Exelixis, Inc. (EXEL) Earnings Call Transcript & Summary

Good day, everybody. Thanks for joining us at the Bank of America London Healthcare Conference. My name is Jason Gerberry. I'm one of the biotech analysts at BofA, and I am pleased to be introducing our next company at the event Exelixis and joined by Andrew Peters, SVP of Strategy at Exelixis and coming fresh off the ESMO medical meeting and some recent updates for the company. But I'm going to turn it over to Andrew as maybe short prepared remarks just regarding Exelixis and the latest and greatest, and then we'll jump into Q&A.

Yes. Thanks, Jason, and thank you for the invitation. Glad to be in London here with everyone. So before I begin, I want to say I'm going to be making forward-looking statements. So please see our relevant SEC filings for appropriate disclosures and risks about our business. So for those who aren't familiar with Exelixis, we're a commercial oncology company, with our lead product, CABOMETYX, focused primarily in the kidney cancer space but with approvals in a host of other indications like liver cancer and some smaller thyroid cancer indications. Behind cabo, we have kind of a whole host of other -- both small molecule and biologics portfolio programs led by XL092. It's kind of a follow-on to cabo, cabo with an improved profile, and then XB002 is our first biologic. It's a tissue factoring targeting ADC, XL102, CDK7 and then XL114, it's a MALT1 inhibitor. So glad to be here and looking forward to the discussion.

Good. Good. So maybe we'll start just with some of the updates that occurred at ESMO pertaining to cabo label expansion studies, for lack of a better way of saying it, and then XL092 or the son of cabo program. Maybe just starting with COSMIC-313. I just wanted to confirm one point. The press release did mention future regulatory discussions that will occur after this data set. And so -- and then it did indicate, I guess, included analysis of future OS data. So I just wanted to clarify, are those discussions sort of pending, more mature OS data? Or -- because there was always a question could you file on the PFS primary endpoint for approval of OS data, which, as we know, is very commercially relevant in the frontline RCC space.

Yes. So I guess, just maybe taking a step back to kind of frame the 313 data and really kind of what happened at ESMO and why we're so proud of that data set. It's really a good example of Exelixis being at the forefront of clinical innovation. As Dr. Monty Pal, the discussion framed following presentation of the data, it was really a series of firsts, what that study really was. So the first pivotal study using a doublet, a standard of care, ipi/nivo is an active control kind of the first successful study for a triplet combination in RCC and really kind of does a lot to move the standard of care for patients. So specific to your question on kind of next regulatory steps. As we outlined on the last quarter call, Vicki, our CMO, mentioned that our next step is to meet with regulators regarding the data. Obviously, both Toni, during the presentation, and Monty, during the discussion, raised a lot of interesting questions with the data. And so I'd imagine that those are the sorts of things that we're going to have discussions with the regulators next.

And so at a high level, this triplet combination, where do you see it fitting? It's kind of -- I understand the premise, right? Take the best of both worlds in terms of your 2 main treatment options right now in frontline RCC and if a triplet would kind of improve efficacy. So where do you see the combination fitting? Or do you think the story is still as an element of TBD as OS data matures?

Yes. I mean I think, as always, it's going to be data-dependent, and it's always a little bit of TBD, in part, kind of pending the maturation of the survival data, but in part, kind of as we interrogate some of the interesting clinical questions that came out of the presentation. As an example of that, looking at kind of the differences in efficacy between some of the patient populations, whether it's the intermediate versus poor risk groups as well as some of the questions that were brought up kind of during both the presentation and discussion, how do clinicians think about the tolerability and kind of that give and take between the increased efficacy, but also the expected increase in toxicity. And then just looking at things like how does this patient population compare to some of the prior data sets like 214 where the differences saying the number of patients who had prior nephrectomies that kind of could have impacted some of that calculus. And so I think from our perspective, obviously, the next steps, so to speak, with regulators. But really, we see the data -- the study met its primary endpoint of PFS, and we're certainly happy with the presentation and discussion and the reception at ESMO. And it really shows that a company like Exelixis can be at the forefront.

