Compugen Ltd (CGEN) 2021 Q3 法說會逐字稿

Ladies and gentlemen, thank you for joining us today. Welcome to Compugen's Third Quarter 2021 Results Conference call. At this time, all participants are in a listen-only mode. 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, Head of Investor Relations and Corporate Communications. Yvonne, please go ahead.

Thank you, operator, and thank you all for joining us on the call today. Today's call is being recorded and will be available on the Investor Relations Section of our website. Joining me to present our prepared remarks are Anat Cohen-Dayag, President and CEO; Henry Adewoye, Chief Medical Officer; and Ari Krashin, Chief Financial and Operating Officer.

For the Q&A session, we will also be joined by Eran Ophir, Vice President, Research and Drug Discovery.

Moving to Slide 2. Before we begin, I would like to remind you that during this call, the company may make projections or forward-looking statements regarding future events, our business outlook, our development efforts and their outcome, our discovery platform, anticipated progress, results and time lines for our programs, financial and accounting related matters as well as statements regarding our cash position.

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 the known and unknown risks and uncertainties, and we refer you to the SEC filings for more details on these risks, including the company's most recent annual report on Form 20-F filed on February 25, 2021. The company undertakes no obligation to update projections or forward-looking statements in the future.

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

Slide 3, please. Thank you, Yvonne. Good morning, and good afternoon, everyone, and welcome to our third quarter 2021 update. Today is an exciting day for Compugen with the announcement yesterday of the expansion of our strategic collaboration with BMS, along with $20 million equity investment. And today, with the release of our 3 presentation at SITC, providing clinical, translational and preclinical data that demonstrates strong execution across our differentiated clinical and scientific strategy and the further support to our path forward.

Indeed, to our knowledge, this is the first data presentation ever of a triple blockade of PVRIG, TIGIT and PD-1, and the first time data on anti-tumor activity of an Fc reduced effector function anti-TIGIT antibody is presented in a scientific conference. These dose escalation studies cleared our path to initiate multiple extension cohort studies that are designed to broadly evaluate the inhibition of the DNAM axis in cancer immunotherapy and advance that are putting development of our 2 leading programs, COM701 and COM 902.

Henry will review the data from our dose escalation studies in the call. But before that, I'd like to share a high-level overview of the 4 key pillars of our long-term strategy and an overview of our clinical strategy. Moving to Slide 4. Our first pillar is to advance the field of immuno-oncology research, with the ultimate goal of developing new therapies. We have an impressive track record of driving the discovery of new targets and pathways in the immuno-oncology space with cutting-edge science published in peer-reviewed papers and development of a clinical stage pipeline derived from this science.

Our clinically validated computational target discovery platform has generated a pipeline of exciting new candidates that are being evaluated in Phase I clinical trials by us and our partners, including our potentially first-in-class anti PVRIG antibody, COM701, our high affinity IgG4 Fc reduced effector function, anti-TIGIT antibody, COM902. bapotulimab, and anti-ILDR2 antibody, which is being developed by Bayer and AstraZeneca's TIGIT bispecific antibody program derived from COM902. I'm proud of our talented team of scientists with expertise in immuno-oncology and translational medicine and with our capabilities of driving the science of new unexplored biological pathways forward as we develop first-in-class therapeutics targeting these pathways.

This takes us to our second pillar, which is to expand the number of patients responding to treatment by developing potentially first-in-class therapies for patients nonresponsive or refractory to currently available therapies. Specifically, the unique biology we unlocked for the PVRIG pathway following its discovery served as the basis for our assessment of the different combinations of PVRIG, TIGIT and PD-1 inhibitors in select biomarker informed tumor types generally not responsive to checkpoint inhibitors.

What makes us differentiated beyond the discovery of this pathway and uncovering its potential role within the DNAM axis, is that we are the only company evaluating combination of PVRIG, TIGIT, and PD-1 blockers in the clinic. It is exciting to be the leaders in the DNAM axis especially considering the recent advancements and raising interest by followers in this area.

Maximizing the value for patients is our third pillar and speaks to our goal of bringing new treatments to market as rapidly and efficiently as possible. To support us on our journey, we seek collaborations and strategic partnerships with leading biopharma companies with complementary capabilities worldwide to expedite our early and clinical stage programs and bring them to the market. In addition to our partnership with BMS, our partnership with AstraZeneca and Bayer also highlights the value of our strategy to broaden the opportunities of our products to reach more patients.

