Compugen Ltd. Stock

Good morning, everyone, and welcome to Compugen's COM701 in Ovarian Cancer and Unigen Platform Update Event. [Operator Instructions] Before handing the call over to the company, please review the forward-looking statements on this slide.

I'd now like to introduce Dr. Eran Ophir, President and Chief Executive Officer of Compugen. Eran, please go ahead.

Good morning, everyone, and thank you for joining us. I'm Eran Ophir, President and CEO of Compugen. Today, we'll provide a brief overview of Compugen's key value drivers, discuss the clinical opportunity for COM701, our first-in-class anti-PVRIG antibody being studied as maintenance therapy in second- and third-line platinum-sensitive ovarian cancer, and share an update on the continued evolution of Unigen, our AI-powered discovery platform.

Dr. Oladapo Yeku, the lead investigator on our MAIA-ovarian adaptive platform trial from Harvard and Mass General Hospital will share his perspective on the evolving treatment landscape in ovarian cancer, setting the stage for the interim analysis of the MAIA study with PFS readout that is expected to be reported by Q1 2027.

Later on, Dr. Sharon Kredo-Russo, our Senior Vice President of Research and Discovery, will discuss how we are expanding Unigen's capabilities to support Compugen's next generation of immune oncology discovery. While we will not be disclosing any new targets today, we believe it is important to keep you updated on the capabilities we are building in this area of high unmet needs.

Following presentations from Dr. Yeku and Dr. Kredo-Russo, Dr. Michelle Mahler, our Chief Medical Officer, will join us for the Q&A session. So let's get started.

This slide highlights Compugen's main value drivers today, all rooted in our discovery capabilities and derived from our AI-powered computational target discovery engine called Unigen. First is COM701, our wholly owned potential first-in-class anti-PVRIG antibody, currently being evaluated as maintenance monotherapy in second- and third-line platinum-sensitive ovarian cancer in the MAIA-ovarian adaptive platform trial, which Dr. Yeku will focus on.

Second is COM902, the TIGIT antibody discovered following our computational identification of TIGIT in 2009. COM902 became the TIGIT component of AstraZeneca's bispecific antibody, rilvegostomig, that AZ is currently evaluating in 12 Phase III trials. Compugen is eligible for up to $195 million in regulatory and commercial-based milestones and up to mid-single-digit royalties on rilvegostomig, which AstraZeneca has previously projected to have peak year non-risk-adjusted revenue of over $5 billion.

Third, we're advancing GS-0321, formerly COM503, which was licensed to Gilead in 2023. This program also originated from our computational discovery platform, Unigen. GS-0321 targets the IL-18 binding protein and represents a novel approach to harness cytokine biology for the treatment of cancer. To date, we have received $90 million from Gilead, $60 million upfront and $30 million upon IND acceptance and remain eligible for up to an additional $758 million in development, regulatory, and commercial milestones and single to low double-digit royalties on worldwide future net sales.

Altogether, these programs and partnerships validate Unigen as an engine for identifying novel immune oncology drug targets. Unigen discovered targets currently make up our undisclosed research pipeline. And through our expanded capabilities, we are working to continue to bring more innovative molecules into areas of unmet need for patients with cancer.

With that broader context in mind, let me focus on COM701, our most advanced wholly owned program and why we believe PVRIG inhibition is relevant in ovarian cancer. In this slide, we summarize this biological rationale, including the unique dominant expression of PVRIG on stem-like memory T-cells, which is different from other checkpoints like PD-1 and TIGIT. Therefore, by blocking PVRIG, COM701 is designed to alter the tumor microenvironment, including in less inflamed tumors such as ovarian cancer. Accordingly, we have observed antitumor activity of COM701 in patient populations typically less responsive to conventional checkpoint blockers like PD-L1-negative ovarian cancer.

So as is typical for new drugs, we began our COM701 clinical trials in heavily pretreated platinum-resistant ovarian cancer patients. In these early trials, COM701 in monotherapy and combinations induced clinical benefit and durable responses regardless of PD-L1 status. As you can see here in these turquoise bars that represent clinical benefit in patients with PD-L1-negative disease. Importantly, PD-L1-negative platinum-resistant ovarian cancer patients historically showed zero overall response rate even when PD-1 and TIGIT blockade were tested in combination, highlighting COM701's unique mechanism of action. In addition, we identified retrospectively that patients without liver metastases appeared more likely, as you see here on this slide, almost 40%, to benefit clinically from COM701 even in these advanced settings.

These observations informed the MAIA-ovarian design, where we decided to enrich for patients more likely to respond to COM701 by going earlier in the treatment algorithm on the very right side of this graph, heavily pretreated platinum-resistant ovarian cancer patients into the second- and third-line maintenance settings. We select patients with no liver metastases. Their immune system is less compromised by multiple rounds of chemo and they have low tumor burden as we select only patients that responded to the last-line platinum chemotherapy. These patients have an unmet need for a drug with strong safety and durable antitumor activity to prevent or delay tumor recurrence following response to platinum-based chemotherapy.

So the MAIA-ovarian is designed to evaluate COM701 monotherapy against placebo in these defined maintenance settings with 40 patients randomized to COM701 and 20 to placebo. The goal is to determine whether COM701 can extend the period of platinum sensitivity, delay progression to platinum-resistant disease, and help preserve future treatment options while maintaining a tolerable safety profile for patients in the maintenance settings. As mentioned, data from this trial is expected to be reported by Q1 2027.

With that context, I am pleased to turn the call over to Dr. Yeku from Harvard University and Mass General Hospital. Dr. Yeku will share his perspective on the evolving treatment landscape in ovarian cancer and the rationale for evaluating COM701 in the maintenance settings. Dr. Yeku, the floor is yours.

Hello, everyone. Thanks for this opportunity to share some of my thoughts about the evolving landscape in the management of ovarian cancer. Next slide, please. So when we think about ovarian cancer, it's helpful to understand the patient journey, basically thinking about how patients move through and how we think about the goal of therapy at each setting. So generally speaking, when our patients present with advanced disease on diagnosis, we use a combination of chemotherapy and surgery either before or after each other, whether we use a neoadjuvant version or an adjuvant version. And depending on the patient's genetic findings, they may be eligible for maintenance therapy.

And our goal in this initial treatment strategy is to cure the majority of patients. Unfortunately, some of our patients will recur with disease. And depending on their histology or stage, at least in ovarian cancer, this is something that happens more often than we would like. And this period of remission is defined by the time of platinum sensitivity, which I'll talk about in a minute. But the longer it takes for the cancer to come back, the better. However, the bottom line is that for the majority of patients, they will have relapsed disease and understanding and developing therapies that are both efficacious and also preserve quality of life in this setting is what's most important.

Next slide, please. As I mentioned before, the way we delineate or distinguish recurrence or relapse is by platinum sensitivity. The current definition of this is the time from their last carboplatin or cisplatin exposure. If it's greater than six months, that's the PFI, the platinum-free interval, then they are thought to be platinum sensitive. If it's less than that, they're thought to be platinum-resistant. Eventually, most patients who develop platinum-sensitive disease will eventually develop platinum resistance. So you can imagine that a patient is treated about a year later, their disease recurs, that's platinum-sensitive. They get another platinum doublet combination. That may work for a time. If it works for greater than six months, then they're still platinum-sensitive.

But unfortunately, over time, that platinum-sensitive free interval will decrease and all patients who are platinum-sensitive will become platinum-resistant. And even though this timing seems particularly arbitrary as it is timed to the carboplatin exposure, it does have biological significance. We know that patients who are in the platinum-sensitive stage of their disease or platinum-sensitive state of the disease have more options. Their survival is longer. Patients who have platinum-resistant disease now have multiply treated cancers that are continuously evolving. And in those particular settings, the mortality is higher. But again, it's on a continuum, if you will.

We have more options for platinum sensitivity or platinum-sensitive disease. There are few options for platinum-resistant disease. But for a long time in ovarian cancer, there has been a lot of work into understanding why there is a dichotomy in patient outcomes for these and what strategies we can use to extend platinum sensitivity, both biologically and also clinically. And many of our efforts so far have not been successful in the past. We've used chemotherapy, pazopanib, all sorts of things, but we have not been successful, constantly highlighting that this is a goal.

There is one medication that we often use in this platinum-sensitive setting, and that's bevacizumab. And the goal of bevacizumab is when we combine it with the platinum doublet, we can then continue it as maintenance, providing an option for patients that's reasonably tolerated, even though it has some side effects that I will talk about later. But in some patients, it extends that platinum-sensitive window so they did not become platinum-resistant and have to deal with all of the sequelae that comes with that.

Next slide, please. So one of the key goals of patients who have recurred from a solid tumor perspective is that we know that these patients are no longer curable. So treatment selection, including maintenance options, becomes a balance of both toxicity and efficacy. We want to control the disease. We want to prevent progression of the disease. At the same time, we want to make sure patients preserve their quality of life. I just mentioned the role of bevacizumab being one of the agents that we use in this platinum-sensitive and even in the platinum-resistant setting. But there are many patients who are ineligible for bevacizumab.

So for example, some patients may have received it with their upfront treatment and their cancers progressed while they were on it. Many patients have absolute contraindications. They might have fistulas or they might have high blood pressure. There's a long list of medical reasons why bevacizumab might not be suitable for a patient. In addition, there are quality of life side effects that we see with this drug such that some patients elect to not take bevacizumab and undergo observation. Again, with the understanding that their cancer will progress at a certain time, but they value their quality of life in the current moment. So they go that route.

So for patients who are eligible for bevacizumab, it's something we discuss and we offer as maintenance, we give it with the chemo and then following the chemotherapy. But for some patients, observation is the way to go because, again, it preserves that quality of life. And the way we follow these patients is with periodic or regular clinical exams. We do scans about every three months or so, but it varies on clinical practice, again, because we know the inevitability of disease progression once patients have suffered a relapse.

Next slide, please. So to fill this space, knowing that bevacizumab might not be appropriate for some patients and knowing that even patient preference might skew you away from bevacizumab and knowing that the idea of observation is something that patients might need to default to, we've continued to do research into trying to understand what other therapeutic modalities we can incorporate into this platinum-sensitive space to prolong, again, that platinum-free interval or decrease the time it takes for the disease to progress. And one drug category that's moving into this space are antibody-drug conjugates, either as monotherapy maintenance or in combination with bevacizumab.

Now, there is no doubt that antibody-drug conjugates are effective. We have lots of efficacy data in the platinum-resistant setting. This is where these drugs are. And they really reshaped the way we think about chemotherapy and therapeutic sequencing in this space. So that same logic that if they work in the platinum-resistant setting, we need to move them up to the platinum-sensitive setting, either with chemo and/or as maintenance is something that is actively being pursued in clinical trials. The results are too early to tell. We've had at least one study that showed no real difference in terms of progression-free survival, but this was using a different type of ADC. There are newer ones that are being tested, and those results will be seen soon.

The one thing we do know is that adding an antibody drug conjugate, either with bevacizumab or by itself increases side effects. That is the one guarantee we have. And we know this because these drugs have a well-defined toxicity profile that we've learned from the platinum-resistant setting. So for patients who are still trying to avoid toxic option as maintenance, they're still going to have the same problem. So we will buy some efficacy because ADCs work similarly to chemotherapy, but that will come at a cost of increased adverse events.

Another issue is that the vast majority of our ADCs have a topoisomerase-1 payload. And currently, there's a debate in the field that because most of the ADCs we have are topoisomerase-1 inhibitors, if we use them in the platinum-sensitive setting or as maintenance and the patients progress or relapse, now they are coming in with increased toxicity burdens, maybe platinum-resistant disease. And the best treatment we had, which are the topo-1 ADCs have now been expanded as maintenance, which some people feel is perhaps not the best way to deploy or utilize this medication. Now that said, there's research going on, on both sides of the fence to better understand this. So there are pros and cons or certainly some notable reasons for why antibody-drug conjugates may not be the appropriate choice for all patients as a maintenance treatment, especially in the platinum-sensitive setting.

Next slide, please. So before we go into why COM701 might be an option in this setting, it is also worth mentioning that some of the other maintenance treatments that we've tried in the past, including PARP inhibitors have since been withdrawn for this use by the FDA in the second-line setting. But in case some people were not familiar with that, in the past, we used to do or give PARP inhibitors as maintenance with or without bevacizumab, most of the time without. And this is because these drugs have their own toxicities. And what we found was that even though initially, they appear to be helping, when we looked at overall survival and we looked at the burdens of toxicity from these drugs, they ended up becoming detrimental in some cases and ineffectual in many cases, and that has been withdrawn.

So currently, anti-angiogenics like bevacizumab or what we have, antibody-drug conjugates have pros and cons with the most recent data we have showing that they're not efficacious as maintenance therapy. We need new options. One of the advantages of COM701 as an immune checkpoint inhibitor is its tolerability profile. So we have some data in the platinum-resistant setting about its preliminary efficacy, and there will hopefully be a public report or manuscript to this effect soon. But one of the highlights of this drug was that we found that there was a subset of patients who, based on the mechanism of action of this drug, we're having responses and durable responses without toxicity. And as I mentioned before, the absence of toxicity is a highly, highly prized asset in patients who are looking for maintenance therapy because they've had chemo. So the chemotherapy has debulked the disease and what they're looking for is something that can extend that benefit that is also compatible with their quality of life.

So a key advantage of COM701 is that we know very well that it's very well tolerated, and we know that it can provide a long progression-free survival in select patients. And these characteristics make it a suitable candidate to be evaluated as maintenance therapy. And because we're using it earlier on in the patient's journey before they become platinum-resistant, all of our data currently with COM701 is in the platinum-resistant setting. We know that these patients are fitter both clinically and also immunologically. There's a lot of data in ovarian cancer and other solid tumors that the more chemotherapy we expose to our patients, that physical frailty that we see in the clinic and that decline is also an immunologic reflection of this as well. So for several reasons and the ones I highlighted, looking at an immune checkpoint inhibitor like COM701 in the platinum-sensitive setting becomes something worth pursuing.

Next slide, please. So the question then becomes to what end? Immune checkpoint inhibitors have been studied extensively for GYN cancers with mixed benefits. And it wasn't until recently that we got our first approval in the platinum-resistant setting with chemotherapy for platinum-resistant ovarian cancer. The key goal in the second- or third-line platinum-sensitive recurrent setting is to maintain that platinum sensitivity. Because if we have a drug or a therapeutic that's reasonably tolerated, low side effect, is not chemo-like, like an ADC, and extends the patient's platinum-free interval, then we've kept them in that box that I showed you a few slides ago in that platinum-sensitive setting. This allows them in the future, if they progress to be rechallenged with a platinum agent, a platinum doublet, and these patients can stay in that physiologic or biological or clinically defined box.

