Exelixis, Inc. ($EXEL)

Good morning, everyone. My name is Emily Shutman, I'm a Vice President here on the Targeted Oncology team at Leerink. And I'm super happy to have with me Andrew Peters, Senior Vice President of Strategy and Investor Relations at Exelixis. So thanks so much for being with us today.

Happy to be here.

Great. So yes, maybe we could start out. ASCO GU happened very recently. So I was thinking we could start with RCC and talk about some of the dynamics within the HIF-2 alpha space for RCC. The results from LITESPARK-011 for belzutifan and Lenvima and second-line RCC were just presented. I believe it showed a PFS benefit without a Statsig benefit in survival at the time of the second interim analysis. So all that said, I'm just kind of curious to understand your perspective on this data set and where you think this regimen may ultimately fit in versus cabo in second-line RCC.

Yes. And thanks again for the invite. Happy to be here in sunny Miami. As we were talking about before, the West Coast in California has been quite a nice winter, but I understand from a lot of people that are coming from the Northeast, it's quite a treat. Before I begin, we'll be making some forward-looking statements today. So just as a reminder, please see relevant risks -- disclosures around risks in our business and SEC filings. So yes, ASCO GU kind of a pretty common topic these days. I think from our perspective, the conference overall was as expected. What we mean by that is kind of before, during and after ASCO GU, a lot of the conversation among clinicians is really this idea of is it better to combine modalities? Or is it better to sequence? And the LITESPARK-11 data kind of in our view, and I think what a lot of the KOL feedback has been really kind of is aligned with that. If you look at, say, TKI monotherapy in the second line, followed by, say, belzutifan monotherapy in that third-line segment, the overall PFS benefit is similar, if not a little bit longer than kind of the data that was seen in 11, adding to kind of the additional tox that you saw and the fact that roughly 20% of either bells or Len discontinued. So you have somewhat of a sequence treatment anyway given that a lot of those patients are functionally just on a single agent anyways. And so as we think about kind of the overall impact of the data, it's probably pretty benign to our business. And if anything, where we do see some len bells use kind of in that second-line space will probably drive incremental cabo/nivo use in the front line. Obviously, patients tend not to see individual treatment retreatment. And so if someone say going to get Len/Bel second line, they probably won't get Len first line. And so that's where we'll see some incremental cabo/nivo use. So kind of net-net to us, as expected, pretty overall, not a lot of robust enthusiasm for the data, let's say, given that lack of survival. And then on balance, kind of if it does get used, we'll probably see some incremental share in the first line.

Got it. Okay. No, that makes sense. Yes, we also did some KOL calls on this topic and something that they emphasize that's really important is ultimately seeing that sort of static OS benefit pan out. I guess what are your sort of thoughts on that dynamic? Like what are the -- what's the likelihood that you think the OS could eventually be Statsig? Or what are the pushes and pulls to think about there?

Yes. I mean, really hard to kind of speculate without knowing a lot of the intricacies of the statistical plan, so to speak. But -- so really kind of hard to speculate. But at this time, I'd say it's -- it would be hard to see that they'll have a very, very robust survival advantage, let's say. So if anything, we'll just have to see the data.

Yes. No, that makes sense. I think that's consistent with what we've heard, too, especially some KOLs, I think, are nervous that not really too much works after lenvatinib in that setting, whereas there are more options after cabo, which could be something that could influence the OS as well.

Yes. So we'll just have to see.

Yes, for sure. That makes sense. And then maybe could you speak a little bit about the quality of life and toxicity data that we saw for that doublet versus cabo? I think you touched on it a little bit, but that's something that's definitely risen to the surface in our KOL conversations as being really highly relevant in this setting in addition to survival.

Yes. I mean one of the dynamics to consider again in this question of combination versus sequencing is what's the added tox that you're giving as part of the combination. And then I think importantly, hopefully, it's not being crossed over that the adverse events that you see, particularly around, say, hypoxia and cardiac dysfunction, those can be especially problematic from a clinical dynamic perspective is both for patients and physicians as you have to think about baseline characteristics of patients, hypoxia, how do you monitor that? What's the potential implications of higher-grade hypoxia over a particular amount of time? Is that what's driving kind of that increased cardiac dysfunction. So all those different dynamics are certainly in play, again, contrasted with the sense of, okay, physicians are comfortable with how they currently treat kind of in that second-line space, very comfortable with kind of that third-line bells monotherapy use. On the AE side, one of the things that we've consistently heard is that sequence approach offers patients the benefit of kind of what we think of as like a TKI break with bells. And so for patients who have been on either IO TKIs or TKI monotherapy for 2-plus years prior, all drugs have adverse events, but the difference between, say, a TKI and a HIF is a little bit different. And so having that opportunity to be on as a monotherapy, an active modality with a slightly different AE profile, that's something that we've heard is actually important to patients and clinicians. And so that dynamic, I think, is important to consider in that overall just framework for how patients get treated as part of their journey with kidney cancer.