And finally, our fourth pillar is to protect our innovation while developing drugs for patients through securing IP rights. We have developed a strong IT portfolio with potentially broad protection for drug candidates.

Now moving to Slide 5. We have employed a comprehensive clinical strategy to assess our DNAM axis hypothesis. We designed multiple combination studies evaluating the simultaneous blockade of the 3 pathways and subsets, all are running now in parallel in multiple tumor types selected in a biomarker informed manner.

Each of these studies include comprehensive translational data analytics in order to further support our drug's mechanism of action and discover and evaluate potential biomarkers that may be of relevance for future patient selection. Our multiple combination studies are being pursued in some overlapping indications. This is by itself will allow us to evaluate the contribution of each drug in this combination, and focusing mainly on nonresponsive tumors should help to differentiate the contribution of COM701.

The data from these multiple studies will allow us to select the combinations and tumor types to progress our program.

Moving now to Slide 6. Delighted to say that we delivered on all our targeted milestones for this quarter and for the year. 2021 played a major role in further solidifying our leadership position in the DNAM axis with steady and impressive execution. We've completed all our dose-escalation studies, including our COM701 monotherapy, COM701 plus nivolumab, COM902 monotherapy and triple combination studies. We've now progressed to expansion cohorts across all our programs, which allows us to focus on select indications through our translational data-driven approach.

Our accomplishments through 2021 also reinforced the power of our partnering strategy in diversifying our portfolio, with a milestone payment from AstraZeneca after the first patient dose in their Phase I study of a TIGIT bispecific derived from our COM902. In addition, we were thrilled to announce the expansion of our collaboration with BMS and believe the recent $20 million strategic equity investment in Compugen strengthens our relationship and the goal of both companies to bring innovative therapies to cancer patients through the advancement of our clinical studies conducted under our collaboration.

As part of the expansion of the collaboration, a joint steering committee has been formed to facilitate strategic oversight and guidance for the program fund under the collaboration. This committee will run alongside the existing joint development committee, which acts at an operational level.

With that, I'd like to turn the call over to Henry for him to provide an overview of our data released at SITC today. Henry?

Thank you, Anat. If we move to Slide 7, please. I am pleased to provide you with an overview of the preliminary results from our ongoing Phase I/II study evaluating the combination of COM701, our potentially first-in-class anti-PVRIG antibody with nivolumab and BMS' anti-TIGIT antibody, BMS-986207 released today at SITC.

The study enrolled 13 DLT evaluable patients with a variety of advanced solid tumors. These allcomers who have exhausted all available standard treatments and we're heavily pretreated with a median of 10 prior therapies and a maximum of 19. The study evaluated escalating doses of COM701 in combination with fixed doses of nivolumab and BMS' anti-TIGIT antibody.

The primary objective of this study is to evaluate the safety and tolerability of the combination. The study reported no dose-limiting toxicities, and the combination had a favorable safety profile and was well tolerated.

Reaching this important milestone shares the path for a comprehensive evaluation of competent DNAM axis hypothesis of triple blockade of PVRIG TIGIT and PD-1 pathways in select biomarker-informed indications. Additionally, COM701 20 milligrams per kilogram IV Q4 weeks was identified as the recommended dose for expansion. The expansion cohorts have been initiated and enrolling patients with platinum-resistant ovarian cancer, endometrial cancer and head and neck squamous cell cancer.

We also reported a preliminary antitumor activity and supported stable disease in 3 patients, with 1 patient with prostate cancer remaining on study beyond 100 days of treatment.

Now moving to Slide 8. We and our partner, BMS, found the translational data important and noteworthy, indicating pharmacodynamic activation of the immune system following treatment across key measures, including interferon gamma induction, increased T and NK cell activation and memory T cell proliferation at all doses of COM701. These results are significant for several reasons. They are consistent across measures and patients and the magnitude of the effect was above what has been observed in monotherapy and drug combination settings of COM701 blockade as well as published results for other TIGIT antibodies, mono and combination studies.

This supports a potent activation of the immune system following triple blockade of PVRIG, TIGIT and PD-1, and aligns with our extensive preclinical work predicting this effect. It is also important to highlight that immune activation was achieved in our studies in historically cold tumor types, including ovarian, colorectal and prostate cancer. Together these results further support Compugen's earlier funding for the potential of triple blockade to drive synergistic immune activation, and we look forward to the continued evaluation of our hypothesis in the ongoing biomarker informed expansion cohorts.

