Arcus Biosciences, Inc. (RCUS) Earnings Call Transcript & Summary

Good afternoon, ladies and gentlemen. My name is Loya, and I will be your operator on this call. [Operator Instructions] Please note that this call is being recorded today, Wednesday, June 23, 2021, at 2:00 p.m. Pacific Time; and will be available on the investors section of Arcus' website at www.arcusbio.com. I would now like to turn the meeting over to Kaytee Bock, Vice President of Investor Relations and Corporate Strategy from Arcus. Please go ahead, ma'am.

Thank you. Hi, everyone. Thank you, Loya. And thank you all for participating in today's call on such short notice. Today, we will be discussing this afternoon's announcement regarding the interim analysis from the ARC-7 study. Joining me from Arcus are Terry Rosen, Chief Executive Officer; Juan Jaen, President; Jennifer Jarrett, Chief Operating Officer; Bill Grossman, Chief Medical Officer; and Kartik Krishnan, Senior Vice President of Clinical Development. I'd like to remind you that on this call, management will make forward-looking statements within the meaning of securities laws' safe harbor provisions, for example, statements about our clinical development programs, time lines and potential of the Arcus-Gilead partnership. All statements other than historical facts involve risks and uncertainties that may cause our actual results to differ. Those risks and uncertainties are described in our annual report on Form 10-K and quarterly report on Form 10-Q, which have been filed with the SEC. We strongly encourage you to review those filings. This conference call contains time-sensitive information; and is accurate only as of the live broadcast today, June 23, 2021. And with that, I'll turn the call over to Terry.