Yes, definitely. Yes. Sort of adjacent to that, I want to talk a little bit about your collaboration with Merck with belzutifan. So the LITESPARK-33 trial just started. So maybe, yes, you could talk a little bit about where you see zanza and belzutifan fitting into the sort of evolving treatment paradigm?

Yes. So 33, I think, is an interesting study because when we first started talking with Merck, one of the questions that came up and at least how we think about it is cabo is the TKI of the 2020 how do we define zanza, the TKI, the 2030s. And part of that is just how the evolution of patients with RCC will be treated and how that changes over time. So with pembro with adjuvant pembro, especially now that they have an OS advantage with treatment, our expectation is that, that dynamic of the number of patients who will be treated with pembro over time increases and changes. So one of the questions that we asked and we're hoping to answer with 033 is really not what does the market look like today, but what does it look like in 2030, 2031, 2032. And our expectation is kind of that percent of patients who have received prior adjuvant pembro will be much higher. And so 033 is designed to really, again, define what the standard of care is should be in those patients. With that data being so new, the pembro adjuvant data, it's really kind of an unanswered question right now. A lot of things are used, cabo is used kind of more by default than anything. And so 033 is a chance to say, okay, with a -- we think truly best-in-class next-gen TKI combined with HIF like belzutifan, do we have a chance to kind of redefine and functionally answer that question? And hopefully, the answer is yes, but we'll see.

Got it. Yes. Can you comment on, I guess, what percent of patients get adjuvant pembro today and where you see that going?

Yes. It's kind of still somewhat early days. I think our sense is just with the overall survival data, our expectation is that's going to continue to grow.

Got it. Yes. No, that makes sense. And the space, this HIF-2 alpha space is pretty dynamic now in RCC, a number of competitors. There's one trial called the PEA 1 trial, which is testing another HIF-2 alpha agent, casdatifan in combination with cabo in post-IO RCC. I think it might be helpful if you could talk a little bit about the key differences between LITESPARK-33 and PEAK-1 as well as any thoughts you have on the differences of the potential profiles of these combinations based on what we know about these different agents today?

Yes. So I think kind of a statement of the obvious that oncology is competitive. It's dynamic. It's always been competitive and dynamic. And so it's just kind of the nature of how we think about our business. So 033, as I said, kind of is really designed to specifically answer that question of that post-adjuvant treatment. As I understand it, I think Peak 1 is a little bit of a broader study kind of looking at more of just a generic post-IO, which is a slightly different patient population. As it relates to, say, the differences between PA and Bells, candidly, it's probably too early to say. One of the challenges that we all have is trying to understand and tease out early kind of Phase I unrandomized data and how that will play out and influence in larger randomized Phase IIIs. The challenge is that functionally, a lot of times, Phase I patients can be different than Phase III patients. And almost a truism in oncology, you tend to see a degradation of benefit when you go from Phase I to Phase III just because of how those patients are functionally. So really kind of early days there. One of the things I'll say is we're really excited to be working with Merck. There's no better oncology development company in the world than Merck. And so with them operationalizing the 2 studies that we're working together, along with kind of the financial partnership that we have in terms of kind of the co-funding arrangement, it's something that we're really excited for.

Yes. No, absolutely. And then, yes, I believe based on the collaboration agreement, Merck is also expected to announce a second Phase III trial for zanza in RCC. And I know this is in Merck's hands, but just kind of curious if you could comment at all on when this announcement could be or what the trial might look like, how it might be different than LITESPARK-33? Yes, any color you could provide?

Yes. I think at this point, kind of the best way to say it is just stay tuned there. We've agreed with our partners to not talk about specifics of the study until it's up and running. And so excited to talk about it when that is. But for now, I think kind of the best way is just stay tuned.

Okay. Yes, fair enough. Maybe transitioning to non-clear cell RCC. Top line data for zanza [ and nivo ] in frontline non-clear cell RCC are expected in mid-'26. Maybe you could talk a little bit about this opportunity, the trial and just sort of the broader thesis behind this combination.

Yes. So it have had really great meetings so far this morning and obviously been with data later this year have been a topic of conversation. And as I've -- as we've discussed many times in the past, one of the really striking things that always jumped out to me as we were thinking about running this study was the fact that there's never been a large randomized study in non-clear cell. By default, all of the approved agents in RCC are approved for both clear cell and non-clear cell. And utilization is driven by kind of guideline recommendations, which themselves are a bit of a hodgepodge of single-arm unrandomized sometimes single center IST type studies that define how patients get used. So as a consequence, market share and utilization when we can tease it out is a little bit of a hodgepodge and mismatch of everything. And so what 304 does is it provides an opportunity to define with Level 1 evidence with a large randomized pivotal study, what a new standard of care could be. And so it's a chance with zanza, nivo randomized against SUTENT to exactly define that. And so if you think about kind of the opportunity set, non-clear cell from a pure epi basis is, say, 15%, 20% of RCC. And so that's kind of how we're thinking about the market opportunity overall is do we have a chance to really draw a line in the sand and say this is the new standard of care based on robust evidence as opposed to kind of reading through some pretty limited data sets, cabo included that define the market right now.

