Hey, good afternoon, everyone. My name is Matt Biegler. I'm an analyst here at Oppenheimer and cover oncologies. Really excited to introduce our next company, which is Adagene. Mickael, if you wanna walk us through some slides, and then we can dive into some Q&A if we have time.
Sure. Thanks for having me. I'm Mickael Chane-Du, the Chief Strategy Officer of the company. Cool to be here to talk about Adagene. Maybe I can go into some slide and maybe a quick introduction about what Adagene is about. We are developing a drug in the setting where the standard of care, unfortunately, has a very low response rate, something in the low digit in the setting of interest for us. Very poor overall survival outcomes. We're talking about overall survival between 10 and 14 months. Setting where PD-1 single agent or PD-1 plus something else has not worked.
There's no label for KEYTRUDA in that setting. We have a combination with KEYTRUDA that has so far yielded a response rate between 15% and 30%, depending on the dosage schedule, with a very safe safety margin, as evidenced by the fact that we have a discontinuation rate of less than 10%. So far from the lowest dose cohort, we already have a median overall survival of 20 months and more to come. This is the setup for ADG126, our lead compound, a masked anti-CTLA-4 in combination with PD-1, KEYTRUDA (pembrolizumab) in late-line MSS-CRC without liver metastases. MSS-CRLC standing for microsatellite stable colorectal cancer, right? ADG126, as mentioned earlier, is a masked anti-CTLA-4. The backbone is an antibody called ADG116.
It does differ significantly from ipilimumab, tremelimumab, or a lot of other CTLA-4 in the space due to the fact that it binds to a different epitope, and the differential epitope binding leads to much stronger potency and ADCC. We have a 10x higher ADCC compared to ipilimumab, for example. To that, you had our SAFEbody capacity, our masking technology, which is protease sensitive. It works with, sorry. Activation of ADG126 or cleavage of it is mediated by two things. Number one, high levels of the target CTLA-4, and number two, high expression of certain proteases, member of the MMP family, UPA, et cetera.
I like to say that proof is in the pudding because, as we speak, we are dosing patients at 20 MPK, 20 milligram per kilogram, in combination with KEYTRUDA, which is a really, really high dose compared to historical data and other CTLA-4. Significantly higher dose also compared to the naked antibody, which we used to call ADG116. It's basically ADG126 without the mask. We're very happy with that, proof is in the pudding because, as mentioned earlier, we have, as we speak, very high response, very low discontinuation rate, no grade 4, no grade 5, et cetera, and we will give an update to the market very soon. Before we dive into the context and our data, I would like to remind our viewers that we had an equity investment from Sanofi dating back from July 2025.
Last year, Sanofi committed to an equity investment of up to $25 million. We received the first tranche of $17 million at the cost of $2 a share. The proceeds of that equity investment were allocated to our ongoing randomized phase 2, evaluating ADG-126 plus pembrolizumab in the setting that we talked about, late-line MSS-CRC without liver mets. Not only that, Sanofi committed to a trial collab, they will evaluate, as part of the phase 1, 2, a combination of ADG-126, our compound, with their bispecific PD-1 IL-15 in more than 100 patients with solid tumors. That's number 2. Number 3, keeping in mind that we have had a discovery collaboration with Sanofi since 2022.
Sanofi expanded that collaboration and exercised an option to move forward a third SAFEbody, a third masked antibody. All right? Taking a step back, we believe that it was a significant external validation, coming from Sanofi. Sanofi being a company that knows masking tech pretty well, because if you remember, they acquired a company called Amunix in 2021. They sold the rights to Vir back in 2024, but in 2025, they decided to invest in our company, do a trial collaboration, expand this discovery agreement that we had. Okay, significant external validation, in my opinion. Let's talk about CTLA-4 before we talk about our data in ADG-126.
I would like to remind people that CTLA-4 is a very de-risked mechanism of action from a clinical perspective as well as commercially speaking, because nowadays we have two CTLA-4 therapies on the market, Yervoy, (ipilimumab), and Imjudo ( tremelimumab), both having received multiple FDA approval for several solid tumor indications, and these two drugs together generate close to $4 billion in revenues nowadays. The one thing that I'd like to remind people is that CTLA-4, and here, this is data as monotherapy. CTLA-4 is known to be a very dose-dependent mechanism of action, in the sense that the higher dose is usually better. We had data from BMS and ipilimumab, looking at 10 MPK of ipilimumab versus 3 in the randomized setting of second-line melanoma, and the 10 MPK cohort did outperform the 3 MPK, looking at overall survival.
