All right, so good morning, everyone. My name is Cory Kasimov. I'm one of the senior biotech analysts here at Evercore, and it's my pleasure to host our next discussion with Immunome and the company's CFO, Max Rosett. Max, thank you very much for being here. It's hard to believe we're already in December, but given the time, I'd love to start this discussion by asking you to reflect back on 2025 and talk about what you see as Immunome's biggest accomplishments over the course of the year.
That's a great question. 2025 has been a year of execution for Immunome. If you look at our portfolio, we brought in AL102 in early 2024 and started guiding towards top-line data by the end of 2025. We filed our IND for IM1021, our first ADC, right at the end of 2024. T his year has really been about running those clinical trials, doing IND-enabling work for the next set of ADCs, and from an investor perspective, it has not had a lot of updates, and so it's really exciting to be getting to the end of the year.
To be getting to the end of the year for 2025 and having data coming for AL102 by the end of this year. We've said by the end of 2025 for 18 months, and that window has now become very, very narrow. W e were able to disclose for the first time that we've seen some activity with IM1021. A s I look at where we're positioned going into 2026, we're potentially going to be submitting an NDA, preparing for a commercial launch, getting to a point where we're not just giving one sentence in a press release about IM1021, but are sharing a real data set from the phase one. A t the same time, we're bringing forward other programs into the clinic on the ADC side.
Got it. Okay. A s you mentioned, we have the phase 3 data for AL102 anticipated any day or week now. Can you remind us of what you expect to be the key points of differentiation as it relates to OGSIVEO, the one product in the market for desmoid tumors?
Yeah. T hat we simply have a more efficacious drug. The things that we're going to be looking at and that physicians will be looking at when we get that data, the first thing I would point to is objective response rate. Desmoid tumors are typically not life-threatening, but they're very, very serious. What patients are looking for is a therapy that controls the tumor, that shrinks it, reduces their pain, gives them functionality, gives them their life back. The first thing I would point to is objective response rate.
Nirogacestat or OGSIVEO from SpringWorks, now marked KGA, was a 41% objective response rate in their phase 3. Our phase 2 with smaller end was 64%. T here's a lot of room to improve on that 41% and go from having a drug where maybe most patients don't achieve an objective response to one where most patients do. I would also point to tumor volume reduction, which in some ways is similar to objective response rate, but is a 3D volume metric measurement of how much the tumor is shrinking.
I f you look at our phase 2 data, the median reduction was 88% reduction in tumor volume per MRI. By contrast, the Niro data was 59%. An analogy we like to use is if you start with a tumor the size of a softball and shrink it 59%, you get to a baseball. If you shrink it by 88%, you get to a golf ball. T hese are pretty substantially different responses, and you're delivering a lot more benefit to the patient when you shrink it.
T his front, it seems like everybody from , everybody talking to investors are focused on response rate, maybe overly focused on that, to the point you're making on tumor volume, things like that, and the T2 signal intensity. These other efficacy endpoints, how much do they resonate at current time within the physician community, and how much education will be necessary on your part to talk about these other benefits of AL102?
That's a really good question. T hat one of the reasons we talk a lot about objective response rate is it is well-known and well-understood both on the investor side and on the physician side. When you talk to physicians who really know this space, they do get very excited about some of these more esoteric secondary endpoints like T2-weighted imaging. I actually, your initiation report had a great quote from a KOL about how T2-weighted imaging corresponds with what they're looking for. J ust in brief, T2-weighted imaging is a way of looking at MRIs to determine whether tissue is still actually. It measures cellularity. A re these still active tumor cells, or is this just collagen? O ne of the things that you'll see happen with a drug like AL102, so you have a tumor, and you essentially kill the tumor, but it's replaced with collagen. A system like RECIST maybe shows that as stable disease, but the doctor is saying all the pain has gone away, the range of motion has come back.
Clinically speaking, this patient has had a full response, and that shows up in T2-weighted imaging, but not necessarily in objective response rate, and so certainly we'll be looking at that. I could see T2-weighted imaging being something that we defer to a full conference presentation. We care very much about saving enough to be able to do a really great conference presentation. It's the right thing to do for the doctors. It's the right thing to do to set up a successful launch of the drug, but yeah, some of these less well-known metrics actually turn out to really correlate with treatment, with the benefit derived by the patient, and doctors do care about those,
To follow up on that, because I get this question a lot from investors, is how much information do you plan to disclose in the top-line press release?