And what was the physician feedback like in terms of the data, and overall, perhaps where and that might fulfill an unmet need?

Yes. So I think that was kind of one of the most fun parts for me being at ESMO, again, kind of being at a company like Exelixis and hearing a lot of that enthusiasm and excitement around really kind of running the right study to try and push that standard of care forward and really help patients. And so I think there's kind of that enthusiasm around these are the right studies to do. These are the novel triplets that need to be interrogated, but it's also about what are the patients that are really going to benefit from a study like this? Is it kind of by -- theoretically, by subgroup looking at the data that, that intermediate group is really kind of where the bulk of the benefit was found and probably the poor risk group was a little bit more of a wash? But also, was the study doing what we set out to try and do? You mentioned before the goal of the study conceptually was to look at the best of both worlds, the best of the IO/TKI, which is cabo/nivo in our view, and then IO/IO ipi/nivo and ask the question, well, what benefit does, say, adding cabo on to ipi/nivo confer? And so one of the things that, again, Monty highlighted during the discussion is if you look at the pivotal data for ipi/nivo, about 20-ish percent of patients have progressive disease as their best response. If you look at 313, it's more than half of that or less than half of that. And so cabo clearly is kind of left shifting that response curve, so to speak, potentially at the cost of, again, as Monty highlighted, depth of response. But again, those are the sorts of questions that are going to be interrogated as we continue to dig through the data and understand was that related to kind of the ipi dynamic or steroid use or all of the things that we've kind of mentioned. But ultimately, it's a data set that we're excited about and really kind of shows that we want to be at the forefront.

Yes. Okay. So as we think about maturation of the OS data on the one hand, it looks like the CheckMate or ipi/nivo arm seems to be performing kind of consistent with what might have been expected based on the prior CheckMate 214 study. I don't know if that was at all a concern about that being a variable and how you kind of model your different interim analyses. But I know the discussant raised the point about just wanting to see OS data. I assume there's a second interim data. And I don't know if you can hazard a guess as to like sort of when we might sort of expect to see OS data mature because [indiscernible] took me to enforce a pretty incredible long-term -- 4-year, 5-year long-term OS durability and a shoulder on the curve. So I guess investors wonder how feasible that is.

Yes. So kind of -- without kind of getting into specifics around when to expect survival, the interim -- the next interim look is event-based. So I can't really comment on when that's going to happen. But just to understand the study a bit more, the first survival look was triggered when the primary endpoint PFS was hit. Then there's a next interim, and there's a final survival analysis. And so kind of we're monitoring and understanding kind of the events as they come in, and that's really what's going to drive kind of the timing of the [indiscernible].

Okay. All right. And then the discussion raising point just about potential overlapping toxicities that might interfere with how the individual components of the combination 313 could be delivered to patients. Just curious your thoughts on those perspectives, if that could impact OS or maybe even from a commercial perspective down the line, that could impact physician adoption of a 313 type regimen.

Yes. So maybe kind of, again, coming back to the Genesis study and kind of the relatively innovative nature and forward-looking frontline is we're starting from a position where there's a high degree of physician familiar with both components of the combination obviously, cabo, both from the media and then the 9ER data set and then ipi/nivo. And so we're starting from a place of high degree of physician familiarity, not only on the efficacy side, but kind of from the tolerability and AE management perspective as well. But again, kind of I'd come back to from a next steps perspective, obviously, speaking with regulators on the next steps, the most important one, but there are a lot of kind of clinical questions that we want to continue to interrogate coming out of the data set. And I think a big one is to really understand what is that patient population that's going to kind of benefit most while also kind of balancing the increase in AEs that we also saw as well. So there are a lot of questions coming out of the study. But ultimately, it's going to come down to kind of finding that right balance in that appropriate patient population. And that's something that, again, came up again and again at ESMO where it's depending on how the data ultimately shape out, it's going to be about finding that right population where that increased efficacy is offset by the tolerability.