That's 1 important data point, among others. We all know that CTLA-4 is not about being given as monotherapy. It's really about the combination and especially combination with PD-1, and unfortunately, as evidenced by the data on the right-hand of that slide. Combining EP plus Nivo has led to, unfortunately, synergistic toxicities and very high discontinuation rates due to adverse events, all right? This has been a bottleneck, and a number of players in the industries have recognized that because the 1st generation of CTLA-4 inhibitors have proven to be quite toxic, dosing high, which is normally what you would like to have in combination with an anti-PD-1, is not necessarily doable. As a result, you know, these existing regimens are probably very likely leaving efficacy on the table.
This is evidenced by the fact that looking at bevacizumab, tremelimumab in the exact setting where we are testing our compound, the response rate was unfortunately very low, 2.5%, if you assume that the sole responder in that phase 2 trial was a responder without liver metastasis, okay? Again, that was phase 2 data against PSC in patients with late-line MSS colorectal cancer patients. As mentioned earlier, the standard of care unfortunately also has pretty abysmal outcome, with very limited response in the mid-single to mid-single digit range, median overall survival, rather in the low teens. Looking at patients without liver metastasis, again, our setting of interest in the near term.
Since we're talking about patients without liver metastasis, I want to emphasize the fact that MSS-CRC as well is far from a rare low-hanging fruit. It is a very challenging tumor type for any immuno-oncology agents, as evidenced by the fact that PD-1, Keytruda or Tecentriq in various phase 1, 2, and 3 trials, have not been able to generate more than a 2% response rate as single agent, right? Very difficult, including in patients without liver metastasis. Patients without liver metastasis represent circa 30% of all MSS-CRC cases. We're talking about 10,000 patients in the US, so we believe that this patient population gives, you know, opens the door to a blockbuster opportunity. We do have our eyes also on patients with liver mets.
We just believe that a smarter way to address that patient population is probably through broader combination or broader/novel combination. Stay tuned. I believe that investors should think of late-line MSS-CRC without liver mets, really, as the first attempt of Adagene to tackle this extraordinary challenging but broad market of CRC. All right? Let's talk about data, because I believe that proof is in the pudding. We have evaluated ADG126, our masked anti-CTLA-4, with full dose of pembrolizumab across 4 different dose cohorts. Again, in patients with late-line MSS colorectal without liver mets, we tested 10 mpk Q6W, 10 Q3W. We have a cohort called the loading dose, 20 followed by 10 every 3 weeks. Last but not least, we have 20 milligram per kilogram every 6 weeks.
All these cohorts, again, involving a combination with KEYTRUDA. Okay? We're talking about late-line patients. A third of the patients were fourth-line. Not surprisingly, these patients have not received prior immunotherapy, because, remember, IO agents do not have, unfortunately, a label in MSS-CRC. They do for MSI-H, MSI-H patients, but not for MSS colorectal cancer patients. In terms of mix, we have a healthy mix between U.S. and South Korea, we have not been shy to say that the outcomes between the two regions have been very consistent, looking at response rates in particular. We are focusing on patients without liver metastasis. Now, this is the first important slide that we presented at ASCO 2025. You can see here that across the different dose cohorts of ADG126 plus KEYTRUDA, we don't have grade 4, we don't have grade 5.
The incident of grade three adverse events are very much under control, our discontinuation rate is less than 10%. Okay, I cannot promise that down the road, especially with a much larger sample size, that the discontinuation rate will remain close to 0%. Obviously, we cannot. It does tell you that that table does tell you that we are able to maintain patients on treatment, that ADG126 and its masking the SAFEbody part of the molecule seems to work. We're giving a very, very high dose of an anti-CTLA-4 with PD-1, a dose that should be lethal to patients without a mask, I believe. Safety is a thing, but without efficacy, we cannot really talk about enlarged therapeutic index.
We do not have responses at 10 milligram Q6W, where we started to see them with confirmed responses from 10 MPK Q3W. We had a response rate of 17%, 15% when it's looking at confirmed response rate. At ASCO 2025, we had a response rate confirmed of 29%. Not only we saw a trend towards better response rate with higher dose, we also had a trend towards better duration of response. Stay tuned, we will soon, very soon update this data set with a little bit more patients. A quick focus on the high-dose cohorts, 20 MPK. We had 20 milligram per kilogram every 6 weeks, and the loading dose cohorts, 20 followed by 10 every 3 weeks.
29% response rate in both cohorts. We shared also that longitudinal analysis of adverse events. The bottom line is that we do see grade 3 adverse events with our combination, but these grade 3 adverse events are very well managed. They're transient, as evidenced by the left side of this slide here. We will update these very, very soon with 5 more patients from the 20 MPK Q6W cohort. Same sample size for the loading dose, just a longer follow-up.... Last but not least, let's talk about overall survival, because overall survival is obviously the gold standard, what we care about for the patients.