Thank you very much, Kaytee. And thank you all for joining our call today and your interest in Arcus. We really appreciate your engagement. As Kaytee mentioned, the point of today's call is to discuss our announcement this afternoon regarding the interim analysis from our ongoing ARC-7 randomized Phase II study. We really recognize the intense focus around this disclosure. And I have to -- we're all genuinely excited to finally have the opportunity to see the data, share assessments of this data set and convey the potential opportunities that they afford us. First, I will spend a few minutes reminding you why we have a TIGIT program, which now [ CPS2 ] TIGIT antibodies in the clinic. And then we can do a deeper dive into the ARC-7 study and results from the interim analysis. I'll round up the call with a number of the other exciting things that we're doing here at Arcus as we continue on our journey towards becoming -- we genuinely believe we'll be a fully innovative commercial biopharmaceutical company. So since our founding only 6 years ago, we've built a robust drug discovery capability to create highly differentiated small molecules. Our small molecule portfolio and focus today include both immuno-oncology targets and mechanisms such as ATP-adenosine pathways as well as cell-intrinsic targets subjected to [ alpha and AXL ]. However, we also recognized from day 1 the essential need to own important backbone antibodies against PD-1 and TIGIT, which we could combine with our small molecules to create and develop rationally designed combination therapies, until we invested early and heavily to secure access to high-quality PD-1 and TIGIT antibodies. Through these efforts, I'm pleased to say that we now have 5 clinical-stage molecules, of which the majority have moved into randomized registration trials. We also have another molecule, our HIF-2 alpha inhibitor AB521, expected to enter the clinic in the second half of 2021. This will bring our total clinical-stage molecules to 6. We have also built what we believe is one of the best oncology development capabilities in the industry, particularly when you think about a company of our size, and our goal has been to advance our molecules into randomized clinical studies as quickly and efficiently as possible. And that brings me to TIGIT and ARC-7. So first, let me touch on the rationale for TIGIT as a target. TIGIT is an immune checkpoint receptor expressed on immune cells, including T cells and natural killer cells, and it's typically [ co-spread ] with PD-1. Activation of TIGIT by ligand CD155 results in suppression of these immune cells, leading to impairment of antitumor immunity. Clinical studies have shown that the combined inhibition of TIGIT and PD-1 results in synergistic antitumor activity. Most notable from this context was CITYSCAPE Phase II data, which showed significantly higher response rate and progression-free survival for an anti-TIGIT plus anti-PD-L1 doublet versus the anti-PD-L1 antibody alone. Early on, we recognized the importance of TIGIT as the next backbone immunotherapy and in-licensed our first TIGIT antibody. At that time, we made the decision to silence the Fc function. We believed then and continue to believe now that depletion of intratumoral Tregs was not a requirement for the clinical activity of a TIGIT antibody; and in fact, that long-term depletion of Tregs in the periphery could result in unwanted toxicity; and that the potential depletion of intratumoral cytotoxic T cells, which are the very same cells that we're trying to reinvigorate, could diminish the potential synergies between TIGIT and PD-1 antibodies. Consistent with how we pursue all of our programs, shortly after we selected our first anti-TIGIT antibody, domvanalimab, which we -- also we call dom, we initiated work on a second, differentiated molecule directed against the same target. We recognized that an Fc functional antibody with the potential to deplete TIGIT-positive cells could be beneficial for certain tumor types. More specifically, in solid tumors, TIGIT is found only on immune cells and not on cancer cells. However, in certain hematological malignancies, the cancer cells actually spread TIGIT. In this case, depletion of the TIGIT-bearing cells will be beneficial, so we recognize the optionality that this provided; and specifically designed our second molecule, AB308, to be Fc enabled. This early investment is now paying off. Dom has the potential to be the second preferred TIGIT antibody to the market and potentially the first in certain settings. And we have another anti-TIGIT molecule right behind it in Phase I combination studies with zim, and we expect to initiate our first expansion cohorts with this regimen later this fall. We believe we're the only company with 2 anti-TIGIT antibodies in clinical development. Before I get to results of the interim analysis, I really want to step back, take a moment to thank our patients and investigators for participating in the ongoing ARC-7 study. As a reminder, we designed the study to target patients with first-line metastatic non-small cell lung cancer with 50% or greater PD-L1 before the release of the CITYSCAPE data. The CITYSCAPE study utilized a PD-L1 antibody, and so the results we will discuss today represents the first public discussion of randomized readouts for a TIGIT antibody combined with a PD-1 antibody in this setting. The ARC-7 study includes 3 arms: zim, which is our PD-1 antibody; zim plus dom, to which I will refer as this doublet; and zim plus dom plus etruma which is our adenosine 2a/2b receptor antagonist and which was previously known as AB928. I will refer to this third arm as the triplet. This randomized study has a target total enrollment of 50 patients per arm or 150 total patients. As a reminder. The CITYSCAPE results in the PD-L1-high patient population were based on 29 patients per arm. ARC-7 was designed to provide us with a lot of valuable information, notably not just how the doublet performs relative to zim alone but also whether etruma might add efficacy to that of doublet. If positive, this triplet combination would allow us to develop a completely novel and differentiated therapy for this patient population. This arm is really the type of combination that Arcus is all about, unique and rational differentiation, which creates a potential home run for patients. Let me just give you a reminder that we and our partner Gilead had been blinded to the ARC-7 data. We blinded ourselves to maximize the integrity of the study and to preserve our ability to use this study for regulatory purposes. This first interim analysis, therefore, was designed to provide us with an early look at these data to form our future development decisions for both the doublet and the triplet. Now for the results from this interim analysis. As you know, we've been very clear that we would not share quantitative data from this interim analysis at this time. This is an ongoing study because it's still enrolling. And as a result, it is critical that we not disclose data that could potentially bias investigators or impact our ability to complete enrollment in the study. Also, the [ intent of ] top line release is to preserve the ability to present the results at a medical conference, which we obviously want to be able to do, so today I'll share as much qualitative information as I can. So the initial decision we made based upon the results of the interim analysis is that we and Gilead agree that both ARC-7 and ARC-10, which is our registrational study