Got it. Yes, that makes sense. I guess maybe could you set some sort of guidelines around like what level of efficacy you're hoping to see in this study?

Yes. I mean kind of tough to speculate ahead of the data, but somewhat of a statement of the obvious. We want to see robust benefit across response rates, PFS and ideally survival. The way that our experience is the way that drugs are commercially successful is if they define a new standard of care and shift how patients are treated. And so the way that, that's done is to have differentiating data. And so that's kind of the goal.

Got it. Yes. I guess, are you hoping to see better efficacy than what cabo plus nivo or atezo has shown in the Phase Ib, Phase II experience. I think there's some early data sets showing ORRs in the 30s to 40 range, PFS north of 10 months or so, 10 to 12 months. Is that something that you're looking to see?

Yes. I mean, again, that's almost the point of the study is to kind of show the limitations of those small unrandomized data sets. I mean depending on which study you look at, you tend to see pretty wide ranges of outcomes. And so it just inherently shows the limitations of cross-trial comparisons between small end data sets, which themselves have pretty wide error bars. And so I think there's some inherent challenges of saying 304 versus 20-patient data set here is a little bit of an apples and oranges comparison. So I think our goal is to define with a robust data set what that could be because obviously, within non-clear cell, there are a bunch of different subtypes. And depending on how those were enrolled in these smaller studies, it can certainly influence. And so I think our sense is our job is to just define what that is and answer the question definitively, how should patients get treated.

Yes. No, absolutely. Yes. And I guess, I mean, beyond like the efficacy and the fact that this was the first robust Phase III data set in non-clear cell. What are other sort of qualities of zanza that you think would drive usage over cabo in the setting? KOLs that we've spoken to comment a lot on the titratability advantage over cabo that seems to really resonate with treating physicians. So if you could maybe speak to that and some other sort of advantages.

Yes. I mean I think it gets to kind of the heart of why we develop zanza in the first place. So cabo is a great drug. We -- everything we do at Exelixis, we say we kind of do through the cabo lens. One of the dynamics that does come up with cabo is it's around a half-life. It's around 4 days. And so what that translates to from a kind of clinical practice perspective is all drugs in oncology, all TKIs, patients inevitably develop AEs that are needed to manage. The challenge there, if you kind of go by the treatment algorithm is dose hold to resolution of symptoms and then rechallenge at a lower dose. But because of that relatively long half-life that cabo has, that dose hold period can be 10 days, 2 weeks, sometimes even longer. And so that can be challenging from a patient management perspective, that can be challenging from a combination perspective if patients are on multiple modalities. And it can be challenging just from the perspective that the longer someone is not on drug, especially at subtherapeutic doses, does that give the cancer an opportunity to kind of regain a foothold, so to speak. So zanza was really designed to take that same kinase profile that cabo has, which we really think is the sweet spot that differentiates it relative to some of the other VEGF-targeting TKIs and then improve the PK profile. So zanza's half-life is a little under 24 hours. And so that makes it more user-friendly, so to speak, from a combination perspective, from a down titratability, all of those sorts of things. And so as we think about kind of zanza, not only in that non-clear cell space, but across the 7 pivotal studies that we're running, that's one of the most important dynamics because our view is while monotherapy certainly has a role and we're developing and it's a NET and meningioma, if you look broadly at kind of what we're hoping to do with zanza over time, it's certainly around investing in combinations as well.

Yes. No, definitely. That's helpful color. And then I want to talk a little bit about CRC. The Phase III STELLAR-303 trial for zanza and atezo read out positively at ESMO this last year. I think you have a PDUFA in December. Could you maybe talk a little bit about this data set and some of the features of this regimen versus RGO standard of care versus the SUNLIGHT regimen, which is the standard of care in the setting as well?