In staying with the theme of translational data, I will take a minute here to mention some of the preliminary findings that will be presented here at SITC in our research poster. Following up on our observation, the PVRIG is a potential, unique and dominant checkpoint involved in stem like memory T cells, dendritic cell interaction, we evaluate axiom from the 2 patients who responded to the combination of COM701 and nivolumab. Interestingly, we saw an increase in markers of dendritic cell activation in these patients when compared to nonresponding patients. This may suggest that this clinical response involves enhanced dendritic cell, T cell interaction triggered by COM701.

It is exciting to see that our research relating to PVRIG potential mechanism of action translates to preliminary findings in the clinic and further support the differentiation of PVRIG from TIGIT and PD-1. This also suggests a differentiated profile for COM701, having a potential to address both inflamed and less-inflamed tumor types where current checkpoint inhibitors have not been successful and further explains why one might expect a more potent immune activation when combining COM701 with the other inhibitors.

Moving next to Slide 9 and to the SITC data from our Phase I dose escalation study evaluating a high-affinity antithesis antibody, COM902, with reduced Fc effector function. The study enrolled 18 patients with advanced solid tumors that is all comers who exhausted all available standard therapies and were heavily pretreated with a median of 7 prior therapies and maximum of 16. The primary objectives of this COM902 monotherapy dose escalation study was to establish safety and tolerability, the recommended dose for expansion, and the PK profile of COM902.

Overall, COM902 demonstrated a favorable safety and tolerability profile. In the study, we reported 2 patients with dose-limiting toxicities, 1 patient in the single subject dose cohort with grade 2 vomiting at 0.01 milligram per kilogram dose, and 1 patient with grade 3 atrial fibrillation at the 1 milligram per kilogram dose. These DLTs were based on assessments by the investigator. It is important to note that there were no DLTs reported at any other doses, including the higher doses up to 10 milligrams per kilogram and a maximum tolerated dose was not reached.

Of note, in the COM902 repeat dose preclinical study, no ECG changes were noted. The COM902 3 milligrams per kilogram IV Q3 week was selected as the recommended dose for expansion. Additionally, the PK of COM902 was dose proportional and supports Q3 week dosing. We were encouraged by preliminary antitumor activity with 50% of patients or 9 of 18 achieving a best response of stable disease, with 6 patients, 67%, with confirmed stable disease; and 3 patients, 17%, with stable disease of at least 6 months duration or longer.

And while we consider TIGIT to be a combination therapy, achieving a business control rate of 50% in this heavily pretreated patient population is encouraging.

These preliminary results are particularly significant for Compugen and the broader TIGIT space. As you may know, there is an ongoing debate regarding the role of the Fc domain and its relevance for antitumor activity with TIGIT inhibitors. Our approach has always been to have reduced Fc effector function anti-TIGIT antibody as we believe this reduces the risk of depleting CD8+ T cells, which are crucial for driving antitumor activity. This signals of preliminary antitumor activity with COM902 monotherapy are suggesting that our reduced Fc approach is an appropriate strategy and this is supported by the translational data, which I show on the next slide.

Now moving to my last slide, Slide 10. We see translational data from the study that further explored this. Flow cytometry analysis of peripheral blood from patients treated with COM902 shows no significant changes in PD positive CD4 and CD8 T cells and NK cells. This is important as these results reinforce that our reduced Fc approach avoids the depletion of major PD-positive lymphocytes, which are critical for antitumor immunity. And as I indicated earlier, support our rationale for choosing an IDD4 Fc-reduced effector function anti-TIGIT antibody.

Before I turn the call over to Anat, I'd like to thank the patients and their families, investigators and study sites. Anat?

Thank you, Henry. Slide 11, please. So far, 2021 has been an eventful year for Compugen with strong execution across our differentiated clinical strategy which has solidified our position as leaders in the DNAM axis.

As the only company targeting PVRIG, TIGIT, and PD-1 in the clinic, while maintaining our first-mover advantage for PVRIG inhibition in general and for the free pathway hypothesis in particular, the strength of this hypothesis has grown through new translational data which supports potent immune activation, now also through triple blockade which is consistent across tumor types and immune measures. We've also strengthened our deep expertise in PVRIG biology with preclinical and translational data, which supports the differentiated profile of PVRIG as an immune checkpoint, with exciting data showing increase in activated DC markers in term of 2 patients that clinically responded to COM701 plus nivolumab treatment.