So providing a drug or a treatment that has low toxicity is important, providing a drug that maintains platinum sensitivity is important. And inevitably, the longer a patient stays in the platinum-sensitive box, the longer it takes them to progress or evolve into the platinum-resistant box, the more we can perhaps benefit our patients in the long run.

So to briefly recap, the goal of COM701 studied as a maintenance after platinum doublet therapy in patients with second- or third-line relapsed platinum-sensitive ovarian cancer is to provide a clinically meaningful, well-tolerated option for patients who might not be eligible or intolerant to bevacizumab, which is the current treatment with the hopes that this agent will prolong their platinum-free interval, keep them in that platinum-sensitive state and also provide them adequate quality of life at a time when their immune system can be meaningfully engaged.

Thank you, Dr. Yeku. As a reminder, Dr. Yeku will join company management for the Q&A portion of our call. At this time, I'd like to introduce Compugen's Senior Vice President of Research and Discovery, Dr. Sharon Kredo-Russo. Please go ahead.

Good morning, everyone. Thank you for joining us today. My name is Dr. Sharon Kredo-Russo, and I'm the Senior Vice President of Research and Discovery at Compugen. I'm excited to share today an important capability we have built based on our established discovery engine. We believe this capability has the potential to bring value to areas with high unmet needs.

As discussed earlier today, over the years, Compugen's discovery platform has generated multiple assets. Some programs are in early stages of discovery, some are in clinical developments, and several are being advanced through strategic partnerships. But what is the engine behind all these discoveries?

Behind all these discoveries stands Unigen, our AI-powered discovery engine continuously connecting computation, biological validation and clinical learning. At the center of Compugen approach is our proprietary knowledge base. This is a computational engine, which is actually a collection of specialized AI-powered computational approaches and algorithms that are designed to study very complex biological systems. These algorithms prioritize hypothesis based on their therapeutic potential and select the right tools that are then applied across multiple data types. The discovery engine's output of potential novel immune oncology targets that are experimentally validated, resulting in the therapeutic drug candidates that have over the years entered the clinic. Clinical trial data then generates an additional layer of information that feeds back into our database and supports all our future and ongoing discovery efforts. It is this established engine that we are now expanding into a new opportunity space called the Tumor-Associated Antigens.

Tumor-Associated Antigens provide a tumor address to the target therapies, directing the drug activity or delivery specifically to cancer cells while sparing the healthy ones. A bispecific antibody is engineered with one binding arm to the tumor antigen, while the other arm brings the modality. TAA plays a central role in many of today's promising cancer therapy modalities. For T-cell engager, the TAA brings the T-cells into direct contact with the tumor cells. For other targeted immune activation approaches, it may mean the tumor selective activation of immune pathway. And beyond immune engagers for other targeted modalities such as ADCs, the TAA enables the local delivery of the therapeutic drug directly to the tumor cell sites. And across all these modalities, the target determines where the therapy goes and where the biological activity happens. And as a result, the TAA quality can directly influence both efficacy and safety. And that is exactly where one of the larger unmet needs lies in the field today, the need for novel and selective TAAs.

Focusing now on T-cell engagers, which have emerged as one of the most exciting modalities in oncology. A bispecific antibody is used to bring the T-cells in close proximity in space to the tumor cells, resulting in a direct killing. This powerful modality has attracted substantial industry investments and growing market opportunities in recent years. However, the power of this high immune activation modality also highlights a fundamental challenge. A T-cell engager can only be as selective as the antigen directing it. And only few TAAs have the right profile of tumor specificity, lower healthy tissue exposure, and high tumor coverage.

As of today, 12 T-cell engager drugs have been approved, validating the therapeutic modality. And yet only two in solid tumors, where target quality is still a critical constraint. The figure below shows how we look at target profile signal across healthy tissues, where red means high level in the healthy tissue and green means low level. One example of a clean target is DLL3. The approval of tarlatamab in small cell lung cancer targeting this antigen demonstrates how a favorable clean target profile can help to unlock the potential of the therapeutic modality. You can see how DLL3 is exceptionally clean with no red tissues at all. Meaning it is mostly absent from the normal healthy tissues and expressed almost only on the tumor cells.

On the other hand, mesothelin is an example of a different type of target. You can see here that the target is quite clean beside the red. It means high expression level in the healthy lung cells. As a result, the development encountered dose-limiting toxicity of on-target off-tumor, meaning activity outside of the tumor in the healthy lung that indeed associated with this normal high lung expression of the TAA. And actually, many targets display extensive normal tissue expression, as shown here, while others display different restricted pattern and the TAA selection for the right indication has became a challenging task. As we believe that the powerful therapy is only as good as its target, we have expanded our established capabilities in target discoveries to build UnigenX, a dedicated engine for TAA discovery.

We built upon our proven foundation of Compugen's knowledge base and combining biology, computational, and clinical validation. We start with the knowledge base and the upfront definition of the target's desired profile. Then specialized AI-focused discovery tools integrate multiple layers of biological data simultaneously, spanning different omic types. Moving to the output, the platform first generates explainable target scoring. This prediction then undergo extensive TAA-specific experimental validation, including immunohistochemistry, normal tissue exposure, functional toxicity assessment, resulting in the drug candidates. And ultimately, the output delivers differentiated opportunity, novel TAA, novel drug candidates, and patient selection matching the targeted therapies with the right patient population based on biomarker target expression. But how do we find the right TAAs?

As said, finding the right TAAs is a major challenge. Our approach is differentiated in several ways. First, it is a multidimensional approach. A target is evaluated broadly and deeply using genomic data, bulk RNA sequencing, single cell transcriptomics, proteomics, and spatial biology and is assessed based on multiple parameters together. This is a versatile discovery platform. We match the defined TAA criteria to the desired specific modalities and the analytical approach to the biological questions, which opens opportunity across targeted modalities. And this deep biological understanding and sustainability allow us to identify opportunities that may remain invisible when viewed through a single data set, enabling us to uncover novel targets with the aim of translating them into a clinical stage therapeutics. So how is our multidimensional multi-omics approach allowing us to achieve this specificity and understanding in novel discoveries?

First, we start with search broadly. This means giving our algorithm access to a very rich biological landscape. The platform integrates information for more than 100 million single cells across multiple cancer subtypes, over thousands of tumor and normal proteomic samples and a proprietary normal cell atlases composed of multiple tissues and over 100 healthy cell types. We believe that the data scale, however, is not enough and understanding deeply means creating a multiparameter target profile for every target candidate. We assess tumor specificity, patient coverage, normal tissue exposure, heterogeneity and then calculate the potential therapeutic index. That integrated assessment is a defining element of UnigenX used for an informed TAA selection.

We integrate these multiple biological layers into explainable target scores. The graph shows two major scores the platform produces, predicted efficacy or activity on the Y-axis and predicted safety or toxicity on the X-axis, with each dot representing a potential TAA, in this case, in colorectal cancer. Zooming in now on a specific target A that we have discovered with a very high scores in efficacy and safety. The figure on the right illustrates the multiple parameters contributing to the total score, divided into different features that originate from the different omics data types, allowing us a deep understanding of the biological profile of each potential target relative to benchmark TAAs. This allow us to adapt each scoring system to the desired modality and its requirement. And importantly, it enables stronger hypothesis and the ability to identify targets with potentially better profile than benchmark. But deep understanding is not only the scoring system as we perform deep validation on every feature.

And here is how the data look like. Evaluating multiple parameters such as RNA expression profile, the cellular resolution of the different cell types and the protein membrane level in both normal and tumor tissues. And importantly, we experimentally validate each target localization through immunohistochemistry to identify tumor specificity across patient cohorts. Individually, we believe looking at each layer separately gives only partial information. But together, we believe that we build the confidence in the target differentiated biological profile. And indeed, we have several target candidates that have been prioritized for further experimental validation in our early pipeline.

But what if we could extend the TAA discovery space beyond a single TAA. We believe the target pairs can expand the TAA discovery opportunity. It is important because some therapeutic challenges may be addressed more effectively through paired targeted approach.

In an AND-gated strategy, two targets, A and B, must be present together on the same cancer cell. The goal is to improve cell activity and sparing the healthy tissue. Therefore, the T-cells will only kill in the presence of both targets on the tumor cell and will not kill the healthy cells expressing only one of the targets. In an OR-gated strategy, either target C or target D can support killing activity of the tumor cells. The goal here is to broader tumor cell coverage within the patient and overcoming potential resistance mechanisms of downregulating or mutating a single TAA.

On the graphs below, light blue represent the tumor cells, while dark blue represents the healthy cells. The top two rows show the expression of each individual target, target A on the top and target B in the middle, while the bottom row shows their combinations. As shown, upon dual targeting of target A and target B in an AND-gated approach, tumor cell activity is improved, sparing the healthy tissue.

The example on the right represents an OR-gated approach from prostate cancer, demonstrated how combining target C on the top, target D in the middle and either target C or target D in the bottom improves the coverage of tumor cells within the patients. We believe these approaches could significantly expand the discovery space from individual targets to rational target combinations that may create opportunities that cannot be addressed by conventional single target discovery approaches.

To summarize what we have discussed today, we started with Compugen's validated discovery foundation, Unigen, with a track record of identifying novel immune oncology targets as discussed earlier today and translating them into the clinic. On the right is UnigenX expansion of this foundation into new and broadly valuable space that include novel TAA discoveries to create differentiated opportunities for multiple targeted modalities such as T-cell engagers.

But ultimately, our goal is to make a meaningful difference for cancer patients. First, we believe that a scalable discovery engine has the potential to generate multiple TAA-driven opportunities across solid tumor indications and modalities. We believe these discoveries could support future potential partnerships. And finally, discoveries from our platform can potentially be combined with our current ongoing early programs.

To conclude, we have created an expanded target discovery engine across IO field to support the next generation of targeted therapies with the aim of achieving clinical meaningful therapeutic effect for unmet cancer indications. Thank you very much for your attention.

With that, I'll hand back the call over to our operator, who will take us through the Q&A portion of our event.

Great. Thank you, Sharon. [Operator Instructions]

We're going to begin with two questions we received over the webcast. So the first one is, what is your platform's differentiation versus competitors? Other companies looking for new TAAs?

Sharon, do you want to take this one?

Sure. Thank you, Tara. As discussed today, we are taking a broad multidimensional and multiomics-based approach, together with a very deep biological rationale understanding. We are not looking into a single layer of data set, but we are looking on multiparameter, multidimensional understanding of the tumor and the biology. Another aspect is that we built upon Unigen foundation in the IO field and the vast assets that we have investigated and digested over the years at Compugen.

Great. Thank you, Sharon. The next question, what would be your first modality to focus on and why?

As mentioned earlier, we can use the TAA discoveries for any desired targeted modalities, for example, antibody drug conjugate that brings the drug. But our first focus is on T-cell engager and targeted immune activation, as we believe in the power of immune activity that can be directed specifically to cancer cells, assuming we have the right TAA to select or the combination.

So our next question will come from Stephen Willey at Stifel.

Yes. Can you guys hear me okay?

Yes, we can.

Perfect. Maybe a question for Dr. Yeku. I know that we've seen some attempt at leveraging immunotherapy within platinum-sensitive maintenance. I think with some differing results. How do you think the mechanism of COM701 has the potential to change the narrative around the use of maintenance immunotherapy?

And then maybe a question for the company. Just curious how you plan to leverage this discovery effort into TAAs? And is this something that you would look to build out your own internal pipeline? Or do you see this more as an opportunity for various collaborative partnerships?

Yes, for sure. I can tackle the first part of that question. I think a lot of our efforts so far with immune checkpoint inhibitors as maintenance therapy for ovarian cancer, much of that stems from our understanding of the role of PD-L1 as a biomarker, and this was way back when. I think now it's better understood that this marker is not the most reliable one that we have because it's very dynamic. And many of these trials involve a combination of immune checkpoint plus chemotherapy, followed by immunotherapy or immune checkpoint inhibitors by themselves. Again, our understanding in terms of the effects of chemotherapy or concurrent chemotherapy with steroids and everything else, and the potential kind of opposite effect that, that might have leading into maintenance, I think is becoming better understood. There are a lot of trials now that are separating both of those things.

So I think the prior studies, one, because we weren't using an appropriate biomarker; two, because there were chemo plus IO followed by IO, which is a different strategy here. Those are the two first reasons. The third and perhaps most important reason is that COM701 works very differently than a PD-1 checkpoint inhibitor of various, PD-1 or PD-L1. The PVRIG pathway has been shown to be specifically important in ovarian cancer just in terms of the way it works via dendritic cells and further down the line, cytotoxic T-cells, but I think a more disease-specific mechanism exists for COM701 that doesn't necessarily or hasn't necessarily existed for immune checkpoint inhibitors targeting the PD-1, PD-L1 axis.

And because we've separated out the cytotoxic debulking phase of our plan from the maintenance phase, again, all of the issues that will be concerned about regarding the myelosuppression or immunosuppression we get from chemotherapy and steroids, that's already behind us without necessarily jeopardizing the debulking effect that we're getting with chemotherapy. So I think for those reasons, at least for me, it's reasonable to expect a different result than some of the other studies that we've done in the past.

Thank you, Dr. Yeku. I will take the second question about the utilization of the platform, either for internal pipeline or collaboration. I think that eventually, the way we always handle it, it was right for our older assets, some of them are in our pipeline, some of them are being partnered. And it's the same for Unigen and UnigenX partners. It really depends on what is right for the program. If we think that for the program, we can move it ourselves and bring value and take it to cancer patients in the right settings and bring it in the right space and eventually bring value to our shareholders, we'll take it ourselves. If we think that there is a good reason why to partner in different programs, it could be in different stages, then we could also partner. So I think overall, I would say that what you saw from us in the past is also the plan for our next asset. Sometimes we can partner, some we can move it ourselves.

Our next question comes from Bill Ling at Leerink.

Bill on for Dana. Just one for me. I'm just curious, what's the path forward for COM701 now that there are several ADC trials ongoing in the maintenance setting? Do you have to run a Phase III against those ADCs?

So maybe I can start, or maybe Michelle can start and then Dr. Yeku can give his perspective. Michelle, do you want to take it?

Sure, happy to. Hi, thanks. So firstly, in order to set up a pivotal trial, we would need to have the ADC already become part of standard of care to be able to compare. So while the ADCs are all being studied in earlier lines of treatment as well as in platinum-sensitive maintenance, they've yet to actually be approved in that setting. So at this point in time, for the patient population that we're currently studying COM701 in, the most appropriate standard of care would be bevacizumab, and that is one of the options that we would consider as part of a path forward. I think one of the other advantages we have with COM701 is because it's well tolerated, it does allow us to combine it with other agents. Let me pause maybe Dr. Yeku would like to also put in his thoughts around this as well.