for dom plus zim in the same population, should continue as planned. In addition, we and Gilead agree that the results of this interim analysis supports the continuation of our ongoing joint efforts to prepare for additional Phase III studies for dom. Before diving into further details, I would first like to remind folks, given that this is an early look, clearly the total number of disease-evaluable patients in this interim analysis was relatively small and the data still needs time to mature in the ongoing study. Also, as with any ongoing clinical study and because this is an early look, the data could change over time. I will now spend a few minutes on each of the arms. The first thing that we wanted to confirm with this study is that zim showed activity similar to that of other marketed anti-PD-1 antibodies in the setting. And in fact, the results to date continue to demonstrate that zim has activity similar to that of the marketed anti-PD-1 antibodies. Having our own anti-PD-1 antibody provides us with a huge amount of optionality, so the performance of this first arm is an important observation for both Arcus and Gilead, which shares the rights of this molecule with Arcus. Second, we wanted to demonstrate that this doublet and triplet provide superior antitumor activity relative to zim alone. Both arms with dom-based combinations showed encouraging clinical activity that Arcus and Gilead believe are sufficient to continue enrolling our existing studies as planned and to continue with our preparations for additional Phase III studies. We're doing that together. In particular, we were intrigued and actually quite excited about the activity seen in the triplet arm, which is the first data set reported for an anti-TIGIT and adenosine receptor antagonist combination. Allowing the data to mature will enable us to better assess the relative contributions of each of dom and etruma. Given the exciting data from the triplet, we look forward to evaluating this regimen in other studies. In summary, we are encouraged by the data we have seen to date in all 3 arms and particularly the 2 dom-containing arms. As with all clinical studies, we understood there would be a trade-off between an early look and data maturity, but we're excited about the data and thrilled that they support the continuation of our ongoing TIGIT development planning strategy. The opportunities in this space are enormous, and our approach is to be thoughtful but expeditious. Generating randomized early data on -- early on to support our future investments in both the doublet and the triplet is an example of just that mindset. A key observation from this initial data set is that the time to response across all arms is rather long with an average time to response of over 3 months, and therefore the study and the data will benefit from greater maturity and more disease assessment. In fact, more than 25% of our responders did not achieve a partial response until after their second tumor scan. We find this extremely encouraging hallmark of the profile and benefits associated with immunotherapy-based therapies. Data from ARC-7 will be submitted later this year, for presentation at a medical meeting. So what does this mean for the Gilead option? The mechanics of the option are such that Gilead has to first determine whether a triggering event for an Arcus program has been met. If they make this determination upon delivery of a data package, Gilead will have a certain period of time, typically 2 months, to make a decision. As a reminder, Gilead gets what we'll say one bite of the apple for dom after triggering event has been met. Also, once Gilead opts into the program, they become responsible for 50% of the joint development costs. Since this was a blinded study, when we unveiled the data to the Gilead team, they shared their excitement to continue to move forward, particularly the dom combinations, including the triplet given the potential differentiation of this combination. I'm sure they will also communicate this with all of you. However, Gilead is determined they will not start the opt-in review period at this time. They expect to make a decision on the opt-in by year-end following maturation of the ARC-7 data to confirm the results that were observed in this interim analysis and to better understand the relative contributions of dom and etruma. Both parties recognize the importance of moving quickly here. This is also why we will continue with our Phase III planning, and the timing for the opt-in should not slow down [ first patient in ] for these studies. Since Gilead's option also includes AB308, I wanted to spend a minute on the status of our second TIGIT antibody, which is actually advancing very quickly. As a reminder: This molecule has been a Phase I/Ib study in combination with zim. This also has the potential to advance into pivotal programs for heme malignancies even as we advance dom in solid tumors. We started this study with 2 important advantages resulting from our extensive clinical experience with TIGIT. One, we started dose escalation of AB308 directly in combination with zim. And two, we started dose level of AB308 that we predicted, based on our prior studies, would be relatively close to the dose level we would want to take forward. And in fact, AB308 achieved full receptor occupancy in our first dosing cohort. We have already completed enrollment of the second dosing cohort and expect that the second or third dosing cohort will likely be our going-forward fixed dose for AB308 and zim, putting us in a position to start our expansion cohorts later this year. Before we move to Q&A, I want to update you on the other events happening over the next several months. First off, we will do an update on our ARC-8 study for AB680, which is our small-molecule CD73 inhibitor in first-line metastatic pancreatic cancer, expected to be presented in fall. This data set will include updated data from the ASCO GI presentation in January from the dose escalation portion of the study and data from the dose expansion. We continue to be very excited about the potential of this molecule in what we all know is a truly devastating disease. Second, we will be initiating the first clinical study for our HIF-2 alpha inhibitor AD521 also in the fall. We also expect to submit ARC-7 data for presentation at a medical conference by year-end. Last, we are currently evaluating etruma in 3 different randomized studies in TKI refractory EGF receptor mutant lung cancer and second-line-plus colorectal cancer and second- and third-line metastatic castrate-resistant prostate cancer. We look forward to being able to share the data from these studies, which we hope will further validate the potential of etruma in multiple patient populations. Finally, I wanted to wrap up with an update on our financial position. First of all, we feel we are fundamentally well aligned with Gilead in how we are looking at these programs, so we think it is unlikely we would advance our programs into registrational studies without the support of our partner. To date, Gilead has made a greater-than-$650 million total investment into Arcus through their upfront payment and equity investments, including having a 19.5% ownership to date, the most recent raised in February. And we believe their commitment to Arcus is as strong as ever. As a result, we have over $885 million of cash and cash equivalents as of the end of March and plenty of capital to prosecute our programs over the near term. We'll now open up the lines to questions.