Yes. So we're really excited about kind of the potential zanza launch later this year in CRC. That third line plus CRC segment is a really interesting opportunity because if you think small G globally, how patients are treated in the U.S., say, given the relative dynamics between, say, KOLs and the community and just how patients are managed over time, share is roughly 1/3, 1/3, 1/3 kind of that sunlight regimen, TKIs and chemo. But importantly, one of the things that historically hasn't been available to patients in CRC is kind of the checkpoint inhibitors and immunotherapy. And so what the 303 data provide is data to show with an overall survival advantage that patients live longer taking the combination of zanza and atezo compared to kind of the current standard of care. And so as we've gone out to the market with research and that's one of the things that jumps out is it's an opportunity to provide patients an option to have treatment with a checkpoint. We all watch the Super Bowl and see all these ads for pembro and nivo and all this stuff. And it's a dynamic that we've observed that patients come in to talk with their oncologists and ask about these checkpoints. And up until now, the answer has been, unfortunately, these aren't available for your type of cancer. So we think, one, that's a powerful message. Two, showing a survival advantage versus standard of care is certainly kind of the gold standard historically in oncology. And we think the 303 data that we showed at ESMO and was concurrently published certainly demonstrate that zanza/atezo should be a new standard of care. And then kind of lastly, on your question on the sunlight regimen, one of the dynamics that, again, has kind of come up in our research is if you think about how patients are treated in the U.S., the vast, vast majority of them will have prior bev use, either once, sometimes even twice. And so that sunlight data, there's actually a pretty stark difference in how patients respond to that combination, whether or not they had prior bev. So kind of overlaying that with treatment patterns in the U.S., there's kind of another dynamic at play, especially when you consider kind of looking at our own data, there's no real difference, say, between prior bev use or not and no real difference between liver met and non-liver met. We just had a robust result across all the different potential stratification.

Yes. No, that's definitely come up in our KOL conversations as well. And I think, yes, that was one of the more positively surprising aspects of that data set for us was the robust benefit in liver mets patients. I guess, could you maybe comment a little bit on any hypothesis you have for why there was such a strong benefit in liver mets patients when historically IO/TKI combos haven't been able to show a benefit in that segment?

Yes. I mean, I think, candidly speaking, it just is a good example and highlights why zanza is different. Taking a step back and why we focused initially on this dynamic between patients with liver mets and not, physiologically, these patients are very different. You would expect if there's metastases in the liver, they have a much poor prognosis, and that's been clear. But one of the things that we've observed in the industry has observed empirically across all of these relatively contemporary studies in late-line CRC is that patients behave differently, especially on checkpoint inhibitors with and without liver metastases. And kind of the most recent study, LEAP-17 that Merck ran looking at lenvatinib plus pembro, functionally speaking, if they would have just looked at non-liver met patients, that probably would have been a positive study. And so that kind of drove a lot of our interest around designing the study to specifically from a statistical perspective, look at those 2 different groups. kind of fast forward the clock and when we read out the top line and then full data at ESMO, we were pleasantly surprised and encouraged that we saw a robust response across both of those cohorts. And to me, to us, that just shows that zanza is actually different. And it's not just that VEGF inhibition that's driving activity, but it's hitting the TAM kinases, hitting [ MET, EXEL, Merck, ] all of the things we think that are unique and differentiated about zanza, kind of myeloid cell biology, tumor microenvironment, all the stuff that we think historically has benefited cabo, say, all that holds true and we think kind of drive a lot of that difference. And so it shows that the combination is very effective and highly encouraged to see that activity across a broad range.

Got it. Yes. No, that makes sense. I guess sort of, yes, related to that, the final OS analysis in non-liver mets patients is expected in mid-'26. I think that one initially wasn't positive at the interim, the liver mets OS was. I guess sort of in that context, how are you thinking about this upcoming non-liver mets analysis? And kind of curious if you think that the benefit in liver mets, does that derisk the non-liver mets to some extent? Or how are you guys kind of...

Well, I think -- so just to clarify, so the data that we presented at ESMO and top line, so that was the ITT population. And so we showed a benefit across all patients. It's inclusive of patients with and without liver metastases. But given the dynamic that I talked about earlier around patients with and without just functionally can be different, the relative maturity, the number of events, survival events for that non-liver met group was still early. And so that's kind of what we're waiting for later in the year. Certainly, kind of directionally, we're encouraged by the early look, but we'll see in the middle of the year. And so that's just kind of a dynamic to consider. But overall, when you kind of squint and look at the tail of the curve and all of that, we're certainly hopeful that we'll see a benefit. I mean, again, a statement of the obvious, one of the better case scenarios or best case scenarios from our perspective is if that data is positive, and we're able to get in a label or show clinicians kind of that benefit in the ITT benefit in the non-liver met benefit in the liver met group, that helps really drive hopefully, uptake in utilization and show that the combination really is a new standard of care in that space.

Got it. Yes. No, that's really helpful. Maybe just in our last minute, talk a little bit about 2026 guidance. I think it's like contemplates 10% to 12% growth year-over-year. Could you talk a little bit about what's driving that in '26?

Yes. I mean, as we talked about kind of in the last earnings call, continued momentum in the base business in RCC and then kind of additional momentum from the net business. So the guidance reflects kind of both of those drivers, which we expect going forward. So kind of...

Okay. Well, thanks so much for joining me.