Compugen at H.C. Wainwright: AI-Driven Immuno-Oncology Strategy

On Friday, 05 September 2025, Compugen Ltd. (NASDAQ:CGEN) presented at the H.C. Wainwright 27th Annual Global Investment Conference, outlining its strategic focus on immuno-oncology. The company highlighted its AI/ML-driven Unigen™ platform for drug discovery, its clinical-stage pipeline, and collaborations with AstraZeneca and Gilead. Compugen’s approach showcases potential strengths in novel target discovery, though challenges remain in competitive landscapes.

Key Takeaways

• Compugen’s Unigen™ platform is central to its strategy, focusing on novel drug target discovery.
• The company has approximately $94 million in cash, providing a financial runway into 2027.
• Key programs include COM701 in ovarian cancer and COM902 in partnership with AstraZeneca.
• The MAIA study aims to improve progression-free survival in platinum-sensitive ovarian cancer.
• Compugen differentiates its TIGIT antibody, COM902, from competitors by its non-FC binding properties.

Financial Results

• Cash Position: Approximately $94 million as of June 2025.
• Cash Runway: Expected to fund operations into 2027 without considering potential royalties or milestone payments.

Operational Updates

• Lead Program: COM701

- Target: Platinum-sensitive ovarian cancer patients.

- Study: MAIA study designed to compare COM701 monotherapy to placebo.

- Goal: Improve progression-free survival by more than three months.

- Interim Analysis: Planned for the second half of 2026.

• Partnerships

- AstraZeneca (relvegostomig):

- Multiple Phase III trials underway.

- Upcoming data presentations at ESMO.

- Gilead (GS0321):

- Phase I study led by Compugen is ongoing.

Future Outlook

• Catalysts

- ESMO 2024: Pooled analysis of Phase I studies with COM701.

- MAIA Study Readout: Expected in the second half of 2026.

- TIGIT Landscape: Monitoring developments from Arcus and Gilead.

Q&A Highlights

• COM701 in Platinum-Sensitive Ovarian Cancer:

- Rationale: Good safety profile and potential to prolong or prevent relapse.

• MAIA Study Design:

- Objective: Enhance progression-free survival in targeted patient group.

Compugen’s strategic presentation at the conference underscores its commitment to leveraging AI in drug discovery, with promising collaborations and a robust pipeline. Readers are encouraged to refer to the full transcript for detailed insights.

Full transcript - H.C. Wainwright 27th Annual Global Investment Conference:

Unidentified speaker, Interviewer: Greetings, and thanks for joining us to have a conversation with Eran Ophir, CSO of Compugen Ltd. Compugen Ltd. is a clinical-stage immunotherapy company that utilizes AI/ML-driven predictive tools to discover novel immuno-oncology drug targets. They have a rich pipeline, as well as development partnerships with AstraZeneca and Gilead. To discuss the company’s development strategy in 2025 and beyond, I welcome Eran to this fireside chat. Good day, Eran. Glad to see you and appreciate you accepting our invitation to talk to our audience today.

Eran Ophir, CSO, Compugen Ltd.: Hi, good to see you and thank you for inviting me.

Unidentified speaker, Interviewer: Yeah, Eran, as we can get our audience to familiarize themselves with Compugen, can you give us an overview of the company and the company’s current area of focus?

Eran Ophir, CSO, Compugen Ltd.: Sure, maybe I can share a slide to give the overview. Compugen Ltd. is a clinical-stage company focusing on immuno-oncology. What we do is use our computational discovery platform called Unigen™ to identify novel drug targets in the field of immuno-oncology. We identified TIGIT around the same time Genentech did. A few years later, we identified PVRIG, another first-in-class potential asset that we use COM701 to target. We developed COM902, an anti-TIGIT antibody that we think is from the right class of TIGIT antibodies, and we’ll discuss it a bit later. We have these fully owned two clinical assets, COM701 and COM902. You also license COM902, or the right to use COM902, which is a potential best-in-class anti-TIGIT antibody, as part of a bispecific for AstraZeneca. They use COM902 as part of their PD-1/TIGIT bispecific molecule, relvegostomig.