Correct. Everything Dr. Mahler said is correct. In fact, we were deliberate when we were designing this study to avoid that particular question. So these are drugs that are in different therapeutic classes, answering completely different questions. Inevitably, there will be patients who will need an ADC maintenance strategy as part of their disease characteristics, et cetera. But we're looking for a population of patients who cannot tolerate the side effect burden of ADCs or who might be better served by receiving their ADCs in the subsequent line, either as a platinum-resistant therapy or someplace else.

Now the reason looking at ADCs as a maintenance strategy and how combination trials in the future might be very challenging is because by the time the data that we're expecting for ADC maintenance comes out, everyone would have been looking at antibody-drug conjugates in first-line maintenance, so newly diagnosed first-line maintenance. And then all of the data that people are looking at for platinum-sensitive maintenance becomes irrelevant. The data that we get from this study is above that rate because we're looking at a different population, answering a different question. So whether or not two or three years from now, the standard of care is to get an ADC in first-line newly diagnosed disease, it doesn't matter as long as you're platinum-sensitive and you received bevacizumab. Looking at COM701 as a maintenance treatment, if the results are positive, will be an option for you.

If ADCs are approved as maintenance in the platinum-sensitive setting, a patient will have to sit with their doctor and decide, do you want to go through all the toxicity, get the benefits and et cetera with an ADC? Or if you've had it in the first line or someplace else, do you want to use a different strategy such as COM701? So looking at this from multiple different angles, I don't see a scenario where we will be forced or required to randomize against an antibody drug conjugate arm as a registrational strategy.

Great. So our next question comes from Leland Gershell at Oppenheimer.

Just a question for Dr. Yeku. Just wondering, you mentioned bevacizumab being an option, yet there are many patients who are either poorly tolerated or ineligible. Could you just tell us in your experience of those patients who respond to platinum, so therefore, would be eligible in the maintenance setting, what fraction do you think that 701 would really be a preferred alternative to bev? Looking at those for whom bev would not be a good option? I have a follow-up.

Absolutely. So I would say, at least in our practice, it's as high as 40%. If you look at other non-U.S. sites where the use of bevacizumab is more strictly regulated either to a platinum-resistant setting or first-line setting or even certain parts of the United States, like the Midwest, some of the sites there actually reserve bevacizumab for either Stage IV upfront when patients have ascites and pleural effusions, when they think it might be helpful or in the platinum-resistant setting. In those particular settings, the proportion of patients who either because they cannot get that or require something different is much higher than the 40%, and I would say as much as 60%.

Also recall that some of our patients, at least the way we designed this study, they may have been previously platinum sensitive. So they may have even received bevacizumab in a previous platinum-sensitive setting. And as long as they're platinum-sensitive again, for a lot of those folks, the cumulative toxicity [indiscernible], or their physician has to decide, do I continue to bear with bevacizumab, or do I undergo -- I would say anywhere from 40% to 60% of patients would be eligible for this kind of study, just looking at the population at risk.

Okay. The connection was a bit choppy, but I think we've got the key message. And then just a follow-up, the ADCs sort of also being studied in the maintenance setting, it sounds like there's also some drawbacks there in terms of tox. So do you see kind of an efficacy bar in which the ADCs could be preferred no matter what? Or do you think that because of the safety advantages, tolerability advantages of 701, that would still be an option that would have uptake kind of irrespective of what ADCs are able to show. How does that calculus kind of play out in your mind?

So ADCs are like chemotherapy. So there are going to be some patients that even after six rounds of chemotherapy, they're still going to have a high burden of disease. They might have liver metastases, a population we have excluded in our study. Some of these patients might be on the verge of progression and you want to give them more chemo. So we think about ADCs as giving more chemo with slightly less side effects than conventional cytotoxics. Those are the patients we'll give -- that will be continually eligible to get more chemo and doctors will continue to give them ADCs.

I think for the patient who's had a good response, so you've had your six cycles, you have no evidence of disease left, maybe you've even undergone a secondary cytoreduction. you have no evidence of disease. For those patients, an ADC, which is like more chemo is overkill, right? You need tumor cells to be dividing to uptake the ADCs, and that's why the risks are justifiable. For those patients, a maintenance option that's very well tolerated when you have no evidence of disease, I think clinically is going to be the choice. So I envision a clinical scenario where if both are available and both have efficacy data, it's going to come down to individual patient characteristics, and both can exist together. They just serve two different and distinct patient populations.

Our next question comes from Chad Messer at Lake Street.

Great. Thanks for putting together this presentation today, two great topics. Just on UnigenX, the data you showed on those targets A through D, very impressive. Appreciate you took the time to do all that. You got four validated candidates with qualified expression profiles, colorectal, prostate cancer. Is it possible to give us a sense of how many candidates have kind of gone through this process so far to date? And maybe just an idea of what the next milestone we should look for from you guys is on the platform?

So as we said and Sharon said along the call that there are a few that passed the computational part into validation. I mean, what we have shown, these are not necessarily yet validated. For us to really say that an asset is validated, there is a long way of validation and going through animal studies and ex vivo studies, et cetera, and we put us at a very high bar before we move it further. So yes, quite a few targets have passed through a few stages maybe or definitely from the computational platform and out. And but there are still stages to go. First-in-class assets are always complicated. And definitely, when an asset will be derisked enough for us to communicate and to take commitments also with our investor community, then we'll disclose exact time lines and number of assets moving forward. In this stage, there are a few assets moving in validation, yes.

So we have another question that came over the webcast from Charles Wallace at H.C. Wainwright. For Dr. Yeku, what size of PFS gain would you change your practice here in PSOC? And does keeping a patient platinum-sensitive matter beyond the PFS number itself?

Yes. I'll answer this question in reverse order. So yes, keeping a patient in a platinum-sensitive setting is very important. I think, as I mentioned before during the prerecorded or the previous discussion section, that serves two purposes. We think and we'll refine our understanding, but patients who are platinum-sensitive have a different biology of disease than those who are platinum-resistant. We know this because many of our trials that we've gone from moving from platinum-resistant to platinum-sensitive, response rates goes up, PFS goes up, everything gets better when you move upstream. So we know there's something about the biology, something about the host, something about the resistance mechanism that makes these two different kind of buckets. Of course, molecular testing will refine that in the near future.

But for now, it's a very good rough surrogate that anybody can wrap their mind around that this group, platinum-sensitive, it's upstream or platinum-resistant. They tend to do better with almost any therapeutic as you go upwards. So there is intrinsic value in keeping patients there. From a practical point of view, patients who are platinum-sensitive regardless, I mean, you don't want it to be 5 days and like sharp midnight and it's exactly six months, but patients who are really are truly platinum-sensitive, they're eligible to receive Carbo/Taxol again or Carbo/Doxil, a different platinum combination. They can then go on to another platinum doublet with an ADC, they can do lots of things that a patient who's now transitioned into the platinum-resistant setting cannot do. If you look at most guidelines all over the world, international and the U.S., the recommendation for those folks out of the clinical trial is single-agent chemotherapy, whereas in the platinum-sensitive portion, we're still encouraged to use platinum doublets, if you can, because people feel there's value with that kind of dose intensification. So yes, there is intrinsic value to the physician and to the patient to remaining in the platinum-sensitive setting.

Now to answer the first question, which is, is there a number that I would like to see. I don't have any specific one in mind. I think in oncology, we have certain numbers that are multiples of each other that make us feel comfortable. It's 3, 6, 12, 18. I don't think they mean all that much. Of course, the longer the PFS is relative to the observation, I would like that. But if you're asking for, "Oh, would I feel more comfortable if COM701 maintenance gave me a 12-month PFS versus a 7-month PFS," that's a much tougher question to answer because I want to see, first of all, are they remaining in the platinum-sensitive setting? If yes, then for how much longer. And I think all of us who've done Kaplan-Meiers or looked at these distributions before, they're going to be on both sides. There's going to be some who are going to be on for a very long time, and we know this from our platinum-resistant COM701 data, some of our patients end up doing very well for surprisingly long periods of time.

And of course, there are going to be some that are just going to be at the margin. But I don't have any magical number in mind, but I want a comfortable buffer between the observation cohort and the active treatment cohort and the maintenance cohort.

Thank you, Dr. Yeku. And then a follow-up to Charles' question. Beyond PFS, which endpoints matter most to you in terms of maintenance, time to next therapy, response to the next platinum, quality of life and how much side effect burden will patients accept?

Quality of life is not an accepted, better yet, an accepted endpoint for our studies. But I would argue that should be, especially for -- so again, we were very deliberate when we designed the study to decouple the cytoreductive chemotherapy that patients were getting from maintenance. And unfortunately, people don't really do that. Most trials combine both of them together, get both drugs or they're one unit. This one is separated for that reason because the goal of maintenance therapy, if you think about it, is to maximize the benefit from the preceding debulking chemotherapy and also to maximize quality of life. And what is the appropriate dial between both of those?

Because, of course, I can give you 20 cycles of chemotherapy and get a longer PFS. The side effect at some point, those burdens will become overbearing. And I think that will be a key distinction between drugs that work like COM701 or have a similar toxicity profile and anything else. Recall that in the past, we've looked at other drugs for maintenance for ovarian cancer. We've looked at weekly Taxol as maintenance as ridiculous as that sounds in retrospect, we tried it. The PFS was longer, but a lot of the patients had permanent crippling peripheral neuropathy. We've tried other small molecule inhibitors like pazopanib, which we thought maybe not so bad, but these patients had appetite issues, weight loss, they had rashes, things that made it not suitable as a maintenance treatment.

So even though I'm an oncologist and I like hardcore endpoints in a maintenance setting, quality of life is important if you can show that it is truly active and not a placebo. And again, that's why we decided to do the hard thing and study this as a placebo controlled trial. Now outside of that, someone who receives or has a break "with maintenance therapy" of 8, 12 months, people travel. People do things. Some patients go back to work. That's an important valuable thing before they go back on chemo, go back on Taxol. So time to next therapy is important. The response to next therapy is important. We know that if we give certain PARPs in the platinum-sensitive setting, people's PFS2 might be shortened. Their overall survival might be shortened. So by giving them certain kind of maintenance therapy, we can make problems that downstream for the patient.

So I want to see time to next therapy. I want to see PFS2 on the next therapy. And because we're doing this in the, we know they can just go back to another platinum doublet and they can expect them to do well. We will not expect the same thing with PARP inhibitors. We've seen that on the PICCOLO study, and we may not see that from ADCs, and we've seen that from some of the recent data presented at ASCO.

Great. Thank you, Dr. Yeku. So this concludes our Q&A session and the event today. We thank everyone for joining, and you may now disconnect.

Ladies and gentlemen, thank you for joining us today. Welcome to Compugen's Second Quarter 2026 Results Conference Call. [Operator Instructions] An audio webcast of this call is available in the Investors section of Compugen's website at www.cgen.com. As a reminder, today's call is being recorded.

I will now hand the call over to Lindsey Trickett, Head of Investor Relations and Corporate Communications to begin. Lindsey, please go ahead.

Thank you, operator. Good morning, and good afternoon, everyone, and welcome to Compugen's Second Quarter 2026 Financial Results Conference Call.

With us today are Dr. Eran Ophir, President and Chief Executive Officer; and David Silberman, Chief Financial Officer. Dr. Michelle Mahler, Chief Medical Officer, will join us for the Q&A portion of the call.

Before we begin, I'd like to remind you that during this call, the company may make projections or forward-looking statements regarding future events, business outlook, development efforts and their potential outcome. The company's discovery platform, anticipated progress and plans, results and timelines for our programs, including disclosure of clinical data, financial and accounting-related matters as well as statements regarding our cash position and cash runway.

We wish to caution you that such statements reflect only the company's current beliefs, expectations and assumptions and that actual results, performance or achievements of the company may differ materially. These statements are subject to known and unknown risks and uncertainties, and we refer you to our SEC filings for more details on these risks, including the company's most recent annual report on Form 20-F. The company undertakes no obligation to update projections and forward-looking statements in the future.

With that, I'll now turn the call over to Eran.

Thank you, Lindsey, and good morning, everyone. Q2 was a quarter of steady advancement, and I'm pleased with the progress we have made across every part of the company. Our science continues to advance in the clinic and our partnerships are advancing on strong footing.

Our MAIA-ovarian trial in platinum-sensitive ovarian cancer is progressing, is on track for the interim analysis by Q1 2027. We're encouraged to see AstraZeneca continue to build momentum behind Rilvegostomig by initiating a new Phase III trial in urothelial carcinoma and with new data at ASCO from the GEMINI study in hepatobiliary cancer and the investigator-initiated I-SPY trial in breast cancer.

Lastly, our collaboration with Gilead on GS-0321 continue to progress as planned. And underpinning all of this is the same disciplined, data-driven approach that has always defined Compugen.

Now, let me take each program one by one, starting with our wholly-owned program COM701, a potential first-in-class anti-PVRIG antibody. We continue to make good progress with MAIA-ovarian, our sponsored, placebo-controlled adaptive platform trial evaluating COM701 as maintenance monotherapy in patients with second and third line relapsed platinum-sensitive ovarian cancer in a setting with no approved maintenance treatment option and significant unmet needs. We anticipate the interim analysis with median progression-free survival data by the first quarter of 2027.

During this quarter, we are pleased to present a trial-in-progress poster on MAIA-ovarian at the ESMO Gynecological Cancers Congress in Copenhagen. The poster underscores the strong biological and clinical rationale for evaluating COM701 in this population, including the differentiated biology of the PVRIG pathway versus other checkpoints like PD-1 and TIGIT, its high expression in ovarian cancer and the durable responses previously observed with COM701 in mono and combination therapy in heavily pretreated platinum-resistant patients.

As we prepare for the MAIA-ovarian interim analysis, we have been keeping close tabs on the emerging external data to keep our own expectations anchored in the current clinical context. Two recent clinical trials in relapsed platinum-sensitive ovarian cancer that included patients who have been pretreated with PARP inhibitors or bevacizumab or both have shown median progression-free survival for the control arm of less than 3 months. While the patient population of these 2 trials is not identical and more heavily pretreated than the MAIA trial population, based on these data, we estimate the median PFS of the placebo control group in our trial to be approximately 4 months.

As a reminder, patients with ovarian cancer are divided into either platinum-sensitive or platinum-resistant categories with the difference being the duration of their platinum-free interval. If a patient relapses in less than 6 months following platinum-based chemotherapy, they move into platinum-resistant category, where further platinum therapy is generally no longer considered effective and treatment shifts to non-platinum options.

