Agenus Inc. (AGEN) Earnings Call Transcript & Summary

Good morning and good afternoon, and welcome to Agenus ESMO GI webcast. We're coming to you live from Barcelona, Spain where we're excited to share with you the remarkable botensilimab/balstilimab combination data we released earlier today in which Dr. Anthony El-Khoueiry presented at ESMO GI's opening session. I am Ethan Lovell, Agenus' Chief Communications Officer, and I'm glad you've been able to join us today. Before we begin the formal presentation, however, I do need to provide our forward-looking statements disclosure. I'd like to remind you today that this webcast will include forward-looking statements, including statements regarding our clinical development, regulatory and commercial plans and time lines, including time lines for data release and partnership opportunities. These statements are subject to risks and uncertainties, and we refer you to our SEC filings for more details on these risks. With that housekeeping out of the way, I'd like to invite Dr. Steven O’Day, Agenus' Chief Medical Officer, to the podium to provide some opening commentary Dr. O'Day?

Thank you, Ethan. So this is both a remarkable and an exciting day for Agenus. This morning, we had the opportunity to present our combination data with botensilimab and balstilimab in advanced MS stable colorectal cancer. The data is both remarkable and unprecedented in this advanced line of therapy and it allows us to highlight one of our many deep pipeline jewels, botensilimab, which is a next-generation CTLA-4 that is much more than that. It's an engager of antigen-presenting cells. And it has been specifically designed to address the unmet need from first-generation CTLA-4s where cold tumors did not respond. So as a pioneer in CTLA-4 early therapy in melanoma, I saw both the great success and the great limitations of CTLA-4. The great success was in hot tumors. The great disappointment was in cold tumors. I've waited 20 years to find a molecule that could address this cold tumor phenotype, which is more than 70% of the solid tumor world. And today, we have some preliminary data that is very exciting. So I'm going to introduce our panel, and I'm going to then finish some opening statements, and then we'll have Dr. El-Khoueiry give the presentation from this morning. And then the exciting part beyond that will be a group discussion with 3 of the top global GI oncology leaders. I can't wait for that. So the three panelists is Dr. Anthony El-Khoueiry, who's the Associate Director of Clinical Research and the Director of the Phase I program at the USC Cancer Norris Center, my hometown of Los Angeles. Anthony is the PI on the trial and presented the data this morning. Dr. Heinz Lenz, his boss, the Deputy Director of the Cancer Center at Norris Cancer Center in Los Angeles. And Manuel Hidalgo, who is the Chief of Heme-Onc at Cornell Weill Cancer Institute in New York City and a GI expert. So what's remarkable about botensilimab is its mechanism of action. It truly is about activating the front end of CTLA-4 molecule to engage these robust T-cells that have memory and the ability to eradicate tumor. But on the back end, it's been Fc enhanced to bring the synapse between antibody and the antigen-presenting cells and bring it to the tumor [ microenvironment ] with the T-cell so that the expansion of the T-cell repertoire into colder tumor antigens can be brought forward and realized in the clinic. That was the hope, that was how it's designed and now we have some real data. As you know, last year, we presented at SITC a Phase I trial. Dr. El-Khoueiry and Dr. Lenz were part of showing activity of this single agent botensilimab or balstilimab in 9 different solid tumors. We have not tested more than that, but in 9 that we've tested. And so this has broad applicability. What we're presenting today is an expanded, the most expanded cohort to date of that trial in MS stable colorectal patients. I just want to highlight that this was a metastatic colorectal, MS stable population that on median average, 50% had already had 4 prior lines of therapy. And 1/3 of them had already seen an I-O. And we know I-O has not had any deliverables in MS stable colorectal cancer. We've also looked at our responders and there's absolutely no biomarker that's really suggesting why a traditional CTLA-4 should respond with a PD-1. For example, TMB was low at 10% or less. PD-L1 was only up-regulated in one of the patients. And there were no poly mutations or other genomic features that might predict from an I-O response. So this is taking it from a traditional CTLA-4 cold tumor with this new molecule and showing clinical activity. The responses were both impressive and durable and the toxicity safety profile was excellent. So I just want to -- we'll finish the session at the end and some of the Q&A about where do we go from here. But as a company, we're very excited about putting this molecule front and center with high priority to move forward, not only in this trial where we'll be delivering more data later in the year; in other solid tumors, not just MSI colorectal, so other cold tumors with exciting activity, but we're also then launching robust global Phase II programs in colorectal MS stable, melanoma and pancreatic cancer. So we've got a lot of work to do, but today is a remarkable day for the company. And I look forward to sharing this with these world-class experts. Thank you. Anthony is now going to go ahead and give the presentation of this morning.