[Operator Instructions] Your first question comes from the line of Alethia Young from Cantor.

Congrats on the moving forward with all 3 arms. Just I guess the question on everyone's mind is that thinking of obviously the Roche data and your internal bar being around the 50% or so. I'm just curious about how to think about views of -- competitive views around double and a triple? I mean, does it need to be kind of in that 60s neighborhood to be competitive? And I know you don't want to give numbers, but like anything qualitatively you can do to help us understand like kind of how to stack this up and how you view positioning this asset competitively?

Thank you, Alethia, and thanks for the question. So as we said, we aren't going to get into any quantitative numbers, but first of all, let me remind you a couple key things. So zimberelimab performed just like the marketed anti-PD-1s. Both of the combinations met our internal thresholds to continue going forward. Our data set is relatively small. It's clearly difficult at this point to compare it to the CITYSCAPE data. Clearly we don't have the maturity of that data set yet, but based upon everything that we've seen, we think that our combination of our anti-PD-1 plus our anti-TIGIT is going to look like any of the other anti-PD-1s plus anti-TIGIT. And we feel that the data that we've generated to date, if it holds up with the triplet, could offer a clear competitive advantage. That's part of the reason why we actually are feeling very good about, having looked early, it's helping to inform how we think about the proportion of registrational studies. Do we actually start to shift towards a greater proportion where we might be taking on triplets? And we feel like the next several months, given where we are in the study, will actually help to see how that doublet and triplet arm actually end up separating with a little bit more time.

And then just as a follow-up. For the triple, I mean, should we expect some sort of kind of further update there by the end of this year, or is it more like kind of in 2022? Since you said it takes like 3 or 4 months, it sounded like.

Thank you, Alethia. So we will present the totality of the data together at that medical conference we were talking about. So full update, both the doublet, the triplet as well as the monotherapy.

And your next question comes from the line of Robyn Karnauskas from Truist Security.

Alethia prepped me well, so let me just ask this clear question. So it sounds like what you're saying is that you -- both combos met the internal marker going forward. I think, going in, I think, I even have written from you all that like 50% or above would be good. And so when people say that encouraging -- you used the word "encouraging." They're worried that, that doesn't mean that, that 50% bar was met. Can you make people feel a little bit better about the word "encouraging?" That's the first question. And then it sounds like this combo -- making the statements around the combo is a little bit of -- or a lot more optimistic than what people, we were thinking. Maybe give us a little bit more color around is Gilead looking at that combo as well. And you said that basically they're going to make the decision end of the year. That means they're going to get data in a few months, because it's they have 2 to 3 months. When will they get that data? Could the opt-in go later than that if they have to wait for more data? So I know there's 2 questions there. Basically make people feel better about the word "encouraging." And second, a little bit more granularity about what that opt-in qualifying mark is.