Yeah. We haven't given the exact list, but just to help you understand how we're thinking about it. Sometimes a top-line press release is just, we hit our primary endpoint, more data to follow. Given the way we talk about this drug, you and I haven't even talked about the primary endpoint, which is progression-free survival. G iven how much we talk about these secondary endpoints, obviously we can't do that. Ob jective response rate, tumor volume reduction, these are things that we think are essential for understanding the characteristics of the drug.
That said, there is a tension and an ongoing dialogue with our clinical team about, okay, what do we have to hold back to have a really good conference presentation? W e understand that people need to come out of this top-line press release understanding that this drug is as good as we think it is.
Got it. Okay. W ith regard to how you define events in your RINGSIDE trial, I believe you're looking at only radiographic progression, right? Whereas DeFi with OGSIVEO used radiographic plus clinical progression. How do you believe this might impact the overall event rate, if at all?
That's a great question. I'm going to take a moment to clarify something, which is this difference in definition that we're about to go through. This is only for progression-free survival. It's only for the primary endpoint. When I talk about objective response rate, when I talk about tumor volume reduction, all of these secondary endpoints, those are defined identically between the two trials. Objective response rate is per RECIST.
Totally unaffected by what I'm about to discuss. Our primary endpoint, though, is progression-free survival. T o have progression-free survival, you need a definition of progression. The gold standard and what we're using is radiographic progression, right? H ow's the tumor grown by a certain amount on a CT scan? What SpringWorks did with their trial is they used a composite endpoint. T hey counted both radiographic progression events and clinical progression events. Clinical progression is inherently a little bit more subjective, right? It's like, hey, my symptoms have gotten worse. I don't think the trial is working for me. I'm going to go off trial and unblind. O verall, we feel like we have the more rigorous metric, and certainly the FDA supported or required us to use this metric. In terms of what it will actually show in the data, it'll be pretty subtle.
If you look at it, it is fairly unusual to have severe clinical progression to the point where you would go off a trial without the tumor also growing and that showing up on imaging. I f you look at the OGSIVEO data, they had a hazard ratio of 0.29 on the composite endpoint. W hen you do the sensitivity analysis and ignore the clinical progression events and do the apples-to-apples number, it's 0.31, 0.29, 0.31. Basically both 0.3. I t's there. We've talked about this because people notice it, and we want them to understand. I t doesn't impact the secondary endpoints in the slide test. T he effect on the primary endpoint is quite subtle.
Got it. That's very helpful. All right. A t the end of the day, what would you consider the "win" in this trial?
For us, a win is coming out of this with a drug that is clearly better than OGSIVEO. Y ou'd love for me to say, oh, it needs to be exactly this objective response rate or exactly that. W e're going to be looking at the totality of the data. O kay, OGSIVEO has a 41% objective response rate. If we come out with a 43% objective response rate, I'm not going to come back to you and say that's clearly differentiated. W e'll know it when we see it in terms of this is a better profile.
Yep. Okay. T he last question I have on this for now is the market opportunity. How do you see the desmoid tumor market opportunity evolving? W e've gotten physician feedback, right, where even if these were the same drugs, it's a once a day versus a twice a day, it was difficult to ascertain differences between the two that it would still be used. H ow do you look at this market opportunity if you have that win where it's a clearly differentiated product versus one where it's a little bit more nuanced and subtle?
Yeah. This market opportunity, we're still in the early days of really building out the desmoid tumor market, right? If you look at the number of patients who are diagnosed each year, that's maybe 1,600 or so. T hese patients live with the disease for a very, very long time. You hear anecdotes about patients who've had this disease for 30 years. T he prevalence pool is substantially larger.
At this point, OGSIVEO has treated maybe 1,000 patients, so there's a long ways to go in terms of really displacing not necessarily other systemic treatment therapies, but things like surgery. Surgery is not actually indicated for these tumors in most cases, so displacing that is important. The patients who are getting active surveillance because essentially there's nothing better to offer them, and so there's a lot of opportunity, not just to displace OGSIVEO as a head-to-head comparison within systemic treatments or within GSIs, but to grow the market by providing a better option for patients who are not getting a systemic treatment at all right now.
Okay. Got it. All right, so let's move over and talk about the ADC platform and what you're trying to build here. T o begin and level set everybody, how is this platform maybe differentiated from what CGEN built and the numerous other competitors that are out there now in the field?
Yeah. A nytime you're looking at the advance of technology, it's like, okay, what are the great things that you're building on? W hat is it that you're doing new? HC-74, which is what we call our payload, what we're building on is it's a Topo I inhibitor. It's a great class. It's a class that has multiple approved drugs, tends to have really nice efficacy, but also a much broader therapeutic index than some of the earlier ADC platforms. CGEN was their big technology was vc-MMAE, and it's great. It's led to some great drugs, but has a much narrower therapeutic index.