All right. Thank you, Steve. All right. We have the slides. Thank you. Okay. So on behalf of my fellow investigators, I'm honored to present data from the ongoing study of botensilimab, a novel innate and adaptive immune activator plus balstilimab, an anti-PD-1 antibody in a cohort of patients with metastatic pretreated microsatellite stable colorectal cancer. There continues to be an unmet need for advanced refractory metastatic colorectal cancer in patients who are microsatellite stable. So standard of care options, such as regorafenib or TAS-102 have low response rates and very modest survival improvement. Single-agent anti-PD-1 and PD-L1 antibodies are inactive in this setting. Responses with first-generation CTLA-4 antibody combinations are quite rare. For example, in the CCTG study of durvalumab and tremelimumab, which was recently reported, there was only one objective response in 119 treated patients. And then lastly, there are several experimental PD-1/TKI combinations that have shown variable results and modest activity overall. You already heard from Dr. O’Day about botensilimab, but this is to show it schematically. So this IgG1 antibody not only block CTLA-4 interaction with its ligands, but it has this Fc-enhanced portion that increases the binding to the Fc gamma receptors on antigen-presenting cells and NK cells, thereby tightening the synapse, the immune synapse with T-cells. This results in several unique properties relative to first-generation CTLA-4 antibodies. This includes increased frequency of dendritic cell infiltration, expanded T-cell priming, expansion and memory T-cell formation, enhanced Treg depletion. There's also a single point mutation designed to reduce complement fixation, which is hypothesized to also reduce complement-mediated toxicities such as hypophysitis. Balstilimab is an anti-PD-1 antibody that's been evaluated in over 650 patients with activity and safety similar to other anti-PD-1 agents. This is the -- these are the baseline characteristics of the MSS cohort from this ongoing Phase I study. So microsatellite stability was an eligibility requirement for this cohort, and it was determined by local assessment. Patients in this cohort were both from the dose escalation and expansion. This was a heavily pretreated population with a median of 4 prior lines of therapy. 50% of patients had 4 or more lines of therapy. 34% had prior immunotherapy. As per design, all patients were microsatellite stable, and the RAS and RAF mutation profile was as expected in this population. Now to be considered evaluable, patients must have had at least one restaging scan at 6 weeks. In the 41 evaluable patients, the objective response rate per RECIST 1.1 was 24%, with 10 partial responses to any patients with stable disease, resulting in a disease control rate of 73% at a median follow-up of 5.8 months. Eight of the responses are ongoing. Three responses are beyond 1 year. And the median duration of response is not reached yet. Here is a waterfall plot showing the change in target lesion diameter. The plus signs reflect an ongoing response or ongoing stable disease. Of note, the responses occurred independent of RAS status. Only one of the responders had -- was reported to have high PD-L1 expression. And all responders had a TMB of 10 or less. Two of the responders were I-O refractory had prior exposure to I-O, and one of them was an I-O PD-1/TKI combination. Here's an overview of safety. These are treatment-related adverse events that occurred in 10% or greater of patients. 76% of patients had [ 8 ] treatment-related adverse event, but the majority of these were Grade 1 and 2. Grade 3 events were limited to 24%. The most common adverse events were GI in nature with 16 patients having diarrhea/colitis, but only 4 of these were Grade 3 in nature. There were no Grade 4 or 5 treatment-related adverse events. The investigator assessed immune-related adverse events occurred in 46% of patients, 17% had Grade 3 irAEs. 10% of patients discontinued treatment with bot, with botensilimab due to an adverse event and another additional 10% discontinued both drugs. In summary, botensilimab plus balstilimab is a novel Fc Enhanced CTLA-4/PD-1 combination. In heavily pretreated patients with MSS colorectal cancer, we've seen deep objective responses with evidence of durability. The combination has a manageable safety profile with enriched responses in patients without active liver metastases. I guess we didn't have this slide at the moment, but in an exploratory analysis of patients with active versus without active liver metastases, the response rate in the patients without active liver metastases was actually above 40% and the disease control rate was 96%. Based on the totality of this data, a global Phase II trial in MSS colorectal cancer will launch later this year. Thank you.