This new preliminary observations suggests that a unique dendritic cells T-cell interaction triggered by COM701 may be driving responses in patients and ultimately may be capable of addressing both inflamed and less inflamed tumors. On the TIGIT front, our ultimate strategy is in combination settings and our position is unique in evaluating PVRIG and TIGIT dual blockade in the clinic with our ongoing COM701, COM902 combination study.

We were pleased to present translational data that support our rationale in choosing an anti-TIGIT antibody with reduced Fc effector function with results demonstrating that COM902 avoids depletion of major TIGIT positive lymphocytes, including CD4 and CD8 T cells as well as NK cells, while producing antitumor activity on par with other TIGIT assets. We continue to be encouraged by the broad and growing interest in the DNAM space and our free pathway hypothesis, while also maintaining our first-mover advantage and strong IP protection.

I'm proud of our execution with multiple key data readouts and the initiation of multiple expansion cohort studies supported by translational data and with multiple data redouts ahead, which will enable us to take our program forward for the potential benefit of patients. I'm incredibly excited about what's to come and I am grateful for the amazing team at Compugen whose dedication has enabled our remarkable accomplishment.

I'll now turn the call over to Ari to review our financials. Ari?

Thank you, Anat. So moving to Slide 12. Our financial results for the third quarter of 2021 released this morning continue to show our strong financial position as we execute across our expanding clinical programs. The $6 million in revenue this quarter related to the AstraZeneca milestone triggered by the dosing of the first patient in AstraZeneca's Phase I/II study of a TIGIT bispecific derived from our COM902.

Cost of revenues of $0.7 million are attributed mainly to royalty and milestone payments. Research and development expenses for the third quarter of 2021 were $8.7 million compared with $5.5 million for the same period in 2020. The increase reflects the expansion and initiation of additional clinical studies during 2021 as well as increased drug manufacturing activities. Net loss for the third quarter of 2021 was $6.2 million or $0.07 per basic and diluted share compared with a net loss of $7.8 million or $0.09 per basic and diluted share for the same period in 2020.

As of September 30, 2021, we had approximately $102 million in cash and cash-related accounts compared with approximately $111 million as of June 30, 2021.

With the recent $20 million equity investment by Bristol-Myers Squibb as well as the collection of the $6 million payment from AstraZeneca expected in the fourth quarter, our year-end cash balance is expected to be in the range of $130 million to $116 million. The company has no debt.

Thank you. And with that, we will now open the call for questions.

(Operator Instructions) The first question is from Mark Breidenbach of Oppenheimer.

I guess I wanted to start with 902, actually, and the dose you've chosen to advance into expansion cohorts. It seems like it's a little bit below what we're seeing used for some of the other TIGIT antibodies, even the fixed dose that the BMS-986207 is being used in your triple study. Could you maybe just comment on why you're confident that you've sufficiently dose-escalated and you're confident around this 3 mg per kg dose going forward and kind of the reasons why 902 might be maybe more potent than some of the other TIGIT antibodies that have advanced into late-stage trial space?

Sure. Thank you, Mark. Henry, would you like to start and maybe Eran will have any additions later?

Sure. I can start. Thank you, Mark, for the question. So we considered several factors when we selected -- we and the investigators selected the dose that we recommended for expansion, which is 3 milligrams per kilogram of body weight IV Q 3 weeks. So one of the things we considered was receptor occupancy, peripheral receptor occupancy. As you know, COM902 is a high-affinity TIGIT antibody. So we observed and we've reported in the poster that we achieved peripheral receptor occupancy at 0.1 milligrams per kilogram body weight dose.

We also escalated COM902 up through 10 milligrams per kilogram body weight dose. And the overall recommendation from review of the data included in the PK data, was that we didn't think, in addition with the investigators and the authors on the study, that going beyond 3 milligrams per kilogram body weight dose was going to add any more clinical benefit in terms of the outcomes for the patients. So that's the reason we selected 3 milligrams per kilogram body weight dose.

Now when you benchmark that dose and you compare it to what the peripheral receptor occupancy is, that's at least a margin of 30. That's significant enough to have tissue penetration in the tumor microenvironment. I wouldn't want to comment on the metrics for other companies. But I think if you go back and look at the way they've demonstrated their receptor occupancy and selected their dose, it's very similar across companies. So we're not -- we're actually in the mainstream.