Yes. Commercially speaking, just curious your best case, I think there's like, what, 20% of frontline RCC patients who roughly get an ipi/nivo-type combination. Is that sort of the incremental opportunity as you see it as sort of getting a cabo-based triplet kind of ingrained into that patient population?

Yes. So I think the opportunity set for 313 and the triplet is -- the good thing is we're not trying to displace a regimen. It's -- we're really trying to be additive onto an existing regimen. And so I think that's kind of more of the dynamic that we think about it. It's not -- it's more changing practice and say, okay, how can patients derive an additional benefit of as opposed to kind of more of a hard switch, but again, understanding which of those patients is really going to be about -- as we gain more insight on the data kind of understand some of those subgroups a little bit more. That's really going to help shape ultimately the opportunity.

But I guess it would be looking to be added on to a regimen that's kind of helped firm at around 20-ish percent or so ballpark?

Yes. Plus, minus, I think.

Okay. All right. Well, I guess it sounds like stay tuned. And over the next 6 to 12 months, maybe a little more clarity on kind of next steps and pathway forward here.

Yes. Next step, obviously, talk to regulators and then kind of continue to interrogate and tease out a lot of the data. But from our perspective, ESMO was an exciting meeting, kind of a meeting of first for us. And as a company of our size, getting a presidential plenary at a conference like that is really kind of a testament to all that we've done and kind of trying to be kind of the leaders in this space.

Yes, okay. Well, let's maybe shift gears to XL092. And maybe you just want to start with sort of as you think about embarking upon developing this molecule, what you -- what the hopes and the aspirations were for what this drug could be as a next-generation TKI follow-on to cabo and what you feel like we've learned at ESMO that helped sort of shape that vision? And sort of what are some of the outstanding questions still?

Yes. So I think taking a step back, it's worth framing what was the clinical question that we were asking with that study, and ultimately, what are we trying to do with XL092? And it's can we replicate the efficacy profile of cabo with a product or with a compound that has a much shorter half life? And so if you compare the data at ESMO versus early cabo studies, I think it's pretty clear that the efficacy profiles are quite similar. And then on the tolerability side, certainly, it's still very early, but we're seeing encouraging signs there as well. But the -- all of the PK data have kind of played out as expected. 092 is about a 23-hour half-life versus around 99 hours for cabo. And so coming back to the initial question of what are we trying to accomplish with 092 is can we generate a compound, can we create a compound with a cabo-like profile with a much shorter half life? And I think the signal is coming out of the data this weekend certainly suggest that we were successful on that front. Now going forward, I think the question that we probably get most around 092 is what's next. We obviously started STELLAR-303 in colorectal cancer earlier this year. We've announced that we expect to announce a second pivotal study by the end of this year. But I think one of the ways that we really talk about and think about 092 is about breadth of development using kind of all of the data that we've generated so far from cabo as well as the emerging data coming from the STELLAR studies from 092 to really kind of point us in the direction around where are the areas, where are the indications that we can really have 092 is that kind of backbone and then looking at different novel doublets and triplets going forward and kind of think about how we can really shift standard of care and help patients. And ultimately, that's what we're trying to do, is we're trying to kind of shift that PFS, that survival curve rightward and can a compound like 092 or a cabo-like compound really help with that.

Okay. I mean you'll hear from some physicians, right, that the idea of putting somebody on a TKI or a cabo for more than a year, right, that's a bit of a commitment, right? And so some of the classical AEs, right, the fatigue, the things that patients don't maybe necessarily want to deal with on a chronic basis. Are there any specific aspects of the 092 safety data that are emerging that you think are particularly encouraging that might kind of carry its way through to the ultimate kind of value proposition of the therapeutic?