For the interim analysis, we define clinically meaningful success as COM701 helping patients remain progression-free for at least 6 months after platinum-based chemotherapy. Achieving these thresholds maintains patients as platinum-sensitive for longer, delays their transition to platinum-resistant disease and give patients a valuable recovery break from the intensity of chemotherapy, thereby improving quality of life, while preserving additional treatment options and potentially changing their disease course.

Overall, we believe COM701's antitumor activity will be best assessed by the totality of the data comparing the treatment effects against our blinded, randomized control arm. MAIA is an exploratory trial designed to evaluate COM701 monotherapy and the magnitude of its effect. It is not a registrational trial powered to demonstrate a statistical difference between the treatment groups. Nevertheless, we believe that comparing COM701 as a monotherapy against a placebo control will allow us to draw clear conclusions about its clinical activity.

Looking ahead, we believe that clear prolongation of PFS in these patients could inform a registration path for COM701 and establish it as a potential backbone for drug combinations in this population, while also enabling a potential broader clinical development plan across earlier and later lines of ovarian cancer treatment as well as in other indications where clinical signals were previously seen for COM701.

Turning to rilvegostomig, the PD-1-TIGIT bispecific antibody being advanced by our partner, AstraZeneca. The TIGIT component of which is derived from our fully-owned COM902 program. In the last week, AZ has added a 12th Phase III trial to the overall rilve program in participants with high-risk muscle-invasive urothelial carcinoma. In this trial, rilvegostomig will be combined with Datroway, their approved TROP2 ADC, and tested in adjuvant settings against standard of care.

This new Phase III trial, TROPION-Urothelial04 follows the Phase II TROPION-PanTumor 03 study in which rilve plus datro combo showed an encouraging efficacy and a manageable safety profile in metastatic urothelial carcinoma.

We're also encouraged by the addition of rilve data AstraZeneca presented at 2026 ASCO Annual Meeting, which we believe continues to support the differentiated profile of this bispecific and its potential as an immuno-oncology backbone across multiple tumor types.

In advanced biliary tract cancer, AstraZeneca presented an updated analysis from the GEMINI-hepatobiliary study of rilve in combination with chemotherapy in the first-line setting. This was the first overall survival data result from rilvi. And as AstraZeneca highlights in their ASCO investor call, the 16.8 months of overall survival was a clear example of prolonged stabilization of responses seen with rilve across clinical trials, and the profile continues to support rilve combination potential.

In comparison, historical trial for first-line BTC showed overall survival duration of less than 13 months. The data showed encouraging efficacy together with manageable safety profile, both of which we view as promising signals in the settings of high unmet needs while recognizing that longer follow-up and randomized data from the ongoing Phase III trial in this setting will ultimately be needed to validate this finding.

As AstraZeneca continues to advance with rilvegostomig across its broad late-stage program, we believe this sustained investment reflects ongoing confidence in rilvegostomig. As a reminder, AZ has previously guided that rilve has a non-risk-adjusted peak year revenue potential of over $5 billion and will remain eligible for future milestones of $195 million and up to mid-single-digit tiered royalties tied to rilvegostomig progress and success.

Moving to GS-0321, formerly known COM503, our potential first-in-class anti-IL-18 binding protein antibody licensed to Gilead. GS-0321 represents a novel antibody approach to harness cytokine biology for the treatment of cancer, potentially overcoming the limitations of direct cytokine administration. The ongoing Phase I dose escalation trial continues to progress as planned. As a reminder, we have received $90 million so far from Gilead on this asset, and we are eligible to receive up to $758 million in additional milestones payment plus single-digit to low-double-digit tiered royalties.

Now, moving to our early pipeline fueled by Unigen, our AI/machine learning powered computational discovery platform, which has been developed and refined for more than a decade to identify novel drug targets and biological pathways grounded in human disease biology. As we have said before, our focus is not on using AI to optimize non-biology, but on uncovering innovative opportunities to activate the immune system against cancer. Unigen has already discovered the targets of COM701, COM902 and GS-0321, and we remain committed to identifying and advancing the next generation of immuno-oncology innovation.

With that, I will turn the call over to David to review the financials.

Thanks, Eran, and thank you all for joining us today. We finished the first half of 2026 with a solid balance sheet and financial flexibility. Cash runway, assuming no further cash inflows, is expected to fund our operating plans into 2029. We anticipate using this runway to continue advancing our COM701 platinum-sensitive ovarian cancer trial, MAIA-ovarian, and to support the progression of GS-0321 in the clinic together with continued investment in our early-stage pipeline.

Going into the details, I will start with our cash balance. As of June 30, 2026, we had approximately $125.3 million in cash, cash equivalents, short-term bank deposits and investment in marketable securities. Revenues for the second quarter of 2026 were approximately $2.6 million compared to approximately $1.3 million of revenue for the comparable period in 2025. The revenues in the second quarters of 2026 and 2025 reflect the recognition of portions of both the upfront payment and the IND milestone payment from the license agreement with Gilead.

Expenses for the second quarter of 2026 were in line with our plans. R&D expenses for the second quarter of 2026 were approximately $6.3 million compared to approximately $5.6 million in the second quarter of 2025. Our G&A expenses were approximately $2.3 million for the second quarter of 2026 compared to $2.2 million for the second quarter of 2025.

For the second quarter of 2026, our net loss was approximately $7 million or $0.07 per basic and diluted share compared to a net loss of approximately $7.3 million or $0.08 per basic and diluted share in the second quarter of 2025.

With that, I will hand over to the operator to open the call for questions.

[Operator Instructions] The first question is from Stephen Willey of Stifel.

I was just curious, so it sounds like you've taken down your control arm assumption in the MAIA trial by maybe about 1.5 months. What do you know about the patient population from these 2 trials that you cited with respect to things like liver metastasis status, and I guess, just general patient eligibility criteria. And I would just be curious to get a better understanding as to your level of confidence now around this revised 4-month number.

Thank you. Michelle, do you want to take this?

Yes, I'm happy to take this. So the 2 trials that we are referring to are European studies. One is TEDOVA recently presented at ASCO. And the other trial is a trial called OReO. Both trials are run in Europe and had similar patient populations because they enrolled patients with platinum-sensitive ovarian cancer and were treated in the maintenance setting. However, the patient population was not identical because the trials did not cap the prior lines of treatment. They included patients that had stable disease as well, which we don't. They also included patients who have liver metastases. And so they also could have had multiple attempts of being treated with both bevacizumab or PARP inhibitors.

So due to this, these patients were actually more heavily pretreated than our MAIA-ovarian trial. And their placebo control arm had a median PFS of 2.8 months. We anticipate that the actual benchmark is somewhere in between the historical data sets, which we took from the original registration trials for the PARP inhibitors, which was approximately 5.5 months, and these new updated trials who have a similar patient population. And therefore, we've adjusted it to approximately 4 months. As such, we currently don't know who is allocated to which arm because our trial is blinded. So we're making these adjustments based on emerging data.

And maybe I could add that eventually, we change -- the approximation is roughly around 4 months. But I think eventually, what is more -- most important for this trial is that that's why we have an internal randomized controlled placebo arm, and eventually, we are comparing COM701 40 patients treated in monotherapy versus placebo arm of 20 patients, whatever antitumor activity we see in the treatment arm is COM701-driven. It's not a combination study. And the assumptions for the placebo control are important, but eventually the critical is the actual data on the trial comparing placebo to COM701 treatment.

Okay. And then maybe just quickly on GS-0321. I guess, you've been dose escalating now for, I guess, around 18 months or so. Have you had a conversation with Gilead about presenting some of the dose escalation data before you move into dose expansion? And is it safe to assume that you are now dose escalating both in combination with the PD-1 inhibitor, and I guess, monotherapy as well?

Yes. So typically, with this kind of arrangement with pharma companies, we cannot say much. I would just remind that as you -- this said, we have dose escalation in mono and in combination with PD-1. We also have backfill course in the monotherapy, meaning more patients in the higher doses, and then, the expansion phase. So we're looking at benchmark studies in this stage. Yes, I think it's reasonable to assume that everything is moving forward as planned. That means that probably we are already doing combinations and other expansions, but -- for sure, I mean, in the backfill course, I would say. But you cannot say precisely where we are and in which stage we'll disclose data. I think it's still early. I mean, if you -- even if you look at other benchmark studies, Phase I studies, 18 months into the study, it's a bit early for reporting data.

The next question is from Leland Gershell of Oppenheimer.

Just 2 questions for me. Just wondering with respect to the MAIA-ovarian trial and the guidance for the data. Just wondering, given the nature of the kind of changing assumptions and event-driven nature, could you see to the extent possible, a readout that might come before the end of the year? And also, I want to ask, are there any particular biomarkers that you'll be looking at alongside the clinical PFS outcome future studies?

Thanks, Leland. So for the first question, we are -- yes, the PFS of the placebo is now a bit shorter, but we are not changing our guidelines. And eventually, that's why we said the results will be by Q1 '27. It is depending on the actual data on the study. And obviously, we're going to report it when the data is mature enough.

Michelle, you want to add something for the second question about the biomarkers and other readouts to look at the study?

Sure. So we're -- our primary readout is progression-free survival. We don't yet -- we don't have a specific biomarker selection strategy other than patient characteristics where we have excluded patients with liver metastases. And the other thing to note is that in our earlier data, we did see activity in patients who were both PD-L1 positive and PD-L1 negative. So other than trying to enrich for more clinical attributes, we don't have a specific biomarker, and we do have an exploratory plan that we will analyze when we unwind the data.

The next question is from RK of H.C. Wainwright.

Eran and team, this is RK from H.C. Wainwright. One quick question. Have you had any interactions with the FDA to see if the MAIA-ovarian trial alone could support either an expedited or an accelerated path for approval, especially in this setting that we don't really have a drug approved?

Thanks, RK. Michelle?

Okay. Sure. So at this point in time, we have not had a meeting with the FDA. Once the trial reads out, we will follow all the appropriate regulatory steps. What I will say to you is we incorporated a lot of the guidelines from the FDA in designing the trial, and it's definitely in line with their guidance on Project FrontRunner, which is one of the reasons why we did go into an earlier line of treatment as well as using the Bayesian trial design, which is, again, part of the FDA guidelines that have recently come out. So we are confident that with robust data, we will be able to have good engagements with the FDA.

Is it possible for me to ask another question?

Sure.

Sure.

On the partnership with AstraZeneca, now that they have 12 clinical studies going on and 12 Phase III studies going on, do you have an idea of what we should expect in terms of the earliest Phase III readout that we could see? And also, does this inclusion of the new trial, does it change either the schedule or composition of the $95 million milestone outstanding?

I will start with the second question. This doesn't change. I mean, just it's another short-term goal in a new indication in combination with ADC, which is, again, very promising, also based on what you've seen from the Phase II study. So this goes for the agreement terms.

Can you remind me the first question, please, RK?

Do you have any idea of which of the Phase III studies we could see data from and anything on either the timing? Or what data we could be seeing or from which study you could be seeing data?

So we could refer only to what AstraZeneca are saying. And while they are reporting continuously data on Phase II studies as this year in ASCO and in the conferences, the Phase III results according to their guidance is after '27, meaning '28, and no, it doesn't mean that it couldn't be earlier interim analysis and other options, but the actual formal guidelines are after '27 for the Phase III studies.

This concludes the Q&A session and Compugen's investor conference call. Thank you for your participation. You may go ahead and disconnect.

Hello, everyone. Welcome to Jones fireside chat with Dr. Eran Ophir, President and CEO of Compugen. I am Danya Ben Hail, an analyst at Jones Research.

Compugen was doing computational target discovery long before AI and biotech became a buzzword. Today, we'll explore what it really takes to turn predictive biology into first-in-class immuno-oncology drugs. Eran, thank you for joining us.

Thank you Danya for inviting.

Let's start with the foundation. Compugen repeatedly uncovers first-in-class targets where our traditional discovery methods stall. Every AI company today claims a data advantage. Getting started, can you give a brief overview on Compugen's platform and what makes it unique?

So first, as you mentioned, we are not just a recent AI story that's saying that we're doing AI, just to have the buzzword in the name of the company. We're doing it a long time ago. We have built the system. We have this engine called Unigen. And so this is one.

I think the second part is the stage of the research and development that we are focused in, which is relatively unique. I mean you gathered a really nice group of companies in this set of fireside chat with AI companies. Most of them are doing, for example, AI to identify the drug targets, to identify easier, maybe better antibodies, small molecules. But we are focusing on the very first stage of the research and development on the drug target itself, which target, which molecule, which protein in the human body we should target to really modify it to have eventually effect in cancer patients. And there are not many companies that are focusing on this relatively challenging first part of bringing novel targets.

And there are definitely not many companies that have shown again and again that they can bring novel targets and that they are generating clinical data and further validation by pharma companies collaborating on the targets. So I think this is another differentiation that the platform is validated.

And finally, is not only the AI and the machine learning and the algorithms and the database that we have that we built along years. It's really the end-to-end capabilities. In one company, we have this know-how, again, developed along years of using the right algorithm, right data sets, asking the right questions to identify the drug targets, but then also how to validate, how to eliminate targets that should not proceed, how to choose the most promising one to take into the clinic. And then in the clinic, we're exactly to take them. I think also this end-to-end approach is relatively unique for Compugen.

Yes. That completely makes sense. So I mean, you through the proprietary algorithms, your decades of accumulated wet lab and you have the specific way to map the immune evasion mechanisms. So all of that together brings it, gives you this advantage? Or is there anything very specific and unique that you would like to say this is what -- this is the point of uniqueness or...

So it's absolutely that. It's the combination of how these pieces reinforce each other. We have great algorithms. We're mapping the tumor microenvironment for years from every different angle you can imagine, transcriptomics, proteomics to try to learn about the human tumor microenvironment. But then also the convergence of disciplines, having data scientists, biologists and clinicians sitting together in the same table, asking themself what clinical problem we can solve, but asking a specific biological question and how we can use the algorithms and data science to help us identify the targets to solve that problem. So this is one.

And second is that this is a flexible approach. So the database continue and feeds itself. We continue to generate data all the time from the preclinical data, from clinical studies we're doing, we're sequencing patients. Whatever sample we can put our hands on, we're sequencing and this feeds back into the database to enable us to choose even better targets for the next round.

Yes. Given the high failure rate of novel biology, so why is finding new targets still the most valuable application of your platform? And how do you protect against the existential risk of novel target failing in Phase II?

So first, I don't think this is only a problem of a novel target, which fails in the clinic. This is -- the biggest challenge is how to really predict clinical success in the preclinical package. For example, if you want to generate a better molecule for a known target, then you need sometimes to be better than the existing one and you need to be differentiated. So you can -- even if your molecule is working because the target is validated clinically, you are not better than the existing drug.