Yes. Maybe just to add to it. Indeed, we compared COM902 to the other leading TIGIT benchmarks. And indeed, COM902 has better binding, better blocking and at least as good or better functional activity in vitro. Regulators translated also into achieving full receptor occupancy in a relatively low level compared to the other published data. And yes, as Henry mentioned, we believe that these dose that we choose will be sufficient to allow full occupancy also in the tumor environment.

Okay. And just with respect to that one Grade 3 atrial fib, is this toxicity that's been seen in other studies of TIGIT antibodies? Is there any chance in your mind that this is a class effect? Or do you think this is a one-off?

Yes. So Mark, it's a good question. We think it's a one-off for several reasons. So we'll come back to look at the preclinical data. We have repeat dose toxicity studies that were conducted prior to initiating the human experiments, the human clinical trades that we have. And the repeat dose toxicity studies did not demonstrate any EKG findings, as I noted in my prepared comments. In addition, we didn't actually see -- we didn't see any additional toxicities when we escalated through 10 milligrams per kilogram body weight dose, like I said in my prepared comments.

Now that particular patient, I can give a little bit of more information. So the patient came in had -- was diagnosed with arterial fibrillation and that resolved on the same day that the patient came in. This was about a week through the first dosing interval in the first cycle. So we think it's a one-off effect. Now we do not know if the other companies that have TIGIT inhibitors report all the data. So you can imagine that sometimes the data that is reported might be limited to maybe 3 patients or 2 patients and the single-digit toxicities might not be demonstrated. But we don't think, based on preclinical data, that this has any bearing or any relationship to COM902. That's our assessment.

The next question is from Stephen Willey of Stifel.

Maybe to follow up on COM902. I think the lack of lymphocyte depletion of TIGIT-positive self is interesting. But can you remind us what is the magnitude of decrease that you would expect to see in these cell subtypes with those TIGIT antibodies that have been engineered to have enhanced effector function?

Yes. So some of the companies did publish these results and what they have shown is dramatic reduction. It's not like a minor reduction, it is in peripheral blood. As we all know, in tumor microenvironment, the depletion normally is much more challenging. So we don't know if there is any kind of depletion to tumor environment. In peripheral blood, they have shown a depletion of CD8 positive, TIGIT-positive cells in the peripheral blood, and that the reduction was dramatic.

Okay. That's helpful. And then on the triple combo biomarker data that you have, again, it did -- I guess it appears to kind of support the hypothesis that PVRIG is contributing something. But another way that the data is presented it's a function of all doses. Is there anything that you can say with respect to there potentially being a dose dependency to some of these changes in biomarkers that you're seeing as a function of escalating COM701?

Okay. So in this trial in the TIGIT study, the dose response range was not as big as we had in the past. So actually, we started in one patient in '03 and then already moved to 1 mg per kg, which is already for receptor occupancy. So in this specific context, we haven't seen much of a dose response. I remind you that in the past and we had a wide range of responses, concentrations. We did see a dose response for COM701 in combination with nivo, et cetera.

Okay. And then maybe you can just lastly talk about, I guess, how the expanded Bristol collaboration, I guess, intersects with the fact that you're now through dose escalation with COM902, you're going to be interrogating various combinations with that agent. How do those 2 things overlap with each other, if at all? And are you guys -- I guess, are the guardrails in place embedded within the collaboration, such that you kind of each stay in your own way?

Yes. Related to that, Steve. So this collaboration actually -- this expansion of the collaboration reflects on the long-term long-standing relationship between the 2 companies. It stands for itself. It is really designed in order to enhance the strategic collaboration, it is enhanced -- it is designed to support the current studies that we have to support Compugen in executing these studies.

At this point in time, we are on the strategy front -- on the clinical strategy front, we are at the stage that we completed all the dose escalation studies that we wanted to do. And we already started the expansion cohort. And remember that we have with Bristol now, 2 combination studies, the COM701 nivolumab and the triplet, and each of this study is conducted in multiple tumor types. We have a broad translational program that goes with the PD markers, immunohistochemistry. As we've stated in the past, also, we are applying advance technologies on patient centers.

So the collaboration is actually growing and growing, and both parties felt that it will be good to have a joint strategic oversight. There is a committee that is actually working in parallel to the joint development committee that is working at the operational level. So at this point in time, having all the studies being expanded, that's the right time for us to align on the strategy of how we're interpreting the data together and how we think about next step.

The next question is from Daina Graybosch of SVB.