Yes. I mean, again, we're still kind of early days from an overall tolerability and safety experience before we can really kind of start to make some definitive conclusions on what the safety looks like. But certainly, if you can compare the early cabo work with 092, some things do jump out like kind of grade 3, 4 AEs. That sort of thing I'd imagine would be the sort of data that we're going to want to kind of truly understand how that evolves over time. But what I'd point to is one of the experiences that I think we've gained from the 9ER launch has been how important that totality of the data aspect is from both a patient and physician perspective. A lot -- in our earnings slides, we kind of have a graphic there that I think really kind of crystallizes for me what that 9ER data set is, and it's about a balance between efficacy, tolerability and things like quality of life. And ultimately, we're trying to understand what's the opportunity for this compound. And it's really going to be about all of those sorts of things together. So you talk about clinician feedback around TKIs, but it's also balanced with efficacy, but also quality of life is a patient feeling better, that sort of thing. So all of that's important. And as the data continues to mature, that's kind of how -- what we close attention to.

Yes. Yes. I guess we're in the -- an enviable role of trying to make odds and ends of early updates. But it's hard to compare the Phase I data against cabo ultimately because a lot of the data sets that we have for cabo presumably longer duration of exposure. We're looking at higher doses and things like that, but -- versus, say, now a lot of combos are going to basically starting at 40 mg versus a lot of the data started at 60 mg. So how do you kind of think about the data from a dose escalation study and putting it in context and benchmarking it against as you understand sort of the profile of cabo?

Yes. I mean I think you kind of hit the nail on the head, and we're all kind of in the same boat in trying to kind of draw the best conclusions we can and making smart decisions going forward. Without kind of getting into specifics, I think you mentioned the 40-milligram dose and kind of how we settled on that for 9ER. And ultimately, kind of that decision that was made by our late CMO, Gisela Schwab, was probably one of the most best decisions we've made as a company because it really informs how to really get that kind of maximum benefit across all of the data. And it's not just about getting the highest efficacy possible because it's a -- there's a saying in oncology, "you can't benefit from a drug that you're not on." And so that dynamic of it, having to either discontinue or completely come off a drug, is something that we pay attention to. So without getting into specifics, I think it's somewhat of a statement would be obvious that we pay a lot of attention to those. We recognize how important it is, how important it can have, especially in combination. And so we look at a lot of data to help [ inform ] those decisions. And even going back to, as I mentioned before, our experience with cabo gives us great insight into how we can move forward with 092.

I guess it's tricky to answer questions about this because, I guess, it makes it sound like you're disparaging cabo in a way, right? And you're really trying to improve upon cabo, and it sounds like being able to achieve more persistent dosing with 092 is one element of all of this.

There's a lot of elements of it. I mean, again, kind of coming back to that initial question, can we develop a cabo-like compound with a shorter half-life? And then secondary to that, what are the clinical implications of a shorter half-life? Just simply speaking, kind of that ease-of-use perspective, potentially better combinability, potentially better tolerability, these are all things that we want to interrogate clinical trials. But from kind of a first principles perspective, can we develop a cabo-like compound with a shorter half-life? I think the ESMO data would certainly suggest that we've been successful in that front.

Yes. Okay. And you've outlined a number of different tumor settings that some are more established cabo markets, others would be kind of more white space opportunities. Are there any like specific tumor settings where you feel like the ability to go to more persistent dosing is of particular value?

So don't want to kind of get ahead of our next pivotal study that we're announcing by the end of the year, but I think you're right in highlighting or outlining the sorts of areas that we want to get into. I think the way that we framed it, both internally and externally, is kind of more waves of development. So the first study, 303, is probably more of that faster to market, high unmet need-type indication where we saw kind of the clear signals of cabo combinations coming out of ASCO GI earlier this year. And then there are these kind of next waves and next waves beyond that, that, as you mentioned, either kind of the white space where cabo has shown signals from, say, the COSMIC-021 study or emerging signals out of STELLAR. And then are there areas that we know cabo has activity, but maybe we can go earlier? So those are the sorts of things that you'll see us kind of interrogate more over time. And I think the other aspect of this is, are there novel doublets and triplets that, to your point, if it plays better with others, obviously, it would make sense to look at those. And so again, somewhat of a statement of the obvious, I think the oncology world is moving towards more and more combinations. And so having a more kind of user-friendly backbone-type therapy, which we think certainly 092 could be that, that drives some of the excitement that we have around.