So there are many challenges also for developing better drugs for known targets, especially, I think these days when China competition rising up, I think that really bringing the target itself and not just trying to develop better drugs on known targets is something which is still a challenge, something that we have a know-how and advantage in how to do it. And again, given the recent competition for China, I think this is still something which is unique and still is -- have less competition and less competitive pressure from others.

Yes. I mean you're right. And the industry has seen like high-profile failures. But I will point that especially in targets that looked great in silico, but failed in humans. So how does Compugen ensure its models are capturing real drug responsive human biology rather than just finding statistical noise in massive multi-omics data sets?

So this is a very important question. It has multiple layers of the way we ensure it. First of all, we are not looking into statistical noise in data sets. We are starting with a unique clinical question. And we always, always, always make our analysis in the most relevant human samples. We never use proxies. We make sure that we have the right data set of human tumors from patients to answer the clinical questions that we have. But then it's the targets that we are working on are never remains in silicon.

Then come as discussed before, the end-to-end capabilities, how to build the biological package to really convince us, first of all, that the target is a valid target that can succeed in clinical settings. So we put huge efforts into doing all the right experiments. And maybe I can take an example. the COM503 that we licensed to Gilead, it haven't yet proven clinical, at least we didn't disclose any clinical data, but at least the package was convincing enough for a few pharma companies to chase it and eventually, we got this $90 million -- $850 million deal, $60 million plus $30 million and all of that. So the package was convincing also to others, let's say that.

So the COM503, we started with a very unique clinical question about resistant mechanism to PD-1. And then we looked into patient samples in Unigen for some of the samples, we actually -- for that specific question, we collected more samples. We have a very good collaboration with hospitals in Israel, and we get every day samples from hospitals, from surgeries.

And then we made the discovery itself. So a disease relevant that source for the data in which we found the discovery. But then also the validation stage. We try to rely as much as we can on human samples directly ex vivo from patients, showing the activity of the drug in the most relevant system. Yes, sometimes mice could be relevant, but definitely, the focus should be on the human systems on a very rigorous validation and high bar for targets to move forward.

And then at the end of the day, also in the clinical settings to really use the computational tools to identify the patient that could benefit, giving all the data that we have. So I think this whole package going from discovery in human samples all the way to the clinical settings with a very strong biological package around it, experimental package is key to what we do. And this is, in a way, taking this initial in silico prediction into a target with full biology around it that is sufficient for us to move forward with.

Yes. And that gets into -- so you're one of the first companies to bring computationally discovered targets into human trials. What have regulators thought to you about how they evaluate preclinical packages that were rooted in predictive algorithms?

So from our experience, we didn't have any issues because eventually, we're not bringing them in silico package. We're bringing them the experimental package, the preclinical package, a lot of focus on safety. So the packages that generated for our computationally discovered targets had never any issues with regulators. We always had a nice path into Phase I INDs, acceptance and all of that. So if you generate the right package, the fact that the target initially was discovered in silico doesn't really matter much.

And okay. So with AI being implemented across the entire R&D path, what is the hardest biological constraint in drug discovery, whether it's tumor microenvironment complexity or immune pathway redundancy that no algorithm can simply optimize away?

So it's exactly that. I think that eventually, the ability to model the complexity of the human in our case, immune system, but in general, the immune system is extremely challenging. So what we are doing to address that, as mentioned before, we are really modeling and mapping the tumor microenvironment from every angle you can imagine, we definitely use the recent AI tools, which allows us to do things that we couldn't have done before, absolutely.

We could navigate now in this complex, huge complexity of data in a way that we couldn't have navigated before, definitely. And still, we're not in the place we can say, okay, we can push on a button and we get a target. This is a very complex system, very complex trial and error approach. But definitely, the tools are improving every day, and we definitely can do today things we couldn't have done in the past.

All right. So there is biological redundancy is something that AI probably can't fully predict today. Hopefully, we'll get better at. So does that explain your shift into looking at combination therapies like in your ERG inhibitors with TIGIT or PD-1 inhibitors? How does the platform account for these complex interactions?

So first of all, I think that in general, oncology is going into combinations. We know that a single drug can achieve, in some cases, significant effects. But in many cases, you need to combine different mechanisms. And this is, again, where Compugen fits in. We are bringing new mechanisms that could work as monotherapy, but could definitely also work in combination. For example, the biology of PVRIG, which we think is very relevant for ovarian cancer.

The focus now in the MAIA study, and we'll discuss it in a second, is on the monotherapy activity of COM701, the blocker of PVRIG that we identified computationally, but the next steps could definitely also be combinations. But this really depends on the patient population, what you can tolerate, the ability to combine.

Again, PVRIG, for example, COM701, the blocker of PVRIG has a very good safety profile. So it's relatively easy to combine. So I think combination is definitely something that is relevant. It's not necessarily for AI discovered or non-AI discovered targets.

Yes. And for PVRIG, what gave you the confidence to take it into preclinical and then clinical studies? What's in silico signals convince you to continue with that?

So first of all, yes, PVRIG identified computationally. There were no publications around it. The academical community didn't know PVRIG when they identified it, which is a good start, but definitely not the important part.

The important part that we started to explore it, we identified it has a very, very different biology, different from TIGIT completely from PD-1. So we don't only have a new checkpoint. It's a checkpoint that could have different consequences when you block it compared to other checkpoints. And this is what we saw preclinically, and this is exactly what we saw clinically.

In patients, we have seen this unique biology translating into activity. As mentioned before, we're sequencing the patients. We looked in patients treated with COM701 before, and we take biopsy also and samples after treatment. And we saw how COM701 can modulate the tumor microenvironment in a way that we think is unique to its biology.

And then also the clinical signals. We have, for example, a patient with PD-L1 negative ovarian cancer that failed all the standard of care treatment could receive. She received COM701 in monotherapy and that computational prediction prolonged the life of that patient. She will remain on the study for 2 years. So the goal of the current MAIA study is to take the signals we have seen -- I mean, so the drug is active. And PVRIG is not in silico prediction now. We know it's an active drug that could modulate and affect the progress of ovarian cancer tumors. We also have seen signals in other tumors.

And now the MAIA study is a study we are doing that after seeing signals in the last line, platinum-resistant ovarian cancer, patients who failed everything. Some of them had 10 prior lines. We are now taking the signals we have seen and a very good safety profile, into an earlier line of platinum-sensitive patients, ovarian cancer, second, third line. And the goal of the study is really to see can COM701 in monotherapy mediate significant monotherapy activity and prolong the progression-free survival of these patients who have no standard of care.

These patients have a huge unmet need in the second, third line, they receive the platinum, and there is no treatment approved to maintain them from becoming platinum-resistant, and this is the goal of the MAIA study.

And we expect interim data readout in first quarter '27 correct?

Absolutely. We expect to have the interim analysis, which will include the meaningful data progression-free survival and any other clinical signals by Q1 '27, yes.

And if the data is strongly positive, does Compugen take the leap into building its own late-stage clinical infrastructure? Or does that milestone simply trigger another out-licensing event?

Yes. It's a good question. It will depend on the actual data, magnitude of effect, whether we will think that -- remember, again, there is a huge unmet need. So definitely thinking about this exact patient population and now we can go to registration fast will be the first priority, either we can do it alone or maybe we think that the best thing will be to partner to move fast and aggressive into Phase III.

In addition, we're talking about an adaptive trial design. So we can also add additional arms to continue to explore the possibilities. For example, we can combine COM701 after showing the monotherapy with bev, we can combine it with ADCs and then open also more opportunities going earlier, later in ovarian cancer and also into other indications.

Yes, especially taking into account it's a very competitive space right now.

It's a very competitive space, but there are very few nontoxic agents, which are fighting for that space. And patients after 6 cycles of platinum, some of them may want to have a bit of an easier drug than an ADC, which is eventually toxic. And the pharma companies, which are fighting between themselves with the different ADCs also want a way to differentiate. So having a combination partner, which is combinable and ideally, again, the durability we have seen in the last line will translate here as well could also be a matter of priority for -- to differentiate in the competition. for pharma companies.

Yes. Makes complete sense. You've secured impressive validation with your AstraZeneca and Gilead partnerships, bringing in substantial milestone potential. Is the monetization strategy to continue offload these assets early? Or yes, so do you intend to keep full commercialization right for future discovery assets beyond what you have currently in the pipeline?

So I think it's really going to be program dependent. First of all, I think with the financial stability we have now after the monetization of a small portion of the AstraZeneca royalty last year. So we now have cash in '29. So we have financial stability, and this gives us the freedom to make our choices because we don't have to rush and out-license necessarily to maintain the company alive.

So if we think for a specific program, if we generate additional data in a Phase II or maybe even a Phase III study, if it's a small and focused one, we will bring most of the value for ourselves and for the shareholders, we'll do so. If we think that we need now a very aggressive clinical program, and we need a pharma partner to move aggressively in multiple fronts, and this will be the path that will bring most of the value to the company and our shareholders. This will be the path. eventually it's going to be program dependent.

Yes. And to summarize this part, so where does your platform create the sharpest shift in return on investment? So is there financial and operational leverage found in accelerating early target discovery? Or is it derisking downstream clinical decisions? Where is it at?

So I think, first of all, the platform is focusing on bringing new biology, bringing new targets. As mentioned before, pharma companies tend to hear -- also biotechs to hear around the same target, same biology, and we're bringing new biology that could really enable more new options for patients.

But also as discussed, the rigorous process that we are doing, and this is yet to be proven, but I think this is ongoing, should bring us not only targets with new biology, but also targets we have more probability of success. We discussed before the COM701 data, and we have the readout by 2027. We didn't discuss, for example, TIGIT. TIGIT is an example of target we identified a long time ago, definitely multiple failures for TIGIT.

I think this is a case of a target that is definitely clinically active. It's clear that TIGIT is active. We've seen it in multiple trials, but some of the initial assumptions that some of the drug makers had on TIGIT maybe overestimated the activity, maybe didn't choose, in this case, the right format.

In this case, for example, I think that with the right format with the bispecific antibody of AstraZeneca, they will leverage also the TIGIT biology that we initially identified. We have 11 Phase III trials ongoing. It was interesting to hear AstraZeneca's management in the recent ASCO event talking about the way they see the evolving data for rilvegostomig.

They talked about stabilization of the responses. They showed in ASCO durable responses in early trials. They talk about combinability. So I think that also TIGIT, which is a target that we identified and had definitely a roller coaster of ups and downs. I think that AstraZeneca with their way of developing it, the clinical strategy, the bispecific will also eventually make the most potential of this TIGIT computational discovery.

Can you just highlight the differences of your TIGIT asset compared to the prior failures, just to clarify to people listening in.

Absolutely. It's very important. So again, this is an example in which doing what we think is the wrong drug format could make a difference. So most of the initial developers who develop TIGIT has an Fc active. I will not go through all that biology, but that was -- we think was not the right format. It caused safety issues. It had challenges to combine.

And eventually, it was difficult to keep patients for a long time on the study because of the safety. But even maybe more importantly is the fact that AstraZeneca are the only ones for developing TIGIT as a bispecific for PD-1 and TIGIT. This has some mechanistic advantages that we see for other bispecifics as well that could be more active and there's some evidence for that, could be more active than PD-1/TIGIT combinations that done -- others have done.

And this also allows a different clinical strategy, much easier to combine, less burden of showing contribution of components of the bispecific. So I think it's the format of the antibody, the clinical design strategy and the combination around all these 11 Phase III trials, most of them or many of them with ADCs that will put AstraZeneca's TIGIT antibody in -- which is partnered from us, obviously, in a different position than the other TIGIT were.

Yes. Thank you for that. So as we look into the second half '26 and beyond, what are the most important clinical and strategic milestones investors should be watching?

So first, obviously, is our own asset, COM701, the MAIA study by Q1 '27, we're going to have the meaningful data to show if COM701 can really drive monotherapy activity in this population of ovarian cancer.

Rilvegostomig, the expectation from AstraZeneca for Phase III readout, which is going to be the meaningful one is after '27. But we see all the time accumulation of data like in the recent ASCO showing again the durability, the safety, the potential of different combinations. So I think with the accumulation of data, it will show eventually that this molecule of TIGIT is doing something else.

And then obviously, the COM503 called now GS-0321 that we licensed to Gilead. We're already in the clinic for more than a year now and the progress through Phase I. We don't have yet disclosure for exactly when, typically when working with the pharma company, but definitely keep an eye for the COM503 readouts and the early pipeline, the computational discovery platform, we have financial stability. We have a validated engine that we will continue every day to leverage and bring more assets. And along the coming year, again, we don't have specific guidelines, but definitely, we will report on the early pipeline when time will come.

Great. Looking forward. So to close up, we'll do a rapid fire questions section. So let's get going. Most overhype claim in AI drug discovery today.

I think that people that say that AI is going to dramatically increase clinical success rate in cure any disease are still a bit early. Things are moving fast, so difficult to make predictions. But for now, human system is too complex to be solved with a push of a button.

So most underappreciated bottleneck?

The clinical development. The community must do something, especially in the U.S., but also in general about making trials cheaper, faster. We have great targets. We need to test them. We need to see in patients if it works or not. And for that, we need a better system for faster and more cost-effective clinical development process.

And hopefully, I can help with that.

Absolutely.

What is the one metric investors should focus on that actually captures platform value?

So eventually, looking at the totality of the assets that we have and that we will have and we will disclose our ability to open new target space, new mechanism to bring additional BD activities to bring more clinical validation for our internal and partnered assets.

One proof point investors should demand over the next 2 to 3 years?

Eventually, the proof is in the pudding. I can sit here and tell about the processes we are doing and the rigorous validation. Eventually, an asset should show success in clinical translation. This is what we work for. This is what the investors should wait for.

Great. Well, thank you for the thoughtful discussion. What you've built at Compugen shows that computational biology isn't about shortcuts. It's about uncovering biology that was invisible to traditional methods, and we are looking forward to the future updates.

Thank you, everyone, for joining us today. Enjoy the rest of the upcoming sessions.

Thank you, Danya.

Ladies and gentlemen, thank you for joining us today. Welcome to Compugen's First Quarter 2026 Results Conference Call. [Operator Instructions] As a reminder, today's call is being recorded.

I will now hand the call over to Lindsey Trickett, Head of Investor Relations and Corporate Communications to begin. Lindsay, please go ahead.

Thank you, operator. Good morning and good afternoon, everyone, and welcome to Compugen's First Quarter 2026 Financial Results Conference Call. With us today are Dr. Aron Ofer, President and Chief Executive Officer; David Silverman, Chief Financial Officer. Dr. Michelle Miller, Chief Medical Officer, will join us for the Q&A portion of the call.