Congratulations on the data, guys. I wonder, first, on COM902. If you could give us any more details about the nature of these stable diseases. Did you see tumor shrinkage? Was there not so much shrinkage and a lot of stable? And any notable case studies that gave you a lot of confidence.

Yes, thank you very much for the question, Daina. So we did look at the metrics for the patients that were enrolled. We, with the investigators, considered that we had enough data embedded in the poster to inform on the clinical profile of the subjects -- of the patients that were enrolled on this study. So the things that we were focused on, essentially, we're looking at the treatment journey or the course of the patients, so which is illustrated in this swimmer plot. We think that the swimmer plot best represents how heavily pretreated those patients were.

We also provided icons in the swimmer plot to illustrate the patients who had received prior immune checkpoints and patients who had prior treatment refractory disease. So one of the other things that we looked at as based on enrollment on the study was that for patients who are in dose escalation, and it's only dose escalation that we're reporting, the assessment of measurable disease is not an eligibility criterion. So we didn't think that would inform on whether patients had tumor reductions.

Yes, patients could be evaluated based on resist, but we thought that the most appropriate way to illustrate the antitumor activity of COM902 was to demonstrate this by the swimmer plot. Yes, we did see a few patients with tumor regressions, but if you look carefully at the swimmer plot, I think it best answers the question of how patients were doing on the study, and that's why we decided to just go with this swimmer plot.

So a follow-up then. So how many of these patients -- or do you know -- I just -- I don't see the notation, which of these patients started with measurable disease and which ones did not?

Yes. We decided not to include that because since it was not an eligibility criteria, not a lot of the patients had measurable disease, and we wanted to convey the finding of the antitumor activity, including confirmed stable disease that we demonstrated in the swimmer plot. So if you see there on that swimmer plot, we have of the 18 patients, 9 of them with either best response assessment of stable disease. And out of these 9, 2/3, 67% with confirmed stable disease, and of course, 3 patients with stable disease that's 6 months or longer. The icons that we illustrate in the swimmer plot, I think provides a surrogate assessment for how COM902 as monotherapy through dose escalation has some form of clinical benefit for the patients that are enrolled. So I think it probably -- if you think very carefully about it, it provides you with enough information on how patients are doing on this study.

Okay. Maybe one more follow-up. For the patients with prior immune checkpoint inhibitor to the -- do they follow, let's say, the SITC guidelines for washout between the periods, the confirmed progression on the checkpoint?

Yes. We had certain benchmarks in the protocol with regards to eligibility for enrollment onto the study. So of all the patients that we had, out of those 18 patients we enrolled, 8 of them had, had received prior immune checkpoints. And if you look at the swimmer plot, it best illustrates those patients, and I can just go through them. So we had a patient with chordoma, adenoid cystic carcinoma, peritoneal cancer, small cell lung cancer, mesothelioma, prostate cancer, renal cell carcinoma and prostate cancer. Yes, we follow the guidelines in terms of washout.

The next question is from Chris Howerton of Jefferies.

Congratulations on the progress . I think a lot of really good ones have already been asked. But maybe just 2 from me, One would be could you comment at all on any safety that you've seen with respect to the triple combination study? Any notable toxicities or things that would suggest things like immune activation or things like CRS, would be one question.

And then the second question that I would have is this might be a little tougher to kind of think about. But one of the things that's notable about the patient populations that we have here is that they're heavily pretreated by 10, median line of 10 previous therapies. So how can we think about the potency certainly of 902 in the context of patients with a healthier immune system? And what are some of the things that you're going to be looking out for as you move forward in that context?

So maybe I can start. So in terms of cytokine release syndrome, we didn't observe this. None was reported on the study for the triplet and also for COM902. This was not something preclinically that we thought we will see. And so far, with the reports that we have, we haven't reported this neither have the investigators, even though they ask these questions and do the lab work on the patient, reported or observed this also.

We think that based on the mechanism of action of COM902 and also as part of the triplet that the toxicities that we've observed, toxicities that are not unique either to COM902 or the toxicities that are unique and start of the triplet, you know that we have accumulated a lot of data on COM701.

So let me speak about the triplet first, on COM701 as monotherapy, and we can say categorically that it is very well tolerated as monotherapy, and it has a favorable safety profile. We also have enough information at this time with regards to COM701 in combination with nivolumab, which we disclosed at ASCO. It is also well tolerated and has a very good safety profile also.