We really like Topo Is because they have responses in a lot of different tumor types, and you can dose them quite a bit higher. What we're doing differently and why we feel like we're innovating within the Topo I class is we're really focused on overcoming resistance. I f you look at ENHERTU, which I would argue is the most successful Topo I ADC at this point, and you segment breast cancer patients by how much they express a certain efflux transporter. Y ou look at the ones who are low expressing, bottom quartile expression, 47% objective response rate. If you look at the ones who are high expressing, 17% objective response rate. Y ou lose two-thirds of your responses if this efflux transporter is upregulated. T here's similar data that shows that it impacts durability of response as well.
One of the things that we are incredibly excited about with HC-74 is it is not a substrate for that efflux pathway. W e've shown that in vitro. W e've also done things like look at mouse models that are refractory both to DXd, to ENHERTU essentially, and to irinotecan, which is also a Topo I inhibitor. W e still have efficacy in those models because we're overcoming this resistance pathway. The other thing I would talk about is bystander effect, right? Real-world tumors are not homogenous, right? Some cells will express your target. Some will have lower expression of the target. A good bystander effect lets you kill all of the cells in the tumor, both target positive and target negative. It only does that in a localized way because the drug is only being released within the tumor.
We've been able to show that with preclinical models of tumor heterogeneity, we have really nice efficacy while still having a safe payload. W e're building on this great class of Topo I inhibitors, but we're trying to overcome the biggest challenges that limit the success of that class.
Okay. It's interesting. W hen it comes to actual products, then the lead ADCs, as you referred to before, IM-1021, targeting ROR1, how might this be differentiated from Merck's ROR1 asset?
Merck's ROR1 asset was built on the same technology I already referenced, the Seagen technology of vc-MMAE. My colleagues know that technology very, very well. W hat they were able to show with that program in lymphoma is some efficacy. What they were limited by was toxicity, right?
If you look at mantle cell lymphoma, they got up to a 43% objective response rate in mantle cell. T hat was at a dose of 2.25 or 2.5 mg per kg where they were having 80% grade three or grade four adverse events. Y ou just can't bring that forward. W hat we look at with that, it's like, okay, there's something to learn about the target there, right? This is a target that has a real path to activity in lymphoma, but you need something with a broader therapeutic index.
O ne thing that captures how much safer our molecule is, is that we were able to start this escalation at two mg per kg compared to 1.75, which is what Merck is taking into phase three, right? That just speaks to the difference in safety profile. W e're really excited to continue to advance that, especially on the B-cell lymphoma side.
Okay. W hen you say your colleagues know it well, can you remind us, roughly speaking, how many people at Immunome came from CGEN, like in the science room?
Yeah. O verall, probably about a third of the company has some connection to CGEN. Doesn't mean they directly came from CGEN, but people with deep experience. Our early research and discovery group has a lot of people who were really excellent at CGEN and really understand how to select an ADC target and build an ADC that is going to have as high a potential as possible of succeeding in the clinic.
Okay. Yo ur third quarter press release teased us a little bit, noting responses across different doses in this study. Should we expect to see that more thorough update at some point in 2026?
Yeah. it's fair to say that at some point in 2026, at a suitable conference, we will have a real data set, right? We understand that one sentence saying, "We've seen objective responses." It's encouraging, and we want to put it out there because we get a lot of questions. T hat's not a data set. U ltimately, what people are waiting for is a real data set of what we're seeing.
Got it. H ow's progress going with regard to advancing other ADCs into the clinic?
We're expecting to have multiple additional INDs throughout 2026. IM-1617, I think we'll get that IND filed fairly soon, and then two others, IM-1340 and IM-1335. T hey're novel targets. They're built with the same technology. All three of them have solid tumor potential. W e're pretty excited to get them into the clinic and see how they work.
If you're fortunate enough to have success with ROR1 with IM1021, do you think it provides pretty good read-through to the other assets, even though different targets, different indications?
Yeah. T hat when you have success with the program, that really shows you're running a program, but you're also running an experiment about the platform and the technology in it. Now, if IM1021 were to fail for target-related reasons that has less read-through. I f you have a successful program, you've proven that the platform is capable of delivering success.
Okay. I've been wrapping up these conversations by asking management teams to give us a little sneak peek into upcoming events that you might have. I t's a little obvious for Immunome, so I can spare you the question. It's a good thing because we're out of time anyway. G ood luck with the data, and thanks for being here.
Yeah. Thank you.