Okay. Well, maybe shifting gears to your commercial business and cabo. The launch of 9ER has been a great success story for you guys. Not looking for guidance for next year, obviously, but maybe just talk conceptually about the puts and takes for growth for this business.

Yes.

Glad you caveat it because I think we have that in front of you.

I know. You didn't let me use my favorite phrase is that we're not going to give guidance on guidance. But -- so I think, obviously, the dynamics in the RCC business are market share and duration of therapy, just generally speaking. And so we expect those to be important going forward. But I think about -- or if you think about the cabo business longer term beyond RCC, it's really going to be about clinical data. So we just talked about 313, but we have the contact trials that are ongoing as well. So big picture, I think, the cabo franchise growth is really going to come out of those data sets, understanding the profile of 313, CONTACT-01 in checkpoint refractory, small cell lung cancer that's expected by this year. Cabo plus atezo in checkpoint refractory, RCC also expected by the end of this year. And then another atezo combination in NHT experience castrate-resistant prostate cancer, that study is enrolling as well. And so those are going to be kind of the key data sets and key drivers for cabo long term not only for kind of 2023, but beyond.

Yes. Because a lot of those feel like 2024, right, if you're getting like the data sets in the second half with the regulatory. I mean -- so I guess what I'm asking is, do you feel like there are still room for share gains? There's this whole concept when you have a drug with 1.5 years duration of treatment that you get patient stacking from when your -- and if you kind of reached a point where you probably start to lap those comps with the patient stacking.

Yes. Again, the more specific I get with kind of those dynamics kind of...

The electric shock.

The closer I get to giving guidance, and I think I probably want to stay away from that.

Okay. Fair enough. Yes. Ultimately, it's a question of market share dynamics, and maybe if you can kind of speak to what's going on in RCC, either kind of present state time or kind of over the past year, where ipi -- or cabo/nivo competitively is in the marketplace gaining share? You do have a competitor that's kind of rolled out a new combination. Is that largely cannibalizing its old combination approach? And just sort of where are the dynamics? And where is the sort of cabo/nivo combination been able to gain share?

Yes. So I think as we laid out on the last kind of quarter call, what we've been seeing -- you kind of hit the nail on the head in terms of overall dynamics. For the most part, cabo/nivo has continued to gain market share, not really at the expense of, say, ipi/nivo, but probably more from other IO/TKI or just IO -- or TKI monotherapy regimens. And then kind of second to that, you mentioned our competitor with [indiscernible], that's probably more cannibalistic at least based on the data we've seen to existing kind of pem/axi combinations. And so oncology is a competitive market. RCC is a competitive market. It's been a competitive market since literally day 1 for cabo, but it's one that we're continuing to kind of go out and make sure that our message around the totality of the benefit, the balance of the data is understood by patients and physicians. And as I mentioned before, that's been a particularly strong message so far. We want to make sure that we're continuing to get that message out there and get that heard. Again, somewhat of a statement of the obvious, if you look at 2020 versus 2021 and now, the difference in a lot of that revenue is kind of that increased market share and increased duration. So we have a pretty good sense at least internally around what each of those metrics are worth and what we're continuing to track.

Okay. Now I talked about like a ton, and they pale in comparison to RCC, but DTC and second-line liver, right? So with second-line liver, there was always this notion that IO-bev was going to open the market for systemic therapies, create like sort of kinetics of patients to your second line where you're standard of care in liver, and then DTC is a newer indication. So are these -- again, they're probably both niche, but are these sort of incremental opportunities that are in front of you, which you really have yet to really realize?