Thank you, Robyn. You know how my brain can't handle -- That was a lot of words, but I think I got the totality of your question. So first of all, what does encouraging mean? One thing to get -- that we want to make very clear going along -- so during 2020, as we evolved from an independent company working by ourselves to an independent company collaborating with Gilead, we had a number of things that we've been working on from a communication standpoint. And what I would say is, at this point, we've gotten to a situation where we're being very highly aligned in how we communicate on the same topic. And one of the things that we've talked about is we don't want to get quantitative about this at this point. With that said, what does encouraging mean? So encouraging reflects a number of things. First of all, that zimberelimab performed as the marketed anti-PD-1 antibodies; second, that both the doublet and the triplet met the hurdles that we've talked about wanting to see for us to go forward. Over the course of the year, we've used the metaphor thumbs up, thumbs down. Gilead has communicated these data to give us a thumbs up. We want to pursue as we've planned. We're continuing the planning with our trial. To your point about the triplet and how should we contextualize the triple based upon what we already have said about the doublet: Our look, we believe, is similar to Gilead. And that's why we're looking to have this additional few months period. If those data continue to hold up, I believe that Gilead shares our enthusiasm of that the triple could represent a truly differentiated opportunity from the other doublets that the rest of the world is pursuing.

Quick follow-up, if I can. So for 308. A lot of questions, obviously, going into this around Fc enabled versus Fc not. Does your decision to continue on with this trial and continue on with all -- with the pivotal trial have anything to do with that the current drug is sufficient? You don't need Fc enabled. Or how do we view the Fc enabled versus Fc not today? Can we learn anything from this press release? And then I'm done. Sorry.

No, thank you, Robyn. One thing I should really make a point because -- and maybe this -- or helps with both yours and Alethia's question. One of the other things we talk about is that we weren't simply looking at a single numerical number. We've really looked at the totality of the data. For example, the spider plots. I mentioned the time to respond. Virtually all of the patients, by far the vast majority, in both the doublet and the triplet remain on study. So all of that encourages us as to how things will play out with greater data maturity. Coming back to the 308 question, you should keep in mind that's exactly the same way we've been positioning that for ages. Basically we've been moving it along, recognize the opportunity that -- and the optionality that it provides us and certain other settings in hematological malignancies. And we're going to be very aggressive about that. In fact, the data and the speed of which that's enrolling is awesome for us and we're excited to have both of those. In terms of Fc versus non-Fc, I think we'll start to switch the narrative or at least from -- coming out of our mouth on that question. As you know, it's something that probably is never going to be proven unless someone literally runs an anti-TIGIT plus the same anti-PDX side by side, but based upon the totality of the data that we've seen to date, we feel we have an anti-TIGIT antibody that looks like an anti-TIGIT antibody. And we don't expect there to be any differences between that Fc or non-Fc function.

Congrats on hitting this milestone.

And your next question comes from the line of Umer Raffat from Evercore ISI.

Maybe 3 quick things, if I may. First, I know there is a lot of focus on the 50% threshold for the combo arm. Let me go to the other side of the spectrum, on the PD-1 monotherapy. And I feel like there's a fair amount of spread even for the commercially marketed PD-1s; for example, -227 for Opdivo's and high 30s, whereas KEYTRUDA and KEYNOTE-024 is in the mid-40s. And as we think about those 2 numbers in the context of 50%, that makes one of them look like only a 5% spread versus combo. The other one looks like a double-digit spread versus the combo, so I'm trying to interpret that in the context of, Terry, how you described the doublet being sufficient to continue and triplet being something you were intrigued and excited about. I would really appreciate any thoughts there. And then if you could also remind us what the median duration of follow-up is.

Well, thanks so much, Umer. And I'm actually going to let Bill Grossman take that question.