Anthony, for that nice presentation. second time I've heard it today. So it just keeps getting better. So we're going to take you off the hot seat right now and your colleagues, 2 world-class GI experts deeply in a field of unmet need of I-O experience to date in GI. And we'll talk about other GI malignancies because I really think this opens the door to a lot. But maybe start with Heinz as the colorectal guru of the world. What do you think? I mean, what's your reaction? And how do you place this in the context of your personal experience with I-O and just in this line of therapy, maybe then?

I think these data are incredibly impressive in a patient population who have very little treatment options and the options available have very low response rate. So I think this could be paradigm changing, showing for the first time that immunotherapies can be effective and can turn a cold tumor into an immune responsive tumor. So I'm extremely excited. It can change the way we look at these tumors because we have been frustrated with immunotherapy [ do ] combinations in MSS colorectal cancer with very little efficacy and this will change our view in how we move forward using immunotherapy in this patient population.

So give the audience some perspective because as you have pioneered the MS high subgroup of colorectal with combination CTLA-4. So obviously, coming from you, saying this is a cold tumor turning warm or hot says something. But do you want to just relate that experience with the subgroup that has clearly benefited from this approach?

Yes. So there is no question about it, MSI-high tumors don't need chemo. They need immunotherapy with big success. Now saying that the MSI-high tumors still have as best responders, 20% to 30% progression of disease. So what I thought would be how would they do [indiscernible] Fc-engineered antibody with the rate of progression, which I think is really essential. 20% to 30% do not respond, will they respond to a better engineered antibody. So I have no doubt they will. To put that in perspective in MSS, second-line chemotherapy efficacy response rate at 10% to 15% to see 24 in all and 40 without liver lesions will change the spectrum, completely an upside down treatment options in the different line of treatments.

Yes. I'm just thinking of melanoma, obviously, ipi/nivo and [indiscernible] has produced great responses. But to your point, that there's still 20%, 30%, 40% that are unmet by that. You can't help but wonder what next-gen CTLA-4 could add even to a hot tumor. So the point is hot tumors, we've got room. But clearly, the message today is cold tumor that has been just refractory to I-O basically.

Yes, I think, and I always believe in the CTLA-4 component with nivo because when you look at all the data, it is more effective. So you have an optimized CTLA-4 makes me even more happy. And that the duration of response has not been reached is further a sign that this is really an immune response which last.

That's the hallmark of CTLA-4 and these durable memory T-cells that don't require continuous treatment.

I agree.

So Manny, we're -- love to hear your thoughts. You span the whole GI tract, but let's first talk about the colorectal data. You're well versed in this -- your reaction?

Thank you, and great to be here today, and listening to the presentation twice as well. Well, I fully agree, the data is quite impressive. Certainly, patients in the third line and beyond for whom standard of care offers as we have seen, extremely limited benefit in response rate and survival and progression-free survival. Other I-O combinations or immunotherapy with other strategies is not there yet, a little bit better, but not substantial. And this combination, which is PD-1 with, I think, a brand-new drug, a drug that targets CTLA-4 but has them an Fc segment that gives very different properties. You have a very clear overall response rate with a manageable toxicity profile. So I fully agree that this is a very important dataset.

So let's -- I'm sorry we didn't have the slide that showed the enrichment for nonactive liver mets. But just to reiterate, in patients who had had prior liver mets that were treated either surgery or ablation or local techniques, and then development tests -- or patients who did not have active liver mets at all at any time, the response in that cohort was about half the patients was over 40%, with disease control [indiscernible] near 100%. Now this is a subgroup where I-O should work in hot tumors the best. It's reassuring to see it working very well there. But there is also activity in this 41 data base, although there's no objective responses in the active liver, we had stable disease at 24 weeks and some minor tumor regressions. So we are not in any way conceding that we wouldn't potentially have an impact on active. But having said all of that, can you talk a little bit about, obviously, debulking liver, regional strategies to liver, when you have a platform of an I-O that's very effective systemically, will this reawaken sort of thoughts about combination directed and maybe encourage surgeons to be more aggressive debulking before we implement? The other option, of course, is to tweak the tumor microenvironment. We have a pipeline that we think can really do that, and we're going to work with others around the world. But let's first talk about Heinz and then maybe Anthony about what do you think about the liver data? And how do we maximize the active and not give up on that in any way?