They are now in Phase III trials with this molecule, using it really as a new backbone to combine in different indications. This is one collaboration. A more recent collaboration is with Gilead, in which we identified, using again Unigen™, a computational platform, another potential first-in-class antibody that was called COM503; it’s now called GS0321. This asset was licensed to Gilead in the preclinical stage, and it’s now in a Phase I study. It’s a really exciting study to a novel approach to harness cytokine biology for the treatment of cancer. Obviously, you have multiple other earlier stage assets, and we continue to use our computational platform to identify novel drug targets to bring more new treatment options to patients.

Unidentified speaker, Interviewer: Perfect. Regarding your computational platform, Unigen™, what are the platform’s main offerings? Also, can you give us some examples where you’ve already validated this platform?

Eran Ophir, CSO, Compugen Ltd.: I think that people can use AI and computation to do many things along the R&D process of molecules. Our unique value proposition is that we focus on target discovery. We all know that there are a very limited number of targets out there for clinical testing. Each target, there are multiple biotechs and pharma companies, but there are very few new targets. What we do, we use a computational platform, Unigen™, and we identify novel drug targets. By that, we bring new options in innovation to clinical testing in oncology. We built this platform along years. We have built a dedicated database with unique data that we collected. We have the expertise, both computationally, also in the lab. How can you really validate and discover new biology in an efficient way?

Most importantly, we do not only say that we use AI and computation to identify new drug targets. We have actually shown again and again, and the platform is validated because we are able to bring already a number of few assets from actual computational prediction to clinical testing. Most importantly, the value of this target is exemplified by the collaboration we have with pharma companies along the years, including a very recent deal that, as mentioned before, we did with Gilead late last year.

Unidentified speaker, Interviewer: Very good. Regarding your lead program, you know, COM701, which is actually a PVRIG antibody, as you said, you recently initiated the MAIA ovarian cancer study. Can you help us understand not only what PVRIG is, but also why it is a good target for ovarian cancer?

Eran Ophir, CSO, Compugen Ltd.: When we discovered computationally PVRIG, there were no papers around it at all. We uncovered most of the biology ourselves. While unlocking the biology, what we realized and discovered is that PVRIG is a T-cell to NK checkpoint, a very, very different one from the other checkpoints, very different biology. Without going into all the details, this biology enables a blocker antibody like COM701 to unleash waves of T-cells into less inflamed tumor types, like ovarian. By that, we believe we can see activity in places where other checkpoints have failed or haven’t shown activity, especially not in monotherapy. Specifically for ovarian, out of the other non-inflamed, less inflamed tumor types, the PVRIG pathway is highly expressed in ovarian cancer. It was a target indication for us before we even started. The clinical signals. We started our clinical trials because ovarian was a target indication.

We did some expansion cohorts in that indication. The clinical signals were there from monotherapy signal, COM701 alone, being able to drive clinical response and also translational response, really seeing the biology in the patient sample screen on treatment. Also moving in other combinations, we continue to see this activity in last-line platinum-resistant ovarian cancer patients. I think it’s the biology and then the clinical signals in the last-line patients.

Unidentified speaker, Interviewer: Okay. Right now, I believe you’re working on conducting the study in the platinum-sensitive ovarian cancer. What makes COM701 a good therapeutic to treat these women with platinum-sensitive tumor?

Eran Ophir, CSO, Compugen Ltd.: It’s a very good question. First, as mentioned, it’s the biology and the clinical signals in the platinum-resistant settings. When we really looked at our data in the platinum-resistant patients, some of them progressed very fast. They had a highly progressing tumor, probably an exhausted immune system, a compromised immune system by all the multiple prior lines of therapy. They probably never even had a chance to respond to a chemotherapy-free regime like the one we used in these patients. The patients who did respond had really durable responses with excellent safety profiles. We thought that the exact right settings for this kind of agent could be going earlier in the platinum-sensitive. In the MAIA ovarian cancer study, we’re taking patients who responded to chemotherapy. They have lower tumor burden, less compromised immune system.