Umer, thanks for the question. As you kind of referred to, there are a number of different studies out there for monotherapy PD-1s, PL1s -- and PD-L1s [ in that population ], but if you look at the majority of those studies and primarily KEYNOTE-042, IMpower110 as well CheckMate -227, the range really is much closer, I believe. I think it's really, I mean, in the 35% to 40% range. KEYNOTE-024 did have a higher rate -- response, I think, around 45%, but we think we're right in the typical range of all PDX antibodies out there. So our performance is right in line with what we had hoped. And we were hoping that we wouldn't see necessarily an underperformance in a patient population, which was not the case so far.

And your next question comes from the line of Geoffrey Porges from SVB Leerink.

This is Na Sun on for Geoff. Congratulations on the progress. So are the study responses confirmed responses? And are they investigator assessed or independently verified? And then a follow-up is what is the longest duration of response in the combination arm.

So I'm going to let Bill Grossman handle those questions as well.

Yes. So right now we're just primarily looking at unconfirmed response rates, and they're based out of investigator assessments right now. We will be looking at with longer follow-up in the study to be -- and confirming all those responses. As Terry mentioned, it's early days in the follow-up period, so we will be looking at that as well. And your last question was around the response. Is that correct, the longest responder?

Yes, the longest duration of response in the combination arm.

Yes, longest duration is looking at -- correct, yes. So we're taking numbers between the 2 different combinations arms. The longest durable response we have is right around 8 months at this time.

And your next question comes from the line of Salveen Richter from Goldman Sachs.

Could you just help us understand what the gating factors are for Gilead to opt in here on this program?

Yes. Thanks, Salveen. I think the -- obviously you'll have a chance to ask them, but I think that part of this is they have the opportunity to let the data mature a little bit more. They're able to do it. If you ask me, it's actually smart. The data looks quite encouraging to us, but I think that they find it valuable when you look at how these data are playing out, as we talked about, the time to response, the spider plot. It gives them a few more months to just see if everything holds up the way things are looking today.

And your next question comes from the line of Mara Goldstein from Mizuho.

Just I had a question just on the -- a, the presentation of data and where you anticipate that may occur, as we're close to the second half of the year and much of the clinical -- bigger clinical conferences are really coming up very quickly. And then secondarily, I'm just curious given that you're approaching that Phase II dose in ARC-3 later this year. By the time you round out the year and have a data packaged for Gilead with dom, where will you be with the 308 study?

Obviously, Bill, you can probably handle both of those. Why don't you go ahead?

Yes. For the first question. Look, the medical conference is still to be determined. Once we get more data maturity, the plan is to submit an abstract by the end of the year. As you point out, there's not that many conferences towards the end of the year, but we'll be looking at our opportunities for presentation at the end of the year once we see the data -- the next data cut. And then the sort of like AB308 package: As we mentioned, as Terry mentioned, we're in our third cohort for dose expansion right now. We do expect either second and third cohort expansion to be our recommend dose for expansion to go forward. We are exploring additional side dosing combinations in that study as well as expansion cohorts. And then we'll be looking at presenting that study, again, at a medical conference as soon as we can in the near future.

Okay. And if I can just also ask just on like the doublet versus the triplet combination. And I understand you're not necessarily giving specific data, but could you maybe help us understand what you are seeing in terms of inflection differential between those 2 different arms?

So all I will say at this point is that both look very interesting, but the triplet certainly looked like it was doing something more. And we think we simply will -- with a few months more data maturity, we'll also bring in not only more data but more patients. That will help us to ascertain whether and how meaningful that difference is.

And your next question comes from the line of Yigal Nochomovitz from Citigroup.

Just following on the last question, Terry, regarding the double versus the triple. And I know you're not giving numbers out, but can you please confirm -- or sorry. Can you -- sorry. Can you confirm that the double and the singlet was comparable in terms of the separation -- rather the separation between the triplet and the doublet? Did you see a similar separation there between triplet and doublet, and doublet and singlet?

Well, we don't want to get into specific quantitative numbers, but certainly there was a separation from -- both the doublet and the triplet were separated from the singlet.

Okay. And then another question: Just you mentioned that the time to response was over 3 months for all the arms and that the study would benefit from greater duration. And then you said that 25% didn't respond till the second scan. I'm just wondering if there was a slower -- if the time to response for the triplet was in fact a little faster than the doublet.