Yes. So the development of immunotherapy for MSS colorectal cancer really hit the wall with the liver metastases. And it's still not clear what the tumor microenvironment does, that it's so immune suppressive. I think this will be a really important challenge for the future to find options to overcome this immune suppressive environment. There are a lot of activities in this microenvironment, which can play a role, including the endothelial cells, including the fibroblast, including the dendritic cells, including the neurotransmitters. So I think there will be a lot of things to learn in the future. Saying that, I think that the liver-directed treatments or surgical local-regional treatments may become differently looked at when you have a treatment efficacy of 40% outside. I think that could challenge a little bit the lines of treatment segments we usually consider. So if I have liver lesions, which I can go easily after and have multiple lung lesions, you may want to consider a liver-directed treatment -- surgery, [ SBRT ] and then have a treatment of -- in fact, we have already done that in some patients, where we know that treatment options open up for patients like this. So I think it may challenge.

So I hate to bring it back to melanoma, but that's exactly as we had more active systemic treatments, both I-O, I-O combinations in BRAF [indiscernible] the approach, which was always more surgical and debulking when we had nothing also increase further when we had more for that very reason. So I would expect that possibility. Anthony, you're in hepatocellular and -- in addition to colorectal, but you focus a lot on cholangio, hepatocellular, where regional therapies. Just same question to you that -- what do you think about this data and how we might optimize combination?

I fully agree with Heinz, it is something that absolutely needs to be explored. The optimal liver-directed modality is not clear. The extent of debulking that needs to happen is not clear. So there's a lot of unknowns. But I'll refer to actually a very important proof-of-concept study that was done at NCI by [ Tim Gratton's ] group in HCC, where they actually gave patients who were refractory to treatment with HCC just tremelimumab and just partial ablation of single lesions in the liver. And I'm not talking about the clinical efficacy, I'm talking about a biologic proof of concept where they showed that with this partial ablative therapy, they had a definite positive immune modulation, right, increase in CD4, cytotoxic T-cells, et cetera. So the impact on the micro and -- microenvironment could be positive, and I think it's worth exploring.

Disrupting the tumor, releasing antigen and disrupting the myeloid and other suppressive group. So it also begs the question earlier -- now let's talk about -- this was a median of 4 prior lines, an extraordinary activity. Obviously, it begs the question of first line, earlier lines where liver mets may be less dominant. But even adjuvant and neoadjuvant where liver is not even present, if you truly have a systemic treatment that is this active. Manny, do you want to just...

Sure. I think this is the beginning of a developmental potential, which is way beyond the second and third line. You could envision strategies to move this into first line. Certainly, first-line patients, you reduce then the tumor with chemotherapy, with antibodies and then maybe the liver disease is less prominent at that time. And those sort of local treatments that we alluded to before, we already made some of that, you can start thinking of maintaining the strategies. And then going to earlier lines of disease, we have seen some very recent examples in rectal cancer, MSI rectal cancer, how effective immunotherapy is if we have now a regimen that is effective in MSS, can we think on development in rectal cancer in situations where surgery would be mutilating or there will be very large -- high toxicity of treatment. I think it's going to be a fascinating spectrum of possibilities down the road.

So Heinz, you've obviously pioneered both the metastatic MSI-high and also the neoadjuvant approach. Correct me if I'm wrong, but in the neoadjuvant high-MSI, obviously, the response rate is over the top. But even in the stable with ipi/nivo, weren't there some pathologic response, which correlates with the melanoma data that clinical size change in lymph nodes in the neo has radiographically underestimates the path. So if there is already a hint with first-gen ipi/nivo in MS stable where liver is not involved, couldn't this really be a platform to study it further?

Yes. So I think there is a difference between local disease versus metastatic disease. The inflammatory response in these tumors are completely different, which is one of the argument when you do nivo/ipi and MSI rectal, localized, you have 100% pathological response. So with the same argument, you would think that an MSS local tumor would also respond better to an optimized treatment regimen like you presented today. So I have no doubt. Will it be 100%? I don't know. But certainly, that window of opportunity should be really evaluated. For metastatic disease, I think you have now a really effective backbone. I think the challenge will be, can you further optimize it? Can you find an optimization and combination, which targets liver metastasis? So I think you have now a pool of data and preclinical models where you can play around and further optimize. In the meantime, you can treat the extrahepatic disease very effective there. It should go to a second line. There is no doubt in my mind because in second line, we have really no effective treatment. First line, I think I would do more homework and optimize the combination.