We did not observe any of the toxicities that you've mentioned. The most frequent toxicity we've observed as monotherapy for COM701 in combination also with nivolumab and also as part of the triplet is fatigue. And it typically grade 1 or 2 fatigue that we've been able to show. The other toxicities that we've seen in combination are things that are constitutional in nature and related to the cancer that the subjects have, either they have elevated liver function test abnormalities as a result of progression in the liver or they have obstruction of certain organ components because of the prevailing cancer itself.

So there's nothing that is unusual in the studies that we've seen so far. For COM902, specifically, it's very well tolerated also. Mark asked about the patient with atrial fibrillation. We think it's a one-off based on the preclinical data that we have. We didn't observe cytokine release syndrome in that. Also, we did note that in addition to most of the patients doing well on study treatment, only one subject came off, and that's the subject that had the treatment discontinuation that we described. That's the patient with prostate cancer with atrial fibrillation that I mentioned. So I'll stop there.

The next question is from Tony Butler of ROTH Capital.

Yes. The question or series of questions. One is in the triple study, a small number of patients, but what is considered a high expresser of PVRL2? That's question one. And then question 2, 2 parts. Going back to the translational data, which I have a lot of respect for, a question was actually asked about dose response. But I am curious about the pharmacokinetics of 701 and especially given the low dose versus high dose. And importantly, is there any correlation to dose level and duration? That's part a.

And then b, do you have a hypothesis as to why if, in fact -- and I realize that the translational data is all peripheral, we don't really know what goes on at the tumor site unless you have a view about that. But what's really -- why don't we see greater tumor shrinkage? And it may be simply because these are very, very sick patients who've had multiple therapies, I'm respectful of that. I'm just curious of your hypothesis on that point.

Okay. I think I can take it. So first of all, for the high expresser of PVRL2. So as you may remember, we published quite extensively in the past that certain tumor types have higher expression of PVRIG and PVRL2, and these are the indications of which we are moving forward to now. So just in the case of PVRL2, in terms of age score, if you wish, it's probably 200, 300. So this was the first one. Then for the dose and duration, Henry, probably you want to take this one, and I can take the third one.

Yes. So you're correct that the duration is a reflection of the heavily -- heavy pretreatments that the patients are on the studies were on. Just as a matter of fact, we did indicate that in our prepared comments and also on the poster, that a median of 10 prior therapies with a range of 1 through 19.

If you think of the COM701 nivo study, this was a median of 5. So it's actually essentially a doubling of the median prior therapies. So if patients are unable to stay as long, then you probably will not be able to see this correlation that is of interest to you and that is also of interest to us. So it's all based on the heterogeneity of the patient population and the length of time patients stay on, those are the key metrics that inform on the relationship with treatment outcome.

And then for the third one, so about the translational data. So we are very excited about this data and also our partner, BMS. We're starting this pathway for over a decade. And we established the biology and we established the preclinical data that shows that if you combine these 3 agents together, they're all modulating the BMS access, and they eventually are synergistic in multiple models and to clinical settings. And now for the first time we see, in patients, such a potent immune activation, reflected by the potency, by the consistency and evident by multiple readouts. And this is very exciting to see this transit now into patients.

Now in order to actually see -- and maybe just to mention again, as mentioned before, that if you compare this data to our own data with COM701 on mono and in combination, and also to all published data that we're aware of TIGIT and mono and combinations, the effects you have seen, again, is more robust and more potent, again suggesting that the triplet is doing something different from each of the combination of the agents by themselves.

Now what about the tumor environment? So as mentioned, these are heavily, heavily pretreated patient population in this specific trial, and more things need to happen in the tumor microenvironment and also they actually see tumor shrinkage. So where we are following also biopsies and when we expand the trial and we also look tumor microenvironment, this peripheral immune effects are potent. And probably in this heavily pretreated patient population, we're not sufficient in tumor microenvironment to modulate it enough to mediate tumor shrinkage. And we are looking forward to see how will this regime look like in our biomarker-driven approach and the expansion cohorts.

The next question is from Reni Benjamin of JMP Securities.

I guess just starting off with the triplet and, Henry, some of the comments you have made regarding the expansion cohorts. Did I hear right that prostate wasn't one of the ones that you'll be expanding into? And if I did hear right, I was just kind of curious, given the long-standing SD that you have here, why that is?