I think it's way too early and way too few patients to start to make those type of comparisons. I will just say, if you look at the spider plots, they look like you would expect for an immunotherapeutic regimen. Too early to try to call out differentials.

And your next question comes from the line of Peter Lawson from Barclays.

Just as we think about the data being presented, could that be presented ahead of Gilead's opt-in? Or would the Gilead opt-in delay data presentation?

We probably can't get that granular with any knowledge we even have about exactly how that's going to play out, so I will just say that we don't have any -- enough fidelity to make a call on that. I'm sorry, Peter.

Got you. And then could you remind us about the opt-in time and event for Gilead around adenosine?

I'll let Jen handle that one.

Yes. So we have not disclosed what the triggers for the adenosine programs are, nor are they. And then I just want to come back to Salveen's question as well that also related to the opt-in particularly. It's patient number-based. So we have not disclosed what that is. It's been redacted in the agreement, but it's patient number-based. It's not based on meeting some sort of efficacy [ framework or ] anything along those lines.

Got you. And then just final question, just around the spider plots. Do they look encouraging, pointing downwards? Or is there any kind of worry around the...

[We -- wonderful ], Peter, the spider plots. And that's one of the things we talk about, is -- or were talking last year, that -- we got asked sort of qualitative questions around the data. That's why we're excited. The spider plots are the telling story. And the pattern is very, very, very consistent. That's why we mentioned so many patients remain on study. That gives you how those spider plots are looking today. Give it a few more months, where I think they're going to look pretty beautiful.

Perfect. Congrats.

Thank you, Peter.

And your next question comes from the line of Zhiqiang Shu from Berenberg.

Congrats on the progress. I have a few questions to ask. First is can you comment on the PFS. I know it's probably immature, but have you seen similar data around 5 months, 6 months as compared to CITYSCAPE? And the second question is around safety. I think, Terry, in your prepared remarks you mentioned, without depleting Tregs, you may have better safety profile. I guess, have you noticed any better safety profiles compared to other Fc-enabled TIGIT antibody? And the third question. I think I want to get a bit more clarity on the opt-in option from Gilead. I wonder. The decision for them to delay to make decision to opt in, would that be associated with the fact that they want to see more data from your 308 data package given these 2 modules are bundled?

Well, thank you very much for those questions. So I'm going to let Bill answer the PFS and safety questions, and then I'll comment on the opt-in.

Thank you. So from a PFS perspective, that's a [ time achievement] end point that we have part of our co-primary study. However, at the current maturity of data, we don't have PFS data, so that will be forthcoming as the study matures with further enrollment as well. So no PFS at this time. As far as safety, again reiterating what Terry had said before, there is no unexpected safety signals across any of the 3 arms. The current safety profile really appears consistent with all the known checkpoint inhibitors or combinations that have been presented to date for TIGIT antibodies. So we haven't seen anything that is unexpected across all 3 arms.

So coming back to the opt-in question, actually I'd like to frame that. So interestingly -- this is not actually a delay in the opt-in. And in fact, well, this is -- would have been like an early opt-in. And so Gilead has the time to look. It's smart. It shouldn't be interpreted all in the context of AB308. So I think Gilead and us are both motivated to go as fast as possible. I should point out that the way this is all staged and how the timing works out is that, both the ongoing studies and our planning, it won't be affected by when -- if they choose to opt in prior to the end of this year. It won't slow us down in anything that we're either doing or planning.

[Operator Instructions] And your last question -- your next question comes from the line of Jonathan Miller from Evercore ISI.

You mentioned a couple of times that the data is not mature enough to get a real sense of some very important metrics and that we should wait for that data to mature. That makes sense. When we see this data at a medical meeting later this year, at that point, do you expect to have enough follow-up to be giving us not just detailed ORR estimates but to have a meaningful sense of PFS curves and separation there as well?

Go ahead, Bill...