Well, we look forward to engaging you in your center around that as well as others. We have an ILT2, that's about to hit human -- first-in-human any day now that we -- is sort of really a myeloid drug that will really not only affect myeloids but also lymphoid cells and repolarize them to activated [ MH ] TAM-1 cells. So we're -- and the liver, of course, is a microenvironment that's particularly myeloid hostile, so to speak. So again, that's just one pipeline type approach, but I agree. So I have another question here about, let's take this out of colorectal right now. Obviously, we have evidence in gynecologic -- cold gynecologic tumor, cold sarcomas, even the warmer sarcomas like angiosarcoma, there's visceral angio in cutaneous and traditional generation ipi/nivo type regimens have not really managed to show response in these visceral low TMB tumors. And so we have evidence of other diseases that we're going to be very interested in exploring further. But let's talk -- we haven't really gone much outside of the colon and the GI tract. Some comments about pancreas is going to be a focus based on some preclinical data with our botensilimab in mouse models, very high-level mouse models showing curative potential with botensilimab and chemotherapy. And Dan Von Hoff has been pioneering that work with us and Manny has also, and we've sort of -- we want to engage that study in the Phase II in the coming months. But -- so Manny, why don't you talk about pancreas maybe? And then what about these warmer GI tumors in the upper tract and even the liver, what's the potential -- where some I-O is already established in itself?

So as you mentioned, the mechanism is designed to warm up these cold tumors. And the coldest is probably a pancreatic cancer for which really there has not been any hint of activity, even the MSI patients, they respond, but they don't do that well compared to colon or endometrial. So very different biology. So testing this in pancreas cancer, I think it's necessary. The preclinical data you mentioned is very, very strong in the [indiscernible] of the strongest data I have actually seen. So I think it would be very important to do these studies. And there's some other tumors in the GI tract that are perhaps not so cold, but not that warm, cholangiocarcinoma, liver cancer, maybe Anthony, you want to comment on those. But I can see a broader spectrum of development in GI.

Anthony?

Yes, sure. Thank you for the hand off there. I think I would say hepatobiliary cancers are very heterogeneous, right? So there is a subset in both HCC and cholangio that is probably more immunogenic and then a majority that's not. So there is an opportunity here to expand the benefit of immunotherapy more broadly in these 2 areas. And I think the combinability of botensilimab is outstanding because of the safety profile. So you could add it to first-line regimens. You could explore it in second-line combinations. Certainly, nowadays with cholangiocarcinoma, first-line therapies really largely mirror pancreas, right, with gemcitabine-based combinations plus durvalumab based on the TOPAZ study. So adding there is another opportunity. So I think that's definitely excitement there.

So I think in the sort of warm tumors, PD-1 chemo combinations have been quite -- I mean, [ Merck ] because obviously, it's been a huge platform in lung and maybe upper GI. But in the cold tumors, PD-1 chemo combinations have really not done well, ovarian, pancreas. And it's my sort of thoughts that the partners not -- shouldn't always be PD-1. If you don't have any engagement of the tumor by T-cells, there's no exhaustion to sort of recover from. And so I see a great opportunity in particularly colder tumors. We're not just reflexing to PD-1/CTLA-4 combinations, but actually using next-gen CTLA-4 as a foundational partner with chemo, for example, in pancreas, but other diseases, PARP and other opportunities. So I think there's going to be a lot of opportunity to position this with PD-1 when the biology and the immunology makes sense, but not necessarily waste PD-1 on cold tumors that aren't recognized. Now you could argue that once they're recognized, you can drive them a [ prevent exhaustion ]. So there may always be some additive role. But I think the proof of concept in the cold tumors may be to really look at CTLA-4 that's active in cold tumors and then combine without PD-1 with thoughts on just that kind of thinking.

I think this makes sense. I think a very unique opportunity for this combination would be gastric cancer. It is immune responsive but not [indiscernible]. But there are these subsets where you have a high CPS score where immuno is better than chemo. And I think the combination showed better when you have a CPS of 5. So there is a lot of fluctuation in the gastric field where you have some immune responsiveness. If you optimize this, you may get rid of the chemo. And I think that would be really a paradigm-changing approach. So I think there are unique opportunities where you have the cold and then some immune responsive and then the very immune responsive tumors, I think it makes sense to do cold tumors, like you have shown with the MSS colorectal cancer. But look at the ones which have some efficacy where it is limited or where you can really change the picture and the treatment strategy completely.