Before we begin, I'd like to remind you that during this call, the company may make projections or forward-looking statements regarding future events, business outlook, development efforts and their potential outcome, the company's discovery platform, anticipated progress and plans results and time lines for our programs, including disclosure of clinical data, financial and accounting-related matters as well as statements regarding our cash position and cash runway. We wish to caution you that such statements reflect only the company's current beliefs, expectations and assumptions and that actual results, performance or achievements of the company may differ materially. These statements are subject to known and unknown risks and uncertainties, and we refer you to our SEC filings for more details on these risks, including the company's most recent annual report on Form 20-F. The company undertakes no obligation to update projections and forward-looking statements in the future.

With that, I'll now turn the call over to Dr. Erin Ofer, President and CEO.

Thank you, Lindsay. And good morning and good afternoon, everyone. Before I turn to our business update, I want to take a moment to formally welcome Lindsay, our new Head of Investor Relations and Corporate Communications to Compugen. Lindsay joined us with strong experience in Investor Relations, and we are thrilled to hover leading our communication with the investment community. Welcome, Lindsay, and we're glad to have you on board.

Now let's start with our business update. 2026 is shaping up to be a significant year for Compugen, and I'm pleased to share our progress in the first quarter of 2026 as we continue executing on our strategic priorities. Starting with our fully owned clinical program, COM701, a potential first-in-class antibody target PVRIG, which is an immune checkpoint with unique biology much differentiated from other checkpoints, including PD-1 and agent. We believe this unique biology underlies the clinical activity demonstrated for COM701 in less inflamed indications such as ovarian cancer.

As a reminder, at ESMO last year, represented the full analysis of clinical data showing that COM701 in monotherapy and combinations was well tolerated and showed consistent, durable responses in patient with heavily pretreated platinum-resistant of ovarian cancer. Based on these results, we decided to progress the development of COM701 and test it in earlier settings of ovarian cancer as a maintenance therapy in patients with relapsed platinum-sensitive ovarian cancer that responded to their most recent line of chemotherapy.

The rationale is to allow COM701 to induce its antitumor activity in early relying patients with lower tumor burden, less compromised immune system and by that, increase the likelihood of these patients to benefit for COM701 unique mode of action. For this purpose, we initiated the marioviroAdaptiv platform trial. In such study 1 of this trial, Comsebralis randomized is maintenance monotherapy versus placebo in patients with relapsed platinum-sensitive or bearing cancer. We're actively enrolling patients in clinical sites across the United States, Israel and France.

Having all sites open and enrolling, spanning leading academic centers in U.S. and Israel, as well aside from the cage Ginko French corporative group gives us confidence in our ability to complete enrollment on schedule for having the Mivan median PFS data at interim analysis by Q1 2027.

This patient population comprised of those progressing post PARP inhibitors and/or EM overall lot candidates for sustreatment represent a significant unmet medical need with no current standard of care. We believe that clear prologation of PFS in these patients could inform a registration task for COM701 and make it a potential backbone for drug combinations in this population while also enabling a potential broader clinical development plan across earlier and netilize of ovarian cancer treatment as well as in other indications for clinical signals previously seen for COM701.

In addition, we're happy to see our partner AstraZeneca's progress on their broad rilvugosomic program. We remain confident in real potential based on its differentiated bispecific antibody formats in addition to clinical and combination strategies. Last month, AstraZeneca presented multiple abstracts featuring yield at the AACR Annual Meeting in San Diego, reinforcing our confidence in the differentiated design and growing potential. This includes preclinical data demonstrating potential opportunities for Velvet as an IO backbone for combination and also laboring data from the Destiny gastric 3 Phase II trial evaluating relevant combination with the blockbuster ADC and HER-2 and chemotherapy as first-line treatment for HER2-positive gastric cancers. These data showed promising antitumor activity and also demonstrated combinability of real from a safety perspective.

Overall, these ACR publications continue to reinforce our confidence in Relday as AZ continue to advance it along 11 Phase III trials across multiple indications, including the recently opened trial in gastric in combination with the closing 18.2 ADC.

With that, we are looking forward to the release of additional clinical data along the year, including at the next ASCO meeting at the end of the month. As a reminder, AstraZeneca's previously estimated a nonrisk-adjusted peak annual revenue potential of more than $5 billion per Wheeler and were eligible for additional $95 million in future regulatory and commercial milestone payments plus mid-single-digit tiered royalties on sales.

Moving to GS-0321, formally known as Confio, our potential first-in-class anti-IL-1 binding protein antibody licensed to Gilead. GS-0321, a novel antibody approach to any cytokine biology for the treatment of cancer, potentially overcoming the limitations of direct cytokine administration. The ongoing Phase I dotscalation trial continues to progress as we planned. As a reminder, we received to date $90 million from Gilead of this asset, on eligible to receive up to $758 million in additional milestone payments plus up to double-digit tildalties.

Now to the early-stage pipeline, and enogen discovery engine. Beyond our clinical assets, we continue to invest in our early-stage immuno-oncology pipeline. UNIGEN our AI-powered computational target discovery platform has already discovered the targets of COM701, COM902, NGS. We remain committed to identifying and advancing the next wave of innovative programs, grounded in novel mechanism of action designed to activate the immune system against cancer. Importantly, we have a solid financial position with a cash runway expected into 2029 following the December 2025 transaction with AZ, through which we received $65 million in nondiluted capital by monetizing only a small portion of our future rebroalties.

Our financial stability allows us to fully focus on advancing our pipeline and reaching key value-creating buystones with both our internal and partnered programs. And throughout all of this, we continue to benefit from a deeply talented and highly committed tumor Compugen I am proud of what we have built and energized but opportunities ahead.

With that, let's hand over to David for the financial update before we open the floor for Q&A.

Thank you, Evan. And I would like to add my own warm welcome to Lindsay as well. It is a pleasure to have you join the Compugen Ltd. and Team Lindsey, and we look forward to working together. I am pleased to say that we continue to advance in 2026 with a solid balance sheet and financial flexibility. Cash runway, assuming no further cash inflows is expected to fund our parity plans into 2029. We anticipate using this runway to continue advancing our COM701 platinum-sensitive ovarian cancer trials, may ovarian and to support the progression of GS-0321 in the clinic, together with continued investment in our early-stage pipeline.

Now going into the details, I will start with our cash balance. As of March 31, 2026, we had approximately $134.9 million in cash, cash equivalents, short and bank deposits and investments in marketable security. Revenues for the first quarter of 2026 were approximately $2.2 million compared to approximately $2.3 million of revenue for the comparable period 2025. The revenues in the third quarter of 2026 and 2025 reflects the recognition of posted of both the upfront payment and the IND milestone payments from the license agreement with inlet. Expenses for the first quarter of 2026 were in line with our plans. R&D expenses for the first quarter of 2026 were approximately $6.9 million compared to approximately in the first quarter of 2025.

The increase is mainly due to an increase in clinical expenses related to my ovarian trial as well as higher drug supply costs supporting our trials. Our G&A expenses for the first quarter of 2026 were approximately $2.3 million compared to approximately $2.4 million for the comparable period 2025. For the first quarter of 2026, our net loss was approximately $7.7 million or $0.08 per basic can be look a share compared to a net loss of approximately $7.2 mill or $0.08 per basic and diluted share in the first quarter of 2025.

With that, I will hand over to the operator to open the call for questions.

[Operator Instructions] The first question is from Lana Grab of [indiscernible] Partners.

Thank you for the question. Lindsay, welcome. Nice to see you here. Going into ASCO, I wonder if you could talk about more specifically the data sets going to prevent witruvagostimig? And helps set the context for what we should expect to see? And are there benchmarks that would -- that we should be keeping in mind when we review the data set?

Sure. Thanks, Dana. So we're talking about 2 DAS assets, clinical data. Obviously, the actual data is not released yet, and I would be cautious on setting expectations AstraZeneca but over, we talk about on the I-SPY trial in the testing will be cost to me in elegant settings with Synet, which is by itself a blockbuster drug, which is very exciting to see these combinations. .

Again, I would be cautious about the expectation. But I think looking again, and this is a platform trial, so really trying to look across the nutrandomize study, but trying to look about driver versus other data sets. The combinability is again going to be very important and to show again how the Fc reduced format of rivagostamig is easy to combine with such ADCs.

And then the second set is the Gimi which is in combination with chemotherapy. And here again, we'll be good to see. I think it's a bit of a longer follow-up from what was reported before. So we to see about the long-term effect of the PSS -- I'm not sure if there will be an OS data, but how the long-term effects are shaping, including the long-term safety in combination with chemo, having in mind that there is -- for this trade, there's an ongoing Phase III study ongoing. So I guess the comparison to sterocontrol should be with caution and still probably is going to be made.

The next question is from Stephen Wiley of St.

Maybe you can just talk a little bit about how you're thinking about disclosing future development candidates that are discovered off the Inogen platform. I think the IL-18 binding protein antibody wasn't announced until it was ready for clinical development. Is that kind of how we should expect incremental assets to emerge out of the pipeline once they're ready for an IND submission?

Thanks, Steve. So I think it's really dependent. Eventually, definitely, the biggest group in Compugen is the 1 that continue to work to bring additional innovative assets like COM, which is called today 031 -- specifically for that asset, it was right for this asset and for Compugen at these times to out license it in the clinical stage. So this also influenced the stage in which we disclosed it. It was relatively early. But it doesn't mean necessarily that we have any specific guidelines that we are porting on early assets only when it's ready for IND or only on its election. It really depends on the actual assets on the stage of derisking in which you want to start comment and committing.

So again, I wouldn't learn too much from the story filing by the protein other than the fact that it was another demonstration of our competition platform can bring such innovative approaches in that case, not only first-in-class asset, but the first-in-class approach to highlight cytokinebology for equites cancer and we are looking into different MOAs not necessarily similar to that to bring again another innovative options that could really make difference to patients.

The next question is from Renan GershofOpeneier.

Wondering if -- could you remind us if the ovarian trial, is that stratifying for patients who are PD-L1 or PD-1 expression status. And I also want to ask when we see the interim data in the first quarter, will given that this is an adaptive trial, would that mean that the interim data could inform some change your design? Or would you simply keep going as planned?

Thank you, Live I think Michel can take this one. .

I think to take this one, yes. So the mylovarian trial actually is not stratified according to PD-L1 subgroup. We are stratified by second versus third-line treatment. And in '17. When a result, we have multiple options ahead of us in terms of adjustments to the trial. So we would consider adding additional arms and a lot of it's going to depend on the totality of the data and also plans towards engaging with the regulators and steps towards a pivotal trial.

Equal additional comment a little about the PDL1 certification, I would like to remind you that FIBRA probably because it's unique biology, we saw in other indications, specifically in ovarian cancer, we saw responses at pro-PD1-pst and PD-L1-negative patients. So for now, we didn't see not necessarily like for other checkpoints that the PD-L1 subset is the 1 responding to COM701. And again, I think this is because that unit biology, very much differentiated from PD-1 so again, not necessarily PD-L1 certification here.

The next question is from RK of H.C. Wainright.

So a couple of more questions on the ovarian cancer trial. So now that you have all these sites active, what is -- any commentary on the enrollment status itself and also because this is an event-driven trial, and any commentary on required events that needs to happen for the interim analysis? And the third question is, what -- what are you assuming for the control on PFS? And what sort of hazard ratio do you need to see to consider that as a win?

Michel, do you want to take .

Yes, sure. So firstly, with respect to enrollment, we're not commenting at this point in time, but I will say to you that we are on track our interim analysis as planned in the first quarter of 2027. And our participating investigators have a high level of engagement and are working really well with us. Regarding the events and the benchmarking. So the trial is an exploratory trial. And so at this point in time, we don't know the full magnitude of benefit, but the benchmark for the control arm from prior clinical trials in the second line and third line of maintenance in those trials where patients did not get treatment. The same patient population had a benchmark of approximately 5.5 months, although there was a range. So in some studies, it was as low as 3.8 months and others as high as 5.8 months. we're hoping to be able to show that there is meaningful single-agent clinical activity of COM701 and we've hypothesized that we would like to see a 3-month or greater improvement of the benchmark PFS.

This concludes the Q&A session on Compugen's investor conference call. Thank you for your participation. You may go ahead and disconnect.

Ladies and gentlemen, thank you for joining us today. Welcome to Compugen's Fourth Quarter and Full Year 2025 Results Conference Call. [Operator Instructions] As a reminder, today's call is being recorded. Participants from the Compugen team are Dr. Eran Ophir, President and CEO; and David Silberman, Chief Financial Officer. Dr. Michelle Mahler, Chief Medical Officer, will join for the Q&A. Before we begin, we would like to remind you that during this call, the company may make projections or forward-looking statements regarding future events, business outlook, development efforts and their potential outcome, the company's discovery platform, anticipated progress and plans, results and time lines for our programs including disclosure of clinical data, financial and accounting-related matters as well as statements regarding our cash position and cash runway.

We wish to caution you that such statements reflect only the company's current beliefs, expectations and assumptions, but actual results, performance or achievements of the company may differ materially. These statements are subject to known and unknown risks and uncertainties, and we refer to you our SEC filings for more details on these risks including the company's most recent annual report on Form 28. The company undertakes no obligation to update projections and forward-looking statements in the future. With that, I now turn the call over to Eran.

Thank you, operator, and welcome to everyone joining our call today. On today's call, I would like to highlight some of our key achievements in 2025 and outline our planned strategic priorities for 2026. During 2025, we made progress across our business scientifically, operationally and financially, including the following key highlights.

We extended our expected cash runway into 2029, assuming no further cash inflows through a nondilutive transaction with AstraZeneca tied to relive their differentiated PD-1 TIGIT bispecific, the digit component of which is derived for COM902, our fully owned Fc-reduced anti-TIGIT antibody. We also diversified our leadership team as I stepped into the CEO role in September 2025 and not transition to Executive Chair.

On the clinical side, we are on our clinical programs initiating new clinical trials with our wholly-owned potential first-in-class anti-PVRIG COM701 and our potential first-class anti-IL-1 bilingprotein antibody GS-0321, lesser to Gilead. We also advanced our clinical footprint, opening sites in the U.S., Israel and France in our COM701 clinical trial. In addition, represented clinical updated ESMO and SITC conference for COM701 and GS-0321, respectively.

So let me elaborate on each of these highlights, starting with the most recent update our December 2025 strategic transaction with our partner, AstraZeneca, where we monetize a small portion of our future relive royalties to AstraZeneca. This deal is important today and for the long term because -- it added $65 million in upfront nondilutive capital from AstraZeneca, extend our expected cash runway into 2029.

It provides an additional $25 million at the next milestone payment which is BLA acceptance and thereby increases our total remaining milestones to up to $195 million for around $70 million previously. And we retain the majority of our royalty interest levering our economics fundamentally intact. So both before and after the deal, we remain eligible for up to mid-single digit tiered royalties from real bests. We believe this deal allowed us to unlock value today to continue advancing our innovative immuno-oncology pipeline, including COM701, GS-0321 and our early-stage pipeline and it allows us to reach both internal and partnered catalysts.