Yes. So for -- thank you, Ren. So for the expansion cohort for the triplet, we're focused on the biomarker informed patient population. So we're focused on ovarian cancer. We're focused on endometrial cancer. We're also now focused on head and neck squamous cell carcinoma, and we also have a basket cohort. This is of interest to us also, prostate cancer, but understandably, we think we need to prioritize the indications that are of interest to us, and that's why we're focused on what we have now currently as part of the expansion for the triplets.

Okay. And then as you guys have looked at this data, have there been any new biomarker-based insights that you might be able to provide more color on? And have you learned anything new that you might be able to extend or you might be able to use in your expansion cohorts?

So I would say that, again, the main observation, as just mentioned, the translational data in peripheral blood. Regarding biopsies, they are not mandatory in this part of the trial. But obviously, this is work in progress. We are collecting samples for all the patients, especially now in the expansion cohorts. We will follow a correlation to the different biomarkers we explore. We'll explore also modulation in the tumor microenvironment, but this is all work in progress.

Ren just...

Sorry, go ahead, Anat.

I was just going to tell you that in general, and Eran was alluding to it, the fact that we were observing potent immune activation in the triplet, in the doublet with COM701 alone, from our perspective, it is really highly supportive the hypothesis that we came up with for the PVRIG pathway and in combinations, and that for us was critical in the dose-escalation setting just to get more comfortable entering into the expansion cohort. But now, there's a lot of work ahead of us pushing these studies forward and making sure that we act on all fronts on the translational approach that we're taking together with evaluating the DNAM axis hypothesis. So at this point in time, this is what we have, but we're encouraged with this.

Got it. And I guess just finally, given these results in the triplet and how happy you are with them, why not evaluate 902 and 701 with an additional either PD-1 or PD-L1 inhibitor versus the ongoing 902 701 study? And when we think about PD-1 versus PD-L1, right, if we look at Roche's data, it's -- you see that response rate with a PD-L1 inhibitor. Is there a biological rationale why -- to stick with, let's say, Opdivo? I understand from a cost perspective and the Bristol collaboration. But is there a thought process or something that you guys are thinking about now, where you might want to evaluate some other combination partners?

So I'll start answering this, and Henry, if you have any additional thoughts, please chime in. We're now -- we've now launched the studies that we wanted to do, and we're doing it in cooperation with Bristol-Myers Squibb in nivolumab. From our perspective, obviously, these studies are serving as a way for us to try and address the relevant drug combinations for the different patient populations in the different indications, and try to assess which indications and combinations we should focus on for the future study.

Having said that, we're not ruling out using additional PD-1, PD-L1 inhibitors in the future. At this point in time, under the collaboration with BMS, we're committed to pursue this path forward and get the data that we need in order to dive more deeply into certain indications with certain combinations. Henry, do you want to add anything?

No, I think you've answered most of it. Just one little part of it that I think you've articulated very well before in the past and continuously, Anat, and that's not just to look at the PD-1, PD-L1 axis. But also to rely on the strength that Compugen has with its antibodies. So we also think strategically of a PD-1 and a PD-L1 free regimen, and as a consequence of that, we have this ongoing study with an expansion cohort of COM701 and COM902, enrolling patients with non-small cell lung cancer, head and neck squamous cell cancer and also colorectal cancer. So that's the other strategy, which I think might be of interest to you also. And I know you remember this, Reni, so not just the PD-1, PD-L1 combination but one that is free of those events.

The next question is from Asthika Goonewardene of Truist Securities.

Asthika Goonewardene here, I apologize for any background noise. I'm at the floor in the city here. Just want to check on the patients on the triplet combination that you did show stable disease, are you able to get any biopsies of the tumors on progression? And maybe can you talk to us a little bit about what were the signals that you might have been seeing that could have been driving the progression after the therapy?

Yes. Asthika, that's a good question. So for patients -- because the data that we are presenting is dose escalation, biopsies are not mandatory as opposed to patients who are enrolled on the expansion cohort where there is a pre-assessment with a biopsy and another biopsy after a certain number of period -- time on the study. So the short answer is we do not have biopsies on any of these subjects who are -- on subjects who are either ongoing or the subjects who have stable disease or the subjects who progress.

This concludes the Q&A session. I will now turn the call back to Compugen's President and CEO, Dr. Cohen-Dayag. Would you like to make your concluding statement?

Yes. Thank you, operator. Thank you for joining us today and your continued support. We hope to see you here at SITC. Stay safe and healthy.

Thank you. This concludes the Compugen Ltd. Third Quarter 2021 Financial Results Conference Call. Thank you for your participation. You may go ahead and disconnect.