Yes. My fantasy is, of course, to have a platform of a next-gen CTLA-4 that's active in cold tumors and then to use brief burst of cytotoxicities whether with short couple cycles of ADC or radio ablation -- and then really let the magic of these memory T-cells do their work.

But you may not need it. I mean look at the [indiscernible] treatment. I think for some tumors, you will probably need some combination, and you can further enhance the efficacy that the chemo need maybe not as high as we still think, when you find the right partner. So I think this is a very good start in order to have an effective combination for cold tumors. And it will go potentially in different combinations depending on the site of metastases or the primary tumor, but very unique opportunities and options.

Great. So should we go ahead and -- I think this has been -- I could do this for hours, but -- with this panel. But let's -- I think there -- I want to introduce Dr. Garo Armen, our Chairman and CEO of Agenus, who's in the room; and Dr. Joe Grossman, our Vice President of Early Clinical Development and a GI oncologist who has pioneered this program. Kudos to Dr. Grossman, who I've sort of been his wingman, but he has really driven this MS stable colorectal. So both are in the room, I think we can take some questions.

Okay. All right. It's now time to move to the Q&A portion where we can take some questions from the online participants. First, I'd just like to thank all of our panel members. I mean it's a fascinating discussion. And it's always really enlightening to hear true experts opine on these various fields because you really understand that there's such complexity to the management of patients and introducing various therapies that make a tremendous difference in their future. So I just do have a few housekeeping comments to make about asking and having answered questions. So in order to ask a question on the webcast, you see a line that says click here to ask your question by audio. You're going to need to click on that and then open Zoom meetings, which will be the next prompt that you receive if you're participating online. One other important point, if you are in the queue to ask a question, and you are prompted to ask that question, please mute the audio of the presentation that we're giving here in order to avoid feedback on the webcast presentation. So with that, maybe we'll turn it over to the operator to take our first questions.

[Operator Instructions] The first question will come from Matthew Phipps from William Blair.

Hopefully, you can hear me. And congrats on this update. Just curious if you knew -- if you had what the response rate was in patients who had prior immunotherapy? I was kind of surprised that so many people had prior immunotherapy in this trial. And then Dr. O'Day, just wanted to confirm, you said only one patient had elevated PD-L1 expression?

Yes. So the answer is yes to that. And part of the question was what?

[indiscernible] in prior -- people who had prior immunotherapy?

Yes. There were responses in prior immunotherapy. I think there were 2 response. Anthony knows the data, yes. 2 responses out of the 10 were in prior immunotherapy. About 1/3 of the 41 patients had seen some form of -- and that includes single-agent PD-1, PD-1/CTLA-4 combinations or PD-1/TKI. The 2 responders in the I-O group were a PD-1 monotherapy progressor and a PD-1/TKI progressor. And the TMB was low, 10 or less in all of our responding patients.

You made -- the slide did mention that pneumonitis was rare. Can you give any more details? And I guess, maybe just general thoughts from the panel on the safety profile here versus what they might expect from other I-O combinations or I-O/TKI combinations?

Sure. Yes. So there was no pneumonitis in this -- there was actually one pneumonitis in this cohort. As a matter of fact, in the [ investigator brochure ], there are 2 cases, I think, of pneumonitis mentioned in total. And to date, 203 patients have been treated. So of course, it's an ongoing trial. The data is still evolving, but it's a very rare phenomenon with -- in a large number of patients. And as we had mentioned earlier, no cases of hypophysitis, kind of consistent with the design to reduce complement fixation.

Can I ask one more question? Just on as far as -- if these responses were confirmed [indiscernible].

The next question will now come from Mayank Mamtani with B. Riley Securities.

Can you hear me now?

Yes, we can. We can.

Okay. There's a little bit of a lag, I think, when I come on versus the webcast. But yes, regardless, congrats on the data. So just maybe 2 follow-ups for Dr. El-Khoueiry about the dataset presented today. So the liver mets versus non-liver mets, 50-50 proportion in fourth line plus setting. Is that kind of consistent with sort of prior studies we've seen from I-O drugs and other sort of studies that you referenced initially? And then the second question was that is there a monotherapy data that is accumulated here because you do point to -- in your -- in your safety table, there was some monotherapy data also referenced, and you also talked about the combination that it's all 10% discontinuation. So do you have any information on the efficacy side for monotherapy?