Now we can give COM701 as maintenance treatment and really rely on the strong biology, on the durability of response in the patient who does respond, and again, the excellent safety profile that might be ideal for maintenance settings.

Unidentified speaker, Interviewer: For this patient population, what is the current standard of care? I believe some of these patients are also either BRCA wild type or mutant. Do they have a specific standard of care where you think COM701 actually could do better?

Eran Ophir, CSO, Compugen Ltd.: What is really interesting is that what we identified with multiple discussions with our KOLs and the clinicians who worked on our studies and the previous studies in ovarian, that there’s really an unmet need and there is no standard of care for this patient. I will explain. Patients with ovarian cancer, metastatic ovarian cancer, they receive surgery and then the chemotherapy. If they respond to the chemotherapy and they are still platinum-sensitive, at certain stages, many of them will relapse. If they are BRCA positive or negative, HRD positive or negative, they might receive BEV or PARP as maintenance. In the second or third line, if they already receive BEV or PARP, or they’re not eligible for many medical conditions, there are really no other maintenance options. They receive the cycle of chemotherapy, the platinum chemotherapy, four to six cycles. There is no standard of care.

They basically sit and wait to most of them will eventually relapse. Either they will relapse before six months, and then they’re defined as platinum-resistant, or maybe they’re still platinum-sensitive. They receive the six cycles again. We think this is exactly the place when a drug which has very good safety and durability of response like COM701 can prolong this time from platinum-sensitive until relapse, or ideally, actually maybe even prevent the relapse.

Unidentified speaker, Interviewer: Please discuss the design of the MAIA ovarian cancer study, especially with some additional thoughts on what sort of patients you’re enrolling into the study, as well as if we can use or if you are planning to use PVRIG as a biomarker as well.

Eran Ophir, CSO, Compugen Ltd.: We’re talking about platinum-sensitive patients, patients that have received BEV or PARP or are not eligible to receive maintenance with BEV or PARP. They receive the platinum, they responded to platinum, so they have low tumor burden. We also excluded patients with liver metastases. Overall, we are taking earlier line patients with immune systems that are supposed to be less compromised. We randomize; it’s a blinded study and randomized study. Either the patient will receive a placebo, because there’s no standard of care, so we can actually give them nothing, a placebo, or they are receiving COM701 monotherapy, and they’re randomized two to one. The goal is eventually to see in the interim analysis that we are improving the progression-free survival by more than three months. This is the success criteria, which is a stringent one, but we really want to see potent activity.

Can COM701 indeed make a significant clinical impact in this patient population?

Unidentified speaker, Interviewer: Okay. I believe you are expecting to do an interim analysis sometime in the second half of 2026. In that analysis, what sort of data could you be talking about? Do you think that data will help us kind of think about where this study is going to end?

Eran Ophir, CSO, Compugen Ltd.: Since we have quite solid historical control for this patient population, their PFS of the placebo arm is supposed to be around six months. Already in their interim analysis, we think we should have the data that will allow us if it worked or not. We should have the placebo around six months, and we expect to see COM701 improving on that, again, ideally by three months and more. We should have this data in half 2026.

Unidentified speaker, Interviewer: Okay. Since this is going to be a monotherapy study, when you get some data either at the interim analysis or at the completion of the study, do you think you may be required to do a combination study? Do you think COM701 is good to be used as a monotherapy, at least based on the data that you have so far?

Eran Ophir, CSO, Compugen Ltd.: This is a very good point. What we did in this study, we did an adaptive trial design, meaning we have a lot of flexibility. If COM701 shows very potent activity, and because it’s a blinded study, actually, we can even add more patients blindedly and consider accelerated approval. Having now more idea on the magnitude of effect, we can stop the study and then open a Phase III study to actually use COM701 in monotherapy. We can also add more arms. After establishing that COM701 as monotherapy has a significant clinical impact on patients, clinically meaningful impact, then we can add other combinations. It could be BEV. It could be, again, going to the triplet of the TIGIT PD-1. We have the flexibility to adopt whatever kind of combination we want that we think will be best also in terms of competitive landscape that will be by half 2026.