Yes. I mean, so we haven't really kind of broken out the revenue split between each of the indications, but I think it's suffice it to say that RCC is kind of the majority of the business, and it's going to continue to be the majority of the business going forward. HCC, there are a lot of dynamics, increased patients, but are those patients as amenable to second therapy -- later line therapies? Unfortunately, HCC is a pretty devastating disease, and you have relatively frail fragile patients. So that kind of [indiscernible] can offset some of it. So I don't want to get into specifics around HCC growth versus DTC growth versus RCC growth again because that'll kind of more start to bleed into the guidance question. But I think it's...

It's more plus than minus.

Yes. So I think it's safe to say that from a big picture perspective, RCC is going to continue to be an important driver for us in the near term. And then as I mentioned, longer term, it's going to be about the data we're going to continue to generate.

Sure, sure. Okay. Just a couple of questions on the patent front. So I know the company has been reluctant to talk about litigation details, I'm very well versed on the [indiscernible]. My question is more just around the time lines. You've got one case that's already been litigated. We're expecting a ruling probably in the fall. There's a second case, my understanding is sometime in the middle of 2023 about from a trial standpoint, and that pertains to broader crystal structure type of patent. And then the matter got maybe a little bit more complicated because now there's another patent that was asserted because the company had to amend there -- and so that pertains to a formulation patent. So my question is does that -- do we know if that more recent patent, does that get rolled into the second wave litigation part of that time line? Does it create a third bucket of -- just trying to mentally compartmentalize where we are with all these legal matters because it seems like it's a case that's gotten more complicated.

Yes. So very much appreciate kind of the -- not having to break into jail, so to speak, and try and talk about too much of the existing litigation. But to your specific question, so the 2, MSN 2 is what we call it, those 2 cases have been consolidated into a single case that is going to trial next October.

Next October. Okay. So the formulation got rolled...

The formulation got rolled into.

Yes. Okay, okay. And then the second filer was Teva who had their matter stayed, I believe, while the MSN case was going on, and that matter is still stayed or maybe I'm getting a little too deep into the legal weeds here.

Yes, that's probably a little bit too much in the weeds that -- to kind of talk about now. But I think kind of from a big picture perspective, we remain confident in our patent position, and we continue to defend it vigorously.

Yes. Okay. Yes. All right. Yes. So then maybe we'll jump to XB002. This is a program that's pretty of high interest in your pipeline, I think, from investors, probably of most interest. We have a medical meeting update now scheduled for triple, I believe. So maybe you can talk about 002. And I know you guys have talked a bit about the value proposition and some of the shortcomings of the existing tissue factor, ADC. But maybe the highlights as you kind of see 002, and maybe to the extent you want to frame, what's expected at triple?

Yes. So kind of, as you mentioned, abstract titles are out for the triple meeting abstracts. And then ultimately, the data will be presented next month. But really kind of maybe the best place to start is to come back to what's the question that we're asking with this program similar to how I framed 092. So the question that we're asking with 002 is can we take kind of a validated ADC target and further optimize kind of those components of ADC. So if you think about conceptually, it's an ADC, it is the antibody component, the linker component and the warhead or payload component. And so what 002 has done is looked at tissue factor, which we think is a very attractive target from a expression perspective both -- where it's been validated in cervical cancer, but many other tumor types as well and say, okay, can you take the antibody and optimize it? So relative to the approved product from our colleagues, 002 antibody is noncompetitive with Factor VII. So if you think about the role of tissue factor in the coagulation cascade, what a noncompetitive antibody could theoretically do is improve kind of the bleeding risk profile of that. And then looking at the difference between kind of the more standard MMAE versus what 002 uses, which is Zymeworks' ZymeLinks' modified auristatin, what are the potential benefits from that as well? Don't want to kind of get ahead of the data expected next month, but really kind of the goal here is to understand with that optimized profile, what does that confer both clinically? And what are the sort of things we can look at like amount of intact ADC versus free drug that play an important role in the overall profile of an ADC? Free drug is generally what drives a lot of the, I'm going to say, off-target toxicity. And if you think about what the -- one of the goals of an ADC is kind of more targeted delivery of a potent cytotoxic. So the more free drug you have, kind of the less targeted that is. So certainly, something like that is something that as we're developing 002, we pay real close attention to.