At the lowest dose, so 10 NPK Q3W plus 10 NPK Q6W, 41 patients, we had a median overall survival presented at ASCO of 19 months, with a median follow-up of close to 18 months. We were very happy with that curve. It compared very favorably with certain historical controls, namely Fruquintinib in the FRESCO and FRESCO-2 trials in the same setting that where we are evaluating ADG126, namely patients without liver mets. Same when you look at bevacizumab, tremelimumab, patient without liver metastasis on that doublet had a median overall survival of 10 months. We will share an update of that overall survival curves. Keep in mind that immuno-oncology agents typically do not shine with very early separation of curve.
We care about the tail of the curve, and on top of it, we will share, for the first time, some landmark analysis for the higher dose, 20 MPK. Keeping in mind that for Fruquintinib in the FRESCO-2 trials had a 12-month OS of only 45%-50%, and at 10 MPK, we had a 12-month OS rate of 70%, so keep that in mind. In terms of catalyst for the next for the next few months, or let's say for the next 12 months, so very soon, during this Q1, so in the coming weeks, we will give an update for ADG126 plus KEYTRUDA, an update to our ASCO 2025 dataset that I just reviewed with you. Stay tuned. We're very excited. We will...
You know, I believe that the investors will hopefully have a broader, a better view of the profile of ADG126, because at ASCO 2025, we had an early sign of efficacy with response rate. Now we're gonna have a better view on durability. Towards the middle of the year, we will have data for triplet combinations, one in combination with KEYTRUDA for Fruquintinib in late-line colorectal cancer without liver mets, and another one is a combination of ADG126 plus atezolizumab versus and bevacizumab. I want to spend a little bit of time on that combination because it's not in CLC, it is in first-line HCC. It came from a trial collaboration that was started, unfortunately, very early on with Roche.
Roche had that trial called Morpheus, where they had a common control arm of atezolizumab, PD-L1, plus bevacizumab, anti-VEGF. We know we have a pretty good view on how, you know, Atezolbev would perform in that trial. Our data, or I would say the Roche data is randomized, so you will have roughly 40 patients who receive Atezolbev, and we have a handful of patients on ADG126 plus Atezolbev. The important thing from that dataset is that because it was stopped early, We just finished the safety run-in, so we have data from the lowest dose, 6 MPK Q6W.
We believe that, you know, this is still an interesting dataset to be shared with the public, and we have the conviction that we can do a significantly higher than 6 MPK Q6W with ADG126 in the context of a combination with Atezolbev, right? Number 3, in terms of clinical news flow, investors should expect towards the end of the year, some initial data from a phase 2 investigator-sponsored trial in Singapore, evaluating ADG126 plus pembrolizumab KEYTRUDA in neoadjuvant colorectal cancer patients. Last but not least, we want to remain very active on the BD front. We want to do more licensing deals. We want to do trial collaborations. We believe that these are potentially very capital-efficient ways to expand addressable markets, evaluate novel regimens, et cetera. That's it.
Nice. Thanks, Mickael. Thanks for running us through it. I have a couple questions. Nice progress, by the way. I want to ask a little bit more about the Sanofi collaboration. I realize that.
Mm-hmm
... it's extensive, and it, you know, you have a pretty close relationship with them. Is there some kind of, like, right of first refusal that they have once your randomized data come out, that could spur some type of, like, co-development agreement structure, or is it more of like a, an outright out-licensing? Yeah, what are really like, the metrics there?
Yeah. It's a good question, and I should have been a little bit more explicit about it. We maintain all the rights to ADG126.
Okay
... and Sanofi does not have a right of first refusal. Sanofi did invest, and importantly, we're talking about Sanofi corporate, not Sanofi Ventures.
Mm-hmm.
Sanofi invested in a company. It is helping us to fund our randomized phase 2 trial that we're doing to comply with the Optimus Project requirements.
Mm-hmm.
Yeah, no, we don't have, we have not sold any rights to Sanofi.
Got it. It sounds like they just invested in your stock, essentially.
Yes. They've done that for a number of companies.
Yeah, I see. That's good. You still own it?
Yes.
You know, ostensibly, though, as you look into the future, the way that I view it, in terms of, like, a blue sky scenario, if you really can tap into neoadjuvant, huge markets are also gonna cost a lot of money to get it developed there. Is that kind of, like, where you're thinking strategic partnerships? You know, what do you think you can kind of do on your own versus what indications are you gonna need a big partner to help usher that forward?
We believe that ADG126 is could be a great natural combination partner, not just with PD-1, okay? We have a lot, we have a wealth of data in combination with pembrolizumab, but also toripalimab. As evidenced by the recent trial collaboration with Sanofi. Sanofi is testing ADG126 with a PD-1 IL-15 bispecific. It's public now, so we can, I can more freely mention it.