Unidentified speaker, Interviewer: I believe at this year’s ESMO, we are expecting to see some data. What sort of data should we expect there? I know the complete abstract is not out there yet, but to the extent you can talk, it’ll be helpful.

Eran Ophir, CSO, Compugen Ltd.: For ESMO, we looked back at the Phase I studies we already published for COM701. We did a monotherapy study and did combination in doublet and triplet. We pulled all the data and tried to see how we see the fingerprints of COM701 activity and biology. Looking at all the data, it would be nice to examine what we see independently in the different trials independently about, again, the safety profile and durability of response. Looking at the totality of the data together could probably inform on why we decided to move into the platinum-sensitive setting in the MAIA ovarian cancer study in less compromised patients with less compromised immune system.

Unidentified speaker, Interviewer: Okay. Moving on to some of your partnerships and starting off with the AstraZeneca one, the company or the partner is evaluating relvegostomig, right? That is actually designed to have the TIGIT component of the 902, which is an Fc-reduced TIGIT antibody, which also, I believe, came from the Unigen™ platform. TIGIT has been having a tough time. We have seen more companies give up on TIGIT development. What makes not only you and your partner, AstraZeneca, continue evaluating the relvegostomig in the clinic?

Eran Ophir, CSO, Compugen Ltd.: I think what is very important to note is that not all TIGIT antibodies are the same. Carefully looking at which company discontinued their TIGIT molecules and which ones are continuing, one can easily see that the companies that have TIGIT antibodies from the class that binds FC receptors have the potential to deplete cells to the good and for the bad. These companies have shown a much less favorable safety profile and a higher rate of discontinuation. Roche, Merck, and BeiGene, all of them discontinued their TIGIT program. The active TIGIT programs now are Gilead and Arcus, which have their non-active FC binding, the relvegostomig molecule, which also is from the class that doesn’t bind FC receptors, and in earlier clinical trials are on COM902. I think this is one, the type of TIGIT antibody, which is probably the right way to tackle the TIGIT pathway.

Also, AstraZeneca is using a bispecific molecule. Not only does this help them from a regulatory standpoint to some extent because they don’t always have to prove the contribution of components, but also they have shown cooperative binding of that molecule, potentially having better activity than just combining TIGIT and PD-1. The data that we’ve shown till now, I mean, they have shown, again, non-randomized studies, but excellent safety profile and very good activity. Maybe I can share for a second the clinical strategy. What AstraZeneca is doing is really using relvegostomig as a new IO backbone. They are now combining it across different indications. You can see the phase, so 10 phase III studies across different indications. In some of them they are comparing directly to pembrolizumab. In some of them they are combining with ADC, some in chemo, some different settings, some with inhibitor.

Overall, I think that probably AstraZeneca is quite confident because probably they wouldn’t start this aggressive clinical strategy. Some of the data supporting this clinical strategy have already been published. Some of it probably will be published in the near future. For example, in ESMO, they have additional two posters, one of the ARTEMIDE-01, more data, or probably more follow-up on data already disclosed in the ARTEMIDE-01 study, and an initial look into bladder cancer in which they combine relvegostomig with an ADC. Overall, I would say that we are quite confident that if not most, then some of these phase III studies are going to be successful. We also have to look at how we’re going to prepare our course. This will also reflect on our own COM902.

Unidentified speaker, Interviewer: Okay. It would be good to see some additional data come from AstraZeneca at ESMO. In terms of the relationship with Gilead, where they have licensed what they call GS0321 now, the deal structure itself was interesting. It makes up to nearly $850 million. In general, what’s the deal structure? Also, what is your responsibility within this partnership?