Do you think it would be -- in your mind, as you kind of think about this -- is free drug something where you think you'd be able to even compare to the competitor program or just be obvious from the clinical symptom profile of this and then cross-trial comparisons will ensue? Just thinking about...

I think a little bit of both. It's really going to depend on the data, how mature it is, how many -- all of -- it's -- we all want to draw definitive conclusions as early as we can. But I think ultimately, it's going to be about how the data set emerges over time and if those in initial hypotheses that we have around what's the benefit of an optimized ADC that's more stable. Ultimately, it should ask the question or answer the question, is it differentiated?

Yes. And there was commentary maybe back in the summer about not having observed leading it up to a certain point. This is open label. So I imagine that the company has some visibility to that. Can you just remind me maybe what's been said in the past?

Yes. Probably just don't want to get ahead of the data next month. We'll see it.

I wasn't sure if there was explicit commentary like, call it, last summer, but that's okay. Realize if I put you in a tough spot. Okay. I guess just dose escalation, presumably, you're going to have sort of finite follow-up here. I guess do you have confidence that the safety profile will be something that we'll be able to get more confidence around? It's kind of a question of onset of AE, right, and how you understand the onset of AE to be.

Yes. And I think, again, it's really going to be about how that data emerges over time and how, what, when, where, kind of don't want to get ahead of that. But I think it's that kind of totality of the emerging profile that we're going to be paying attention to. I mean if you look at just from a development perspective, we've announced expansion, not only as a monotherapy, but in combination as well. So clearly, all of those aspects around tolerability play into its profile as both a monotherapy, but what's the potential benefit in combination also.

Yes. Okay. And then maybe just the market opportunity for something like XB002. I think cervical cancer is one of the more validated opportunities. But is there an opportunity to go more broadly with the approach? Or would it be other ADC approaches within the earlier-stage pipeline as you think about possibly getting more invested in ADCs?

Yes. So there's kind of 2 important dynamics in that question. So I think on the first, one of the things that attracted us to tissue factor in the iconic program in the first place is the breadth in really kind of the large number of tumor types where tissue factors is overexpressed. So clearly a target that we like from an ADC opportunity perspective. Obviously, validated in cervical, but I think one of the questions that I think we're all trying to ask is, with an improved optimized profile within theoretically or potentially improved TI, can you expand that into additional tumor types? And we're interrogating that in kind of our expansion studies right now, both as a monotherapy and combination. But the second part of your question is kind of more on our ADC strategy in general. And I think that's one that's perhaps a bit kind of underappreciated in terms of what Exelixis is doing in ADCs. And I think Peter Lamb, our CSO, has done a good job over the last several earnings calls and other forms to kind of frame that. But ultimately, it's really about accessing kind of our suite of technology partners and optimizing the antibody, the warhead, the linker, all of those kind of different pieces of the ADC to look at targets where we think the underlying biology makes sense to interrogate via ADC and what is that right combination of optimal antibody, optimal linker and specific type of payload that should work best in that. So XB010, that's our next ADC that we have announced that targets an oncofetal antigen called 5T4, that's something that we're certainly excited about and is very representative of how we're thinking about the ADC world going forward and how it's a really good combination of everything that we thought at Exelixis, starting with kind of small molecules going into biologics, continuing to build that out and really optimize it.

All right. Great. Well, we're past our time. So I want to thank you, Andrew, for your time and your insights and updates.

Yes. All right. Thank you.

Great.