So I think the first question, if I heard it correctly related to a comparison of the activity that's been seen with other therapies in active and nonactive liver mets. So if you think about some of the waterfall charts that we have with regards to that.

Sure. Look, it's always difficult to do cross-trial comparisons and cross-study comparisons. But in a nutshell, if we look at I-O/I-O combinations, meaning PD-1/CTLA-4 combinations, the biggest study is the 119 patients from Canada with durvalumab, tremelimumab with 1 objective response out of the 119 patients. So you put this side by side, by this, it's obviously this has a much higher objective response rate. When you look at PD-1/TKI combinations, even though that's a completely different concept, right, but let's look at that. The response rates are anywhere from zero to maybe low teens to 15%. When we look at lung mets only in a few studies, it gets up a little bit higher, but on average in the low 20s. So even then side-by-side comparison, this appears more active overall as far as objective response rate and in the non-liver mets patients. So I hope I answered that question.

And so I'm not sure what the second question was. Mayank, would you -- do you mind?

He is asking about single agent. So from the Phase I data that we've shown, and Steve, you can bail me out here if I forget something. There's definitely single-agent activity in cold tumors. So we did see responders with cold tumors. If I try to remember off the top of my head, I believe a pancreas response and in MSS endometrial may have been in single agent botensilimab. So I'm getting nods from Joe.

Yes, that's correct.

Cervical.

Yes, I was asking colorectal because I believe the plan is to look at monotherapy activity and combination and then compare it to standard-of-care regorafenib. Is that sort of -- Dr. O'Day, maybe for you, is that sort of a trial design we -- sort of an update on that, the thinking for going forward? Or is this data kind of informed you to do things differently because it's definitely much better than what we saw in the prior update?

So commenting on the plans going forward around combination potential with regorafenib and Phase II study design.

We're committed to moving forward quickly in melanoma, pancreas and MS stable colorectal. We have not publicly disclosed the details of those trials. We're in active discussion with the FDA on multiple levels. We clearly, in colorectal, the combination is very active and safe, and so we're excited about that. We are obviously going to go into melanoma, and we have had single-agent activity in PD-1 resistant CTLA-4 naive melanoma and in CTLA-4 refractory melanoma. Again, suggesting that we have something different than a [indiscernible] CTLA-4 even a slightly modified CTLA-4. So we're seeing single-agent activity in hot tumors, PD-1 resistant tumors, cold tumors. And even we've reported a response in a CTLA-4 refractory hot tumor. So very exciting on the single-agent activity approach and then obviously, the combinations seem well tolerated and the unmet need is huge. Thank you.

If I just [indiscernible] final question, if that's okay?

Go ahead, Mayank.

And this is for the GI experts. As you think about earlier line settings and also thinking chemo-free regimen, a lot of other indications are benefiting from. What would you think is sort of the combination partner of choice, both from a legacy kind of targeted therapies, but also you guys are excited about some of these emerging pipeline molecules also? So just curious if you guys could comment on that would be helpful.

So I think, obviously, we've commented on our company strategy around clear experiments in melanoma, colorectal and pancreas. But we're really -- we're -- you can imagine the engagement from the GI community, from the sarcoma community, from the GYN and even my melanoma colleagues who haven't got a lot of melanoma data are incredibly impressed with the concept of a cold tumor turning warm. So we are going to be engaging Joe and his colleagues at the company are obviously an investigator [ initiated ]. We're going to take the best minds of academia around the world, global leaders and show them this data as we have. And we're going to have them help us develop proofs of concepts and earlier lines of treatment so that we can rapidly get this drug by itself and in combination to approval. So it's going to be an exciting year or 2 here as we move this forward. The first step is you've got to get a cold tumor warm or hot. I want to reemphasize that is not an easy task. It has been the plateau of the second revolution of immunotherapy. First revolution, CTLA-4; second, PD-1. We have been flat in cold tumors. And these are early days, but this to me, excites me to see something start to transition a cold to warm and hot. The future is wide open, but we're going to need help. We're going to need our academic and KOLs worldwide, our investigators, but I can say that all of our investigators in these diseases that we're sponsoring trials in the coming months are really excited to join and it's been a pleasure to introduce this drug to that.

That is all the time we have for questions today. This will now conclude our webcast. Thank you for your participation.

Thanks, everyone, for joining. We appreciate it. Have a wonderful day.