Eran Ophir, CSO, Compugen Ltd.: It’s a very interesting, unique deal structure because we received on a preclinical asset $60 million upfront. We continued the development, then we received additional $30 million upon IND approval. We’re eligible, as you mentioned, for additional roughly $750 million in milestone and up to low double-digit tiered royalties. What is unique here is that we are leading the Phase I. Even though the asset is fully licensed to Gilead, they thought for such a first-in-class asset that we have the biology, we have the expertise, they already saw the capabilities of our clinical team. They trusted their molecule in our hands because both of us thought together that the right thing for the molecule is for us to take it into the first stages of the clinic. We’re leading now the Phase I. We dosed the first patient earlier this year.

This is a really exciting study, and we’ll see how it develops.

Unidentified speaker, Interviewer: Okay. Can you describe to the audience what GS0321 is? What’s the mechanism of action? How did you identify this target?

Eran Ophir, CSO, Compugen Ltd.: We used Unigen™, our computational platform, to look for a resistant mechanism in a tumor environment. We are not looking at all for cytokine targets. Actually, we thought that cytokines are problematic drugs. They have a terrible therapeutic window. This was not the goal of our computational search. We identified that in a tumor environment, there is a very potent anti-tumor cytokine called IL-10 that is found there in an inactive form, complex with IL-10 binding protein that binds it with a high affinity and prevents its activity. What we realized is that what we could do is develop a blocker antibody that inhibits the inhibitor. I believe we were the first to do so in general, and for sure specifically for IL-10 binding protein, to unleash the natural activity of IL-10.

By that, we really harnessed cytokine biology for the treatment of cancer in a very different way because we are modulating only the tumor environment because IL-10 is expressed in the tumor environment and not in the periphery naturally. We are not modulating at all the periphery. Preclinically, it was very exciting. Clinical signals probably are, I mean, the clinical study is ongoing now.

Unidentified speaker, Interviewer: In terms of the clinical study itself, the Phase I study, what different tumor types are you evaluating? Also, since you started the study in late 2024, should we expect any clinical data update during this year or early next year?

Eran Ophir, CSO, Compugen Ltd.: We started the first patient in the beginning of 2025. It’s a monotherapy in combination with Zym. We have some expansion cohorts and the indications are not disclosed. Overall, I would say that preclinical data have shown that many indications are relevant. Probably it would not be the best to start with a very, very desert. We need maybe some level of immune infiltration. Again, it could be quite broad. About the exact timing, it’s going to be very dependent on the dose, what will be dosed for monotherapy, for the combination. I think it’s a bit too early to commit for an exact time. The trial is progressing nicely. We’ll definitely, at a certain point, report data.

Unidentified speaker, Interviewer: In terms of catalysts, what should investors be looking out for, not only in the rest of this year, but also in 2026?

Eran Ophir, CSO, Compugen Ltd.: First of all, we have ESMO, in which we’re going to report the pooled analysis of our phase I studies with COM701. This, again, maybe will shed a bit more light on the reason we moved into the MAIA ovarian cancer study in the platinum-sensitive patients. The MAIA ovarian cancer study is going to have readout in H2 2026. Being maybe one of the only companies with a non-active Fc, active TIGIT program, I think that in general, TIGIT readouts are going to be important. Arcus, Gilead, with their, I have a phase II readout or OS readout this year and phase III readout next year. Obviously, relvegostomig readouts. We have an ESMO and probably we didn’t disclose exactly what and when, but more data will be disclosed along time. I think we have our own internal catalyst, but in general, TIGIT landscape catalysts that are going to be relevant.

Unidentified speaker, Interviewer: Very good. To close out, in terms of your balance sheet, how strong is the cash position? What sort of a runway do you project from that?

Eran Ophir, CSO, Compugen Ltd.: At the end of June, we had roughly $94 million in our cash. This will allow us, in a very conservative approach, we’re not relying, we’re not discussing any royalties from any milestone from our collaborations or new deals. Just this amount as is will allow us to go into 2027 and to support all our activities from the early pipeline to the clinical trials. Into 2027, we think we’re in a good cash position.

Unidentified speaker, Interviewer: Very good. Thank you very much, Eran. Thanks for spending the last half hour with us. Appreciate it. Good luck and talk to you soon.

Eran Ophir, CSO, Compugen Ltd.: Thank you.

This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.