All of this without compromising our long-term upside in Relay, a potentially multibillion-dollar assets. And to put this into context, rilve is being advanced by AstraZeneca, a broad late-stage development program, including 10 active Phase III trials. AltraZeneca, previously estimated a nonrisk-adjusted peak annual revenue potential of more than $5 billion for rilve.

Next, let me briefly touch on the leadership transition. I'm excited and privileged to have had the opportunity to step into the role of President and CEO in September 2025. With Anano serving as Executive Chair, we believe we had a leadership structure that combines operational focus and strategic continuity, a strong foundation for Compugen next phase of growth.

Turning now to clinical execution, starting with COM701. In the myOranclinical trial, in the platinum-sensitive ovarian cancer, we initiated dosing and expanded our trial footprint globally by opening trial sites in the U.S., France and in Israel. I also want to highlight the data we presented at ESMO last year from the Cove pooled analysis of Phase I clinical data in platinum resistance ovarian cancer. The pooled analysis demonstrated that Comseveran was well tolerated and showed consistent, durable responses in patients with heavily pretreated platinum-resistant ovarian cancer, particularly in those without liver metastasis, representing patients with oral disease burden and potentially less immunosuppressive tumor macroenvironment.

We believe this data support the rationale for advancing COM701 in the early line settings as maintenance therapy in platinum-sensitive ovarian cancer. These programs underscore our commitment to pioneering innovative biology. Regarding GS-0321, our partner program with Gilead, we initiated dosing in a Phase I dose escalation and expansion trial and subsequently presented a trial in progress update at CIT.

Overall, we believe that our achievements in 2025 set the stage for continued execution in 2026, which transitions me nicely to our 2026 strategic priorities, which includes the continued execution of the MAIA-ovarian conductive trail. The first HAP trial is a randomized trial, comparing COMSEvoRan monotherapy to placebo in the maintenance setting of platinum-sensitive ovarian cancer, a setting where there is a significant unmet medical need and no current sort of care.

We are on track to have an interim analysis in Q1 2027. This data could lead to maintenance monotherapy past registration and for a potential bedbone for drug combination in this population. We're also enabling potential broader clinical development plan across ovariccancer lines of treatment and in other indications or clinical signals were seen for COM701. In parallel, we're executing on our Phase I trial with GS3 as a reminder, the first pressure was dosed in January 2025. GS0321 is our potential first-in-class anti-IL-13 binding protein antibody licensed to Gilead.

We believe that the key differentiator of GS-0321 is that it is not a cytokine, but an antibody harnessing cytokine biology for the treatment of cancer. It's new in a cool mechanism and venoclinical data this approach by offer advantages on both safety and efficacy. Gilead has already paid $60 million upfront and an additional $30 million when it successfully achieved IND clearance, we're eligible to receive up to additional $758 million in future milestones and single-digit to low double-digit tiered royalties.

The ongoing Phase 1 constitutes 2 parts: Part 1, dose escalation and Part 2, motivation. In addition, we continue to track our partner AstraZeneca's progress very closely as they execute on their broad Phase III relay program. Given the recent history in the TIGIT field, it's worth taking a moment to explain why we maintain confidence. For us, the answer is clear, antibody format matters, and clinical and combination strategy.

So let me explain. On format, relve is an anti-PD-1 TG bispecific antibody that has the reduced FC functionality. This design delivers coordinated inhibition of both PD-1 and TIGIT on the same immune effector cells with cooperative binding, resulting in greater efficacy, the anti-PD-1 plus anti-TIGIT single combinations when tested in ex patients derived models of non-small cell and pence. In addition, this format using reduced safety functionality maybe reduce the unwanted depletion of immunofactor cells and maintain a favorable safety profile.

In vehicle, AstraZeneca trials are designed differently from some other companies digital, and also include novel combinations like ADCs that have not been tested thus far. So to summarize, our confidence in relve is based on its differentiation at different drug formats in a different clinical trial and combination strategy.

Lastly, turning to our early-stage pipeline. With our current cash runway expected into 2029, 2026 will be our continued focus on our early stage pipeline, which is managed by the largest team within Compugen. UNIGEN AI machine-based computational engine that generated COM701, CO09022, NGS-0321, and we remain committed to investing in this differentiated discovery platform.

So stepping back, let me summarize where we are today. We have a unique positioning, solid financial outlook that enable us to continue and leverage our computational drug target discovery engine to deliver the next generation of novel immuno-oncology assets. We have a clinical pipeline grounded in potential first-in-class immuno-oncology science, and we have 2 validating partnerships with AstraZeneca and Gilead representing approximately up to $1 billion in potential milestones plus royalties.

And our team is consistently striving to deliver at the highest levels. I'm incredibly proud of what our team has delivered and equally excited about the opportunities ahead. Thank you to everyone at Compugen for dedication. With that, let me turn it over to David for the financial update before we open the call for questions.

Thanks, Eran. I am pleased to say that we are advancing in 2026 with a solid balance sheet. Cash runway, assuming no further cash inflows is expected to fund our operating plan into 2029, and we anticipate using this runway as planned to advance our COM701 in the powering cancer trial, myaovarian and to support the progression of GS0321in the clinic together with continued investment in our early-stage pipeline.

Going into the details, I will start with our cash balance. As of December 31, 2025, we had approximately $145.6 million in cash, cash equivalents, short-term bank deposits and investments in marketable securities. The cash balance at the end of 2025 included the $65 million upfront payment from AstraZeneca for the monetization of a small portion of regarding future royalties. On the revenues front, we reported approximately $67.3 million in revenues for the fourth quarter of 2025 and approximately $72.8 million for the year ended December 31, 2025, compared to approximately $1.5 million and $27.9 million in revenues for each of the comparable periods in 2024.

Revenues for 2025 include the upfront payment of $65 million from AstraZeneca and a portion of the Avon payment and the IND milestone payment from the license agreement with Gilead while the revenues for 2024 reflect a portion of the upfront payment and the IND milestone payment from the license agreement with Gilead and the $5 million clinical milestone payment from AstraZeneca.

Moving to expenses. R&D expenses for the fourth quarter of 2025 and for the year ended December 31, 2025, were approximately $5.5 million and $22.8 million, respectively, compared with approximately $5.9 million and $24.8 million for the comparable period in 2024. The decrease in 2025 was mainly due to lower clinical expenses resulting from winding down prior clinical trials, partially offset by an increase in clinical expenses related to the May variant trial initiated in 2025.

Our G&A expenses fourth quarter of 2025 and for the year ended December 31, 2025, were approximately $2.1 million and $8.9 million, respectively, compared with approximately $2.2 million and $9.4 for the comparable periods in 2024. Finally, on net profit. For the fourth quarter of 2025, we reported a net profit of approximately $56.8 million or approximately $0.60 per basic and diluted share compared to a net loss of approximately $6.1 million or approximately $0.07 per basic and diluted share in the comparable period of 2024.

Net profit for the year ended December 31, 2025, was approximately $35.3 million or approximately $0.38 per basic and diluted share compared with a net loss of approximately $14.2 million or approximately $0.16 per basic and diluted share in the comparable period of 2024. With that, I will hand over to the operator to open the call for questions.

[Operator Instructions] The first question is from Diana Graybosch of Leerink Partners.

This is [ Raviv ] on for Dana. First question I would have is, can you help us level set on what to expect in the 1Q '27 update with COM701 in terms of what we expect to see in that update? And then the follow-up to that is, can you help us understand the time line of what is required for that path to registration that you mentioned in the call.

Michelle, do you want to take it? .

Sure. I'm happy to take it. Thanks for the question. So the current trial is an adaptive trial design, and we anticipate that there will be data maturation in 1Q of 2027. And regarding the time line and what we'll be requiring for registration is really depend on the totality of the data. And we are planning for success and would have to consider other subsequent plans or trials which we are still in discussion. -- at that point in time, ready to disclose.

Yes, I think we can say a high level, as we said in the past, that there are a few opportunities here. I mean 1 is to continue if the data indeed is -- has meaningful clinical clinically, we can continue to pass for the trace monotherapy. It can open a path for combination strategies in that population. And of course, because we know that we have signal also in broken under education, a positive monotherapy signal in this trial also opened many other options. But I guess our first steps to be in that specific population following a positive results. .

The next question is from Josh Nickerson of Stephen. Please go ahead.

This is Josh on for Steve. Could you just remind us the cadence of potential outlying milestones for rebogostomag? And maybe just provide some color on the next trigger for milestone payment upcoming.

David, do you want to take it? .

Yes, sure. Joe, thank you for the question. Yes. As a reminder, so we did the deal is AstraZeneca in December, and we disclosed our next milestone with BLA acceptance on which we will be entitled to national $25 million on top of what we are already entitled to. So going forward, we will still be entitled to EUR 195 million in milestone from AstraZeneca the rilvegostomig deal. .

We are having some technical difficulties, just a moment please. Eran, Can you hear us? .

Absolutely. .

Maybe the speakers just disconnected. We'll move to the next question. The next question is from Swayampakula Ramakanth of H.C. Wainwright.

This is RK from HC Wainwright. So trying to think through the OVARIAN trial, previously you had stated you will have some interim analysis done in the second half of now it's moved to the first quarter of 27%. Just trying to understand the shift, is it -- is it more because you're adding additional centers? Or is it because you'll see a slower accumulation of events than what you initially modeled for.

Thanks, RK. So we reported that shift already in the previous quarter. And the reason back then was a bit slower. By the way, it's not only Compugen issue, but again, for us, it was a bit slower opening of the major academical U.S. sites, which we're very glad that now all of them are open. Actually, now all the sites are open, all 28 sites are open. We mitigate for that, gladly, we were approached by the Kajanio Group, which I actually have experience in very -- in that specific patient population. And they approached us to contribute to the study.

So gladly, they joined as well. We now have all the sites open fully on track to have the readout in Q1 27. Obviously, like in any trial, we need to see that the events are cumulating as expected. But other than that, everything is on track, and we continue to be in Q1 27 as we report also in the last quarter. So no change in this quarter For that. .

Okay. And then in terms of the AstraZeneca relationship, obviously, the recent monetization speaks to the alignment to the deep alignment that AstraZeneca wants to have with the drug? Are there any discussions of expanding the use of the COM902 derived jet in additional multispecific pharmas within the AstraZeneca portfolio.

So AstraZeneca are controlling givagostomig. We don't discuss with them the on plan. I mean we did see recently that -- and this is again, I think, illustrating the commitment for the program, we did see recently a new Phase III to now in gas treatment combination with coding ADC, which is now in clinicaltrial.gov. So this would be when it's activated the 11 Phase III trial.

So they are expanding with rilvegostomig. It's not specifically COM902, but sort of a custom which contains comment for comment too specifically, we fully own it. And obviously, it's a different opportunity that you can leverage in other collaborations, probably AstraZeneca with the Rivagostomig, that will move with that 1 and not with comment. And then the last

Last question for me is on the 0321. In terms of the data that's expected, would that be in any of the medical conferences or where would we see that data? And will we see more than just initial safety.

So -- we initiated the first patient was dosed at the beginning of 2025. By our agreement with Gilead, obviously, when we report data, it has to be fully aligned with them. For now, we don't have guidelines. But typically, and also I think what Gilead will do themselves is to report it in a scientific conference. And typically, it will include activity plus safety. But for now, we are not making any commitment because it will need to be in line with with Gilead.

The next question is from Leland Gershell of Oppenheimer.

I'm just wondering as we await the update on 701 in about a year from now, first quarter of '27. Just wanted to ask what you may plan to be presenting at the various oncology meetings this year, ESMO, SITC and so forth. Can you give us a flavor of what we might see out of Compugen through '26.

So overall, from what we currently disclosed and obviously, a long deal we might update it from what we currently disclose, we -- for the Gilead has just mentioned, we don't have any specific guidelines, but it could be a long medical conference along the year. AstraZeneca and again, it's AstraZeneca program and Atrzadecision, but they do have some clinical readouts this year and they might report it in some of the static conference. I didn't disclose yet when. And this is basically what disclosed for this year and obviously, next year would be the myStudyrit, which is an important one. .

This concludes the Q&A session in Compugen's Investor Conference Call. Thank you for your participation. You may go ahead and disconnect.

Ladies and gentlemen, thank you for joining us today. Welcome to the Compugen Third Quarter 2025 Results Conference Call. [Operator Instructions] An audio webcast of this call is available in the Investors section of Compugen's website, www.cgen.com. As a reminder, today's call is being recorded. I would now like to introduce Yvonne Naughton, VP, Head of Investor Relations and Corporate Communication. Yvonne, please go ahead.

Thank you, operator, and thanks, everyone, for joining us today. Here with me from Compugen team are Eran Ophir, our new President and CEO; and David Silberman, our Chief Financial Officer. Michelle Mahler, our Chief Medical Officer, will join us for the Q&A. Before we begin, we would like to remind you that during this call, the company may make projections or forward-looking statements regarding future events, business outlook, development efforts and the potential outcome, the company's discovery platform, anticipated progress and plans results in time lines for our programs, including disclosures of clinical data, financial and accounting-related matters as well as statements regarding our cash position and cash runway.

We wish to caution you that such statements reflect only the company's current beliefs, expectations and assumptions, but actual results, performance or achievements of the company may differ materially. These statements are subject to known and unknown risks and uncertainties, and we refer you to our SEC filings for more details on these risks, including the company's most recent annual report on Form 20-F. The company undertakes no obligation to update projections and forward-looking statements in the future. With that, I'll turn the call over to Eran.

Thanks, Yvonne. Good morning and good afternoon, everyone. I'm delighted to speak with you today as Compugen's new President and CEO. I'm really energized stepping into this role as such pivotal time for our company. Having led our scientific strategy in my forward position as CFO, I've seen firsthand how our science has evolved, and I believe we can deliver significant value for patients. So where do we stand today. Our fundamentals are strong and our strategy is clear. We are pioneers in computational drug target discovery. And we believe that our deep expertise in TIGIT biology is now gaining clinical momentum. I think that now it's a great time to highlight what makes us different in the TIGIT drug development space and why you should be paying close attention to our differentiated Fc-reduced anti-TIGIT programs and their advantages over Fc-active anti-digital antibodies. Reflecting on the history of drug development, one can appreciate that indeed choosing the right therapeutic target to cure disease is critically important.