Yes. It's a little bit hard to say right now, so I'd say that's a little bit TBD but from a PFS perspective probably not because that -- if you look at the CITYSCAPE data, they're beyond 6 months mark, which I think is where we need to be across all of our enrolled patients, but we will be able to at least express and show information on some of the durability of the responders. And so that type of data, we'll be able to definitely include in the presentation around our response rates.

Great. And at that point -- just of course, obviously this depends on enrollment and follow-up, but about how many patients would you hope to have by that medical meeting later this year?

So we've made it clear we're not actually disclosing anything quantitative about enrollment, but I'll tell you that the study is enrolling quite well. We've got over 60 sites up. It's doing well.

[Operator Instructions] And you have a follow-up question from Robyn Karnauskas from Truist Security.

I -- sorry. I'll ask another one -- 2 questions, just thinking about the competitive landscape. We've got a lot of people going after TIGIT. How much of your press release and lack of color is given around the thought to other people reading the data? And second, you've got a lot of catalysts for adenosine in the back half of the year. Could you set the bar for people who don't seem to give much credit to adenosine? What would be the bar for the next readout you have for adenosine? And why people should care about that data set.

Sure. So certainly there are elements of keeping things, well, a little tighter now due to the competitive nature of things. We feel we need to give qualitatively, let the world understand what we're seeing, but I think we recognize this is a quite competitive field. I mean particularly as you start to think about these differentiators like a triplet that start to move you away from just how fast you can execute on a doublet in a given setting. I'll remind you that's what we've been saying all along, is the unique aspects of Arcus. And so we want to give the color as to what we're seeing, but we will be a little tighter on that information. I also think maybe Jen can comment a little bit on how the rest of the year is going to play out insofar as the adenosine programs, which by the way we'll -- we're extraordinarily excited about those. And in fact, we've been saying for ages the triplet is a key aspect of what we were interested in. And so I think it's a -- very logical just to position the question. So thank you, Robyn, appreciate your questions.

Yes. So the most likely next data set will be the ARC-8 update. So that's for AB680, our small-molecule CD73 inhibitor in first-line pancreatic cancer. So we've submitted an abstract for presentation at a conference. So we hope that will be in the fall. That's something we're very excited about. As I think we've mentioned in the past, we've also started the second-line cohort in that study, sort of continuing what we started with ARC-8, but that will be a more mature data set from both the dose escalation data we presented earlier as well as the initial data from the expansion cohort. So that will be 20-plus patients in that expansion cohort. We also hope to have, as we've talked about, by the end of the year, the presentation of the ARC-7 data at a medical conference, but that only includes data on the triplet arm, which I think we've tried to convey today we're very excited about. So those will be the big events in the remainder of the year. And then the first half of next year, we should have randomized data from all of the studies that Terry pointed out; and particularly ARC-6, which is a study evaluating etruma in metastatic prostate cancer, which we're very excited about. It will be randomized data and -- yes, randomized data; and again for etruma plus chemo in EGFR-positive lung cancer.

And your last question comes from the line of Mara Goldstein, a follow-up question, from Mizuho.

Just quickly. One of the things that you guys have said is that the ability to do this interim allows you to make any adjustments to ARC-10. And I'm just curious as to, at this early juncture, whether or not you've learned anything that has prompted you to revisit any of your assumptions.

Thanks, Mara. So that's an easy answer. ARC-10 is proceeding just as planned. The one thing that I would like to highlight, though, because I think it's an important aspect of this: This took on a very public and external flavor as we went through the year, but as we always talked about, this was intended as an internal decision-making enhancing tool. And in fact, probably the biggest thing that came out of it for us taking that early look is helping us think about what we might do differently, insofar as these doublets versus triplet. You can imagine we were in full planning mode on a number of doublets. We're still continuing there, but we're building into planning the possibility that we may shift just how we think about doublets versus potential triplets and when we might start to execute on those, pending the data holding up as they get more mature.

And these are all the questions we have. Please continue.

So let me just thank everybody on the call again for joining. We really appreciate your continued interest and all of the awesome questions. So we look forward to maintaining dialogue. Thank you.

And this concludes today's conference call. Thank you for participating. You may now disconnect.