not going to give us any PD data. These are thin volunteers. Somebody, oh, you'd appreciate this. Somebody today mentioned that in the incretin analogs, even healthy volunteers lost weight. I gently reminded them that that's not because of appetite suppression. It's because of those high doses. They were literally sitting on the toilet the whole day. Sorry to be crass, but that's a very, very effective way of losing weight. If you're vomiting or have major GI discomfort, you're going to stop eating. That's obviously not how they work. CB1 doesn't have that issue. The GI adverse events for rimonabant were about 10%. We expect no PD signals.
That's not the purpose of the exercise here. The purpose is a clean bill of health.
Got it. So you mentioned about the phase I- B study that's going to start in 4Q and then the day 90 and the neuropsych concerns. But are there any specific design elements that you're incorporating to monitor these risks, potentially the types of patients or anything?
Wonderfully standard. C-SSRS and PHQ-9 at baseline. You screen patients. That is a score for them. That is nothing new. Sadly, nobody wants these patients in their studies. I say sadly because one-third of patients are like that for obesity. Nobody wants them. Once you are in the study, every visit, you would administer those two questionnaires. That is it. The questionnaires are highly validated. It helps if you choose sites that are very familiar with all. All experienced obesity sites are familiar with administering these questionnaires. You really just want to come out of this with no imbalances.
Got it. Just to push it back on the neuropsych AE profile, do you believe a three-month treatment duration is sufficient to fully de-risk the profile?
Nothing fully de-risked until it's over. That's just the way it is. Having said that, if we look at the history of CB1 inverse agonists, taranabant is the one that really did have issues. It's a different backbone. In general, in pharmacology, neuropsych happens early. It seems to be independent of the drug mechanism of action. It's when you first encounter it. The other interesting thing about neuropsych is it's a hump. You get over it. If you can, you get over it. They tend to never come back. I think if we see that 90 days is, again, quote-unquote, "clean," A, it's immensely encouraging. B, it would seem to be very differentiated from whatever monlunabant generated, with a caveat that we don't really know what they generated. We just have that press release.
In that press release, the impression I think all of us got was, while thankfully none of the horrible adverse events happened, so they de-risked for suicidality and depression, which in and of itself is actually remarkable, they still have some sort of a background squeaky wheel. That would be especially interesting. Novo have created a moat. The advantage of Corbus is not that we're delighted with it because it was a very painful moat, but the advantage of Corbus is that we are on the inside of the moat. If we can show that we are differentiated, and remember, their study was 120 days. Ours will be 90, so it's pretty similar.
If we can show differentiation, then the moat remains and the narrative becomes, see, it's not enough to be peripherally restricted because monlunabant wasn't sufficient at about half the restriction of rimonabant. You need to be highly peripherally restricted. Remember, we are 15 times less in the brain than monlunabant, but with a similar peripheral presence. I think that narrative is both simple and accurate. If the data fits to it, I think it will become quite powerful.
Makes sense. Talking about the peripheral restriction, that's interesting. The CB1 inhibition using antibodies could potentially reduce the brain penetration further than small molecules. What is your view on antibodies targeting CB1 versus small molecules?
It's a great question. It's very hard for me to answer because I'm biased by the mechanism. CB1 inverse agonism with the small molecules is highly clinically validated. It's almost comically clinically validated. Every single one of them led to weight loss. That's not the issue here. Juxtapose that there have been two attempts to target the same GPCR with antibodies. One was a Chinese company called RuiYi that became Bird Rock Bio and is now in the hands of Skye. The other one was Takeda that then became Goldfinch. It's a very different mechanism. It's a completely orthogonal mechanism. To date, that mechanism in either hands, whether it was Bird Rock Bio or Goldfinch, has not shown clinical efficacy, but in their defense, nor was it tested in obesity. We just don't know. We'll see. I think it will be really interesting.
It's just a completely different animal. It's a MAB. Obviously, it has to be given systemically, et cetera. It is just very different than a small molecule that's taken orally that binds to the site and inverse agonizes it. Like I said, the one thing where no one's wondering about is if we're going to see weight loss. It is highly, highly, highly likely that we will see weight loss.
Makes sense. So 913 is entering a competitive obesity landscape. It's dominated by incretin-based therapies, oral GLPs, also gaining traction. What is your vision for where a CB1 inverse agonist fits in as a monotherapy, a combo partner, or a niche option?
All of them. We'll see. Let's start with a niche, right? About 10% of patients are incretin insensitive. They just don't respond. We have no idea why, but they don't. A larger chunk of patients respond. They just can't tolerate it. Then you have another chunk of patients who can respond and can tolerate, but they're at risk of sarcopenia. For all of those, you can imagine a situation where you go, look, I have a pill that you'll take every day. On its own, it's probably not going to be as efficacious as tirzepatide because almost nothing will be. You can't take tirzepatide. You can take this. That's one. That's monotherapy. The second one is combination with incretin pathway drugs. We've shown pre-clinically liraglutide, semaglutide, and tirzepatide. The idea is to augment their efficacy.
This is especially interesting potentially with the oral incretin agonists, which tend to be less efficacious than the injectable ones. This could bring them up to the level of the injectable ones. The last one is maintenance therapy. Simplistically, we have drugs now that are very good at making you lose weight very well, very quickly. Where they struggle is lifelong chronic therapy. They're just not particularly pleasant to be on. There are issues of cost, et cetera. You can imagine a world where you lose weight on an injectable incretin analog. You've done so very well over six months or 12 months. When you've reached your desired weight, you just switch to a once-a-day pill. That's it. It's no different than a statin or a proton pump inhibitor. You feel nothing. The only thing that happens to you is you don't regain weight.
You just stay where you are. My guess, because that market does not exist yet, but my guess is that is probably the largest long-term market, lifelong therapy for weight maintenance. That is where oral small drugs are especially interesting. A reminder to our audience, outside of the incretin pathway, the only other clinically validated small molecule pathway that I am aware of is CB1 inverse agonist. I cannot emphasize this enough. It works. That is not the issue here. Is will it be safe? I think peripheral restriction is a very sensible way of addressing a very specific safety issue.
Makes sense. We're almost out of time. Two minutes left. Maybe just a little bit on 601. Anything on the update that you'll be having in 4Q? Anything on the competitive landscape, maybe from Pfizer and your cash position?
Short cash position is easy. We ended the quarter with $133 million in the bank. We have runway till the middle of 2027. We are immensely, immensely, immensely grateful for them. 601 is an interesting one. It is our highest risk program. It is not a validated mechanism of action. We are not the only ones. Competition are typically big pharma. What is so enticing about it is it is TGF beta. Billions of dollars have been spent on trying to go after TGF beta because it is, other than P53, the most important target in oncology. This, we, Pfizer, Roche, AbbVie, are going after this with a novel mechanism. We are in the clinic. We will have data for a monotherapy dose escalation study. We are setting a high bar for an IO asset. We want to see this thing work on its own.
We will have that data by the end of the year. We will share it. We will see what it looks like.
Great. This has been a great conversation. Thank you.
Thank you so much. I appreciate it. Thanks to our audience as well. Take care.
Take care.