But choosing the right drug format, which fits that specific target is just as important. We know that anti-TIGIT antibodies with the Fc-active format have not lived up to expectations. And most of these programs were discontinued. However, this did not surprise us because Fc-active anti-TIGITs can deplete TIGIT positive effect on T cells and T regs. This is not desired because 1 on efficacy. TIGIT is present on effector T cells. So similar to the action of anti-PD-1s -- invigorate these exhausted cells and avoid their depletion Two, on safety, TIGITs present on Tregs depleting peripheral there could result in immune-mediated side effects.

Fc reduced anti-TIGITs like our own COM902 in contrast, preserve and reinvigorate the factor T cells avoid depletion of peripheral Tregs and therefore, have the potential for improved immune activation and a better safety profile. It is notable that as earlier Phase II trials with Fc-active anti-TIGITS, safety was a concern with high rates of discontinuation due to sevens. This was also even more evident in the Phase III trials. For example, during the recent ESMO meeting, the presenter highlighted that TRPR-07 trial, adding Fc-active TIGIT to atezo resulted in these patients only receiving bid number of doses of 12 versus 17 in the atezo-only arm.

As a result, the patients receiving TIGIT PD-L1 combination received 30% less the -- antibody versus controller. So safety really impacted the ability to administrate treatment and, therefore, probably impacted the outcome. We've always advocated for the Fc-reduce formats, and we believe that the data starting to support our convention, not all anti-TIGIT antibodies are the same. And we believe the market is missing this. We believe our assets are positioned to capture the upside as new data emerges with readouts anticipated from 2026, provided our conviction proves correct.

This moves us to our strategy, which is rooted in science and focused on patients. We have 5 key value drivers, starting on Fc-reduced TIGIT programs. COM902 is one of the only 2 clinical stage Fc reduced antigenic monoclonal antibodies currently in clinical development. And importantly, it's fully owned by Compugen. Positive Phase III data from Arcus Gilead with the only other known Fc reduced anti-TIGIT monoclonal antibody is expected in 2026 and could be a real catalyst for COM902. Notably, recent overall survival data from their Phase II frontline gastric cancer study, which is the same setting as Arcus Gilead ongoing FET trial was presented at ESMO recently and showed a median Argo survival of 27 months versus 15 months or less for benchmarks, a meaningful signal for the Fc reduced class.

Next is Rifugostomix. Our partner, AstraZeneca Fc reduced anti-PD-1 TIGIT bispecific with the TIGIT component derived from RFC reduced high affinity COM902. Interestingly, cooperative bispecific binding might provide even further efficacy into PD-1 and TIGIT blockade, while in addition, potentially supporting an easier regulatory path. The potential commercial opportunity for real way substantial with AstraZeneca estimating nonrisk-adjusted peak year revenue target of more than $5 billion. We understand that AstraZeneca's ambition is for forever to replace PD-1 PD-L1 therapies and to service the backbone for future combination treatments.

Their broad development program spanning 11 Phase III trials across lung, gastrointestinal and endometrial cancers represent a potential significant value driver for Compugen as we're eligible for regulatory and commercial milestones and mid-single digit tiered royalties payments. Moving to Fc-reduced PVRIG, COM701, fully owned and the only Fc-reduced monocolonal -- antibody in the clinic, which again, we believe is the right Fc-format. The biology here is truly differentiated from PD-1 and TIGIT checkpoints, providing advantages that we believe could translate into clinical benefit for patients with platinum-sensitive ovarian cancer.

Positive data in our ongoing MIOviral platform trial could support a product like development program aimed at addressing a significant unmet medical need. And finally, to our KulanSmalt potential first-in-class antibody program, adversing cytokine biology, GS-0321 previously COM503 is a potential first-in-class anti-icinbinding protein antibody licensed to Gilead. GS-0321 represents a novel antibody approach to harness -- biology for the treatment of cancer, potentially overcoming the limitations presented by illustration of therapeutic cytokines. It represents another potential value driver for Compugen, as we're eligible to receive $758 million in milestone payments and single-digit to low double-digit tiered royalties. This program is the most recent example of how our AI/ML power discovery engine is delivering new opportunities.

And behind this, we have early pipeline of what we believe to be truly innovative research programs. As pioneers in the field, we are committed to delivering real breakthroughs, not just incremental therapies. And really innovation is never easy. It takes time, persistence and willingness to tackle the toughest scientific challenges. But I believe deeply in what we're doing here, and we have the best talent and great tools to do this and are truly excited about the potential of our early stage programs.

Next, turning to the progress we have made this quarter. The team at ESMO in Berlin in October represented the full analysis of our 3 previously reported Phase I trials, reflecting the clinical benefit of COM701 as monotherapy and in combination in patients with heavily pretreated platinum-resistant ovarian cancer. The pooled analysis demonstrated that COM701 was well tolerated, showed consistent durable responses in patients with heavily pretreated platinum-resistant ovarian cancer, particularly in those without liver metastasis, representing patients with lower disease burden and potentially less immunosuppressive tumor macroenvironment. The results of the analysis support the rationale for the ongoing randomized myoviral platform trial, evaluating COM71 its maintenance therapy in early lines of treatment.

The Myoviral platform trial is progressing -- sites have been activated across the U.S., Israel and France including major academical centers and multiple sites from the French oncology cooperative group, AkaciGenico, renowned for several recent platinum-sensitive ovarian cancer trials. We now estimate the interim analysis in Q1 2027. We believe Myovian is a significant opportunity to address an unmet need for maintenance therapy in platinum-sensitive ovarian cancer. Next, our partner, AstraZeneca, which presented new real bad data at ESMO as part of 2 mini-oral sessions. AutoMD 1 follow-up shows that river was well tolerated with promising efficacy, confirming its potential in checkpoint-naive non-small cell lung cancer.

Notably, the drug-related discontinuation rate of only 3% further support differentiation of dept-reduced formats. The Tropin pan tumor 3 evaluating combination with -- showed promising efficacy and manageable safety underscoring the potential of next-generation IO bispecific plus ABC. Moving next to GS-0321, our novel antibody approach with Gilead that leverages cytokine biology, the Phase I trial is progressing as planned and represented a trial design at CIT last week. We have strong conviction in our fully owned programs. We are validating partnerships with AstraZeneca and Gilead providing potential for over $1 billion in milestones plus royalties.

Of course, none of this would be possible without our highly committed talented team here at Compugen who continuously performs at the highest level of excellence. With that, I will hand over to David for the financial update before we open up the floor for Q&A.

Thanks, Eran. I'm pleased to say that we are advancing in 2025 with a solid balance sheet. Cash runway, assuming no further cash inflows is expected to fund our operating plan into the third quarter of 2017 and we anticipate using this runway to advance our COM701 platinum-sensitive ovarian cancer trial, myaovarian and to support the progression of ZS-0321 in the clinic, together with continued investment in our early-stage pipeline. Going into the details, I will start with our cash balance. As of September 30, 2025, we had approximately $86 million in cash, cash equivalents, short-term bank deposits and investments in marketable securities. In October 2025, subsequent to the financial results for the quarter ended September 30, 2025, a total of approximately 0.8 million shares were sold through the company's ATM facility, contributing to a net proceeds of approximately $1.6 million.

Revenues for the third quarter of 2025 were approximately $1.9 million compared to approximately $17.1 million revenue for the comparable period in 2024. The revenues for the third quarters of 2025 and 2024 reflect the recognition of respective portions of both the upfront payment and the IND milestone payment from the license agreement with Gilead. Expenses for the third quarter of 2025 were in line with our plans. R&D expenses for the third quarter of 2025 were approximately $5.8 million compared to approximately $6.3 million in the third quarter of 2024. Our G&A expenses for the third quarter of 2025 were approximately $2.2 million and approximately $2.6 million for the same period in 2024.

For the third quarter of 2025, our net loss was approximately $6.98 million or $0.07 per basic and diluted share compared to a net profit of approximately $1.28 million or $0.01 per basic and diluted share in the third quarter of 2024. With that, I will hand over to the operator to open the call for questions.

[Operator Instructions] The first question is from Stephen Willey of Stifel.

Just curious, the extension of the Maya interim analysis from, I guess, the second half of 2016 into the first quarter of 2017. Is that just predicated on enrollment time lines and kind of what you're seeing just from an accrual perspective. Does that have anything to do with the accumulation of PFS events in the study that may be required to trigger the interim, just curious as to what's kind of happening behind the scenes there?

Sure. Thanks, Steve. So overall, as we know, there are a few factors that determine the initial readouts of clinical trials. This opening the sites is the actual enrollment rates. And finally, the actual accumulation of events along the trial. And we saw the -- estimation and the more the trial develops, you understand the kinetics and you optimize your prediction, and this is exactly what we're doing here. We can say today that we opened most of the sites, including major academical U.S. centers and the French Clinical Group. It took a bit more time to open it's mostly academical centers, but we are glad to have them on board. And now again, most of them are open.

And now we expect while opening the sites, and again, and heading the French site also was done because they showed interest and also to support the aggressive enrollment trade that we anticipate. And now this is the time for the start enrollment and see the actual ramp-up. And Michelle, if you want to add something to add some color?

No. I mean, you covered everything. So effectively, we had selected a number of sites. We've had additional academic sites wanting to participate. And those do tend to take a little longer to open. In addition to that, we had also been consulting with Geneco in France when they asked to participate as well. So we're trying to reflect our best estimates. I think there's a lot of different factors that impact when 1 has an interim analysis, and we still believe that we will be able to meet the aggressive time lines that we have.

And finally, we also disclosed today that we have cash runway into Q3 '27. So we also have the cash to support taking into account this shift in Q1. So through an good position to continue with the trial and to bring value profession because we think -- we believe in this study. .

The next question is from Daina Graybosch of Leerink Partners. Please go ahead.

I wonder if we could talk about the upcoming ARCUS Gilead readout with -- and gastric cancer because it could come as early next year. What do you know what you're looking for, of course, if it's successful, then that validates your ingoing hypothesis. But is there anything you would see in the outcomes of that trial that would reduce your confidence in your own TIGIT and more importantly, in the bispecific Rovigo.

Thank you, Daina. It's a very good question. So this will be the first Phase III readout for a reduced digit antibody. The data enabled is promising, but it was a single-arm study, not many patients, but doubling -- almost doubling the overall survival versus -- control was reassuring. And now is the time to see how it evolves in the Phase III, and this is, of course, could be very meaningful for us. But it is only 1 trial. So obviously, if it's successful, this reflects directly on the Fc-reduced and what we're saying about the Fc-active the safety issues, potential reduced efficacy issues, and this will show directly that Fc reduced are active in Phase III results. But even if this trial fails, ARPU deal themselves as additional Phase II Phase III trials, additional 2 ones. And AstraZeneca has 2 additional advantages over the -- just a single monoclonal Fc-reduced. One is they show that they're bispecific has a potential more activity. And actually, it shows up in a very nice ex vivo patient-derived material system, and they have a cooperative binding that allow cooperative blockade of PD-1 and TIGIT on the same cell.

And the result was in that relatively trastational system that is more active than just PG1- TIGIT blockade. And then also in some of the trials, that a potential regulatory advantage is the way we see it and looking just for the size of all the accumulating Phase III trials, for example, run of the the trials in the -- I believe the -- BLRA01 that comparing real plus chemo to -- and because it's a bispecific, nobody can ask for a contribution of components as far as we understand it. And therefore, you don't need to show that TIGIT is active. Yes, we believe TIGIT is active, and they have to reduce. But even if the activity is not sufficient in this case, having both enhanced efficacy due to the bispecific format and just using it -- as a safe backbone to combine with chemo to compare versus chemo. This is definitely an advantage of the bispecific. There are some other trials doing the same. And they also have some trials doing directly head-to-head versus pembro.

And we believe and think that they should have a win there as well. But so they have multiple shots on goal with some advantages of the bispecific in this.

The next question is from Lena Gershell of Oppenheimer.

Just wondering if you -- as we look forward to the interim update from my ovarian, could you remind us of any internal threshold or bar you're looking for from that interim with respect to efficacy.

Thank you, Leland. So I will turn the hand over to Michelle so again, just to remind everyone, we talked about a study which have 40 patients treated with COM701 in maintenance settings compared to 20 patients in placebo. We're relying on solid -- control -- and internal control compared to placebo. This is not a registration trial, but we are looking, and we think this trial is built to allow us to understand if COM701 has a monotherapy signal in this patient population after seeing a signal in the last-line lung resistance setting. And then upon success, this adaptive trial design will allow us to build and to continue to move forward either to adding more arms or to potential -- approval. And Michelle, please add some color on that.

Okay. So the clinical trial is an exploratory study to allow us to determine the magnitude of the effect size of COM701. It is very desirable for us to be able to demonstrate single agent activity, an improvement of up to 3 months above the placebo would be very clinically meaningful. But at the same time, we are looking forward to the totality of the data to be able to determine what the next best steps would be.

The next question is from Asthika Goonewardene of Truist Securities.

So with COM902, this would test to be an unpartnered actually reduced TIGIT antibody in the way new data coming up from the people who were watching on Arcus and making read-throughs here. So I know you've licensed to bind at your AstraZeneca, but does that still give you flexibility to partner in IoT with a separate company. Can you let us know about any restrictions or financial terms we should take into consideration?

Thanks, Asthika. It's a great question. So we licensed AstraZeneca, the rights to use COM902 as part of their bispecific PD-1 and TIGIT and some other bispecifics. But we fully own covenant. We don't have any restrictions. We can either decide to move forward in our own trials, obviously, upon successful results by others to be a meaningful driver, remind love that in the days that the active TIGIT initial deals were hundreds of millions of dollars of upfront deals and being the probably only monoclonal Fc-reduced-digit antibody other. We believe that redos in 26 could bring meaningful interest back into TIGIT, especially in COM902 and we have -- again, we fully own it. So it's we can be fully opportunistic in whatever direction we would like to take this COM902. .

The next question is from [indiscernible] H.C. Wainwright.

Good morning, Eran and David, thanks for taking questions. A couple of quick questions. One is when we you look at the tolerability profile of comps alone. How does that influence its potential use in combination therapies, especially in some of the less immune influent tumors. And the other question is, when you saw the data from the pooled analysis presented at ESMO. There were some great 3 or higher treatment-related adverse ones about 16.7% results. So how does -- how do you plan to include that safety in combination therapies. .

Sure. So firstly, the tolerability of COM701 as a monotherapy is extremely well tolerated. In fact, we didn't have any discontinuations due to adverse events in that pooled analysis when COM701 was used as a single agent. And in the triplet combination groups in the pooled analysis, the adverse events that we're seeing that were grade 3 were in keeping with the same frequency seen in respect of labels for both nivolumab and pembrolizumab. Given the tolerability of COM701, we believe that it is very well set up for being able to be used as a monotherapy or as a combination with standard of care agents or with other novel agents that are coming through the landscape.

This concludes the Q&A session in Compugen's investors conference call. Thank you for your participation. You may go ahead and disconnect.