Mm-hmm.
We believe that ADG-126 could be combined with PD-1 VEGF, PD-1 TGF-β, PD-1 cytokine. All right? This is enabled by the fact that ADG-126 has a very strong, has a very significant safety margin, thanks to our SAFEbody capacities.
Mm-hmm.
All right? When you keep that in mind, and again, MSS-CRC is far from a low-hanging fruit, it's a very, very cold tumor, PD-1 doesn't work, et cetera.
Yeah.
When you think about the best-case scenario, the upside scenarios, a safe CTLA-4 that can be dosed high, can be very efficient, can unlock the true potential of an anti-CTLA-4.
Mm.
can be combined not just in late-line MSS-CRC. You mentioned neoadjuvant CRC.
Mm-hmm.
I would mention first and second-line CRC. There's a lot of different settings of CRC where we can go. We will have randomized data soon in first-line HCC.
Mm-hmm.
We can think of the warmer tumors where CTLA-4 and PD-1 have a label.
Yeah.
Let's talk about melanoma, RCC.
Yeah
et cetera, you know? To address all these tumor types, I humbly believe that this is better done with a partner.
Yeah. Yeah, I agree. I mean, it's interesting kind of strategic question that you just mentioned, you know, like, we know that IPI works in melanoma.
Mm-hmm.
I feel like the path of least resistance for most small biotech companies would just say, "Hey, I've got a better, safer IPI, so I'm gonna go show that it's better and safer in melanoma." Why did you choose CRC as your first indication?
Let me put it this way. Remember the context. We know that PD-1 single agent doesn't work in that setting, so showing any kind of response above 5%, 10%.
Yeah, it's easy.
... for ADG126 plus pembro or plus toripalimab plus whichever PD-1 you can think of, is a tour de force, right? Showing what we have at the moment, the response rate of, call it 15, 9, 29% as of ASCO 2025.
Mm
In that setting, I would say, is at least very interesting in the eyes of investigators, partners, and potential partners.
Mm-hmm.
You know, keeping that in mind, it has positive reach-through, I believe, to developments indeed in the indications with the lesser resistance.
Mm-hmm. What is a pivotal or, like, registrational trial in third-line CRC look like? Is it against, like, a physician's choice, you know, bev, Lonsurf type of regimen?
It's a good question. After we finish our randomized phase 2, we have a dose for the phase 3. We will discuss with the FDA.
Mm.
The current base case scenario would be a randomized phase 3 with a sole primary endpoint of survival, okay? Now the question is, indeed, what kind of control arm can we have?
Mm-hmm.
It depends on the kind of patient that we go after. If you go after third line, or if you go after third/fourth line or fourth line only.
Mm-hmm.
If you go after third line only, very likely we can talk about physician choice.
Yeah
... including the SUNLIGHT regimen, Lonsurf, et cetera.
Yeah.
If we're talking about late line, fourth line, it could be physician choice, including more frequent.
Mm-hmm
... best supportive care, et cetera. It really depends on the patient population, in my humble opinion.
Yeah.
I would say that we need, number 1, to generate data, and then we will discuss with the FDA. For now, again, the current base case scenario is to have a randomized phase 3 trial looking at OS as a sole primary endpoint. We did notice in certain instances that the FDA allowed sponsors to do phase 3 trials, randomized phase 3 trials with two co-primary endpoints.
Mm-hmm
involving ORR and OS. We're not telling people that this is exactly what to expect, we did notice that precedent, among others.
Yeah, that would definitely speed things up if they allowed that. It'd be interesting to see, 'cause I know that there's some others in your space, like the developers of BOT/BAL, that are kind of broaching this subject now. As it relates to that, how do you kind of see differentiation versus some of the other combo approaches, like, I guess BOT/BAL is the best analogy here?
First of all, I believe that BOT/BAL is Botensilimab is an extremely potent CTLA-4. There's no question about that, in my opinion. I also believe that we probably have a significantly better safety margin, as evidenced by the lower rate of grade 3, lower rate of discontinuation due to adverse event. Which is very key, you know, when you want to maintain patients on treatment, this is how you unlock also long-term survival benefits, okay? I believe the masking technology does allow that enlarged therapeutic index versus BOT/BAL, right? I would say that's 1 thing.
The other thing I believe is, besides the core fundamentals of ADG126, we believe that combining with just PD-1 is just part of the answer. We believe that combining beyond PD-1 is very important for a safe anti-CTLA-4.
Okay. Great. Great. Congrats on the progress. It's always nice to chat with you, and good luck for the rest of the year.
Thank you, Matt. Thanks for the time.
Thanks, everyone. Thanks, everyone, for joining. Bye.