All right, we'll get started here. Last fireside of the day for me, but very excited to have Exelixis here. We have Andrew Peters, SVP of Strategy from the company. Andrew, great to have you. Look forward to the discussion here.
Yep, looking forward to it. Thanks for the invite.
Yeah. Maybe just to start off, you know, kind of give us kind of high-level, you know, status of the business, kind of update, you know, what's going on, you know, kind of commercially, but also, you know, within the pipeline. Then we can kind of drill down on some of these more specific questions that we have.
Yep, sounds good. You know, as I said, thanks for the invite. Glad to be here. Great set of meetings this morning. A lot of good, engaging conversations. Before I begin, I'm going to be making some forward-looking statements today, so please see relevant risks and disclosures in our regulatory filings with the SEC. Yeah, so cabo, you know, it's a pretty exciting time for Exelixis. Pretty rare that you can have two positive Phase III studies in the same week.
Mm-hmm
With CONTACT-02 and then the CABINET study. And so really good to see kind of that momentum. And most importantly, you know, ultimately, we're in the business to help patients, and those data are really important from that perspective. So cabo itself, you know, it's a global now, you know, billion-dollar franchise. I think we did a little over $400 million in the U.S. for Cabometyx this past quarter. Guidance for the year, $1.575 billion-$1.675 billion in U.S. product revenue. So certainly a rarity in biotech these days.
You know, one of the things we framed on the first quarter call is cabo is actually the most successful biotech-launched oncology drug since it was approved in 2016 or so. And that's something that we're especially proud of. That's something that we want to continue to execute well on, and then as we look to our clinical stage pipeline, really use that cabo as the gas to invest in the Exelixis engine. So looking at kind of our next gen VEGF-R TKI, zanzalintinib or zanza, and then XB002, the tissue factor targeting ADC. So it's those two programs and then kind of a lot of stuff coming up behind it from our internal portfolio, and then being opportunistic on the external innovation side from a business development perspective.
You know, firing on all cylinders, going, going forward, and just an exciting time to be at the company.
Awesome. Well, a lot to unpack there. So maybe, you know, starting with cabo, I guess, you know, as you said, great launch, you know, continued growth. Like, but where do you kind of see the growth trajectory from here? And I guess, again, is the incremental growth, you know, kind of more penetration into existing areas, or again, really coming from, you know, what you were just kind of talking about, like CONTACT-02, for instance, and some of these other opportunities. So, you know, how do you, how do you kind of—if you put the growth kind of drivers in separate buckets, like, what are those buckets?
Yeah. So I kind of don't want to specifically get into what drivers of each are. But you know, what I can point to is something like the ASCO GU data. We presented a 44-month update of the 9ER study. That was a pivotal study looking at the combination of cabo and nivolumab in frontline RCC. And as that data continue to mature, we think it further differentiates it versus other IO TKI combinations. A lot of the messages coming out of that data set, a lot of the conversations that we were having with physicians at the conference, are really around kind of the totality of the data, the balance of the data. It's not just about efficacy, it's about tolerability, it's about things like quality of life, it's about things like discontinuation rate.
You know, unsurprisingly, one of the things that we think are so compelling about the data set is, as that overall survival signal continues to remain steady, we think it really has to do with the fact that the discontinuation rate is so much lower. And so, you know, unsurprisingly, patients benefit from drugs that they're on. Patients don't benefit from drugs that they have to come off of because of tolerability issues. And so that's one of the key kind of differentiating aspects for the, you know, the cabo combination there. And then, you know, you rightly pointed to the opportunities for new indications going forward, whether it's the, you know, neuroendocrine tumor side and the, the CABINET study or, you know, the castration-resistant prostate cancer from CONTACT-02.
You know, those are certainly two data sets that we're quite excited about and, you know, looking forward to sharing those data because, you know, ultimately, we're in the business to help patients, and we think those data will certainly help inform, you know, that.
Got it. Yeah, we'll touch base on that in a little bit. I just want to understand, you know, in terms of how you think, like, at least in renal cell with cabo, obviously, you have a really solid toehold there, you know, solid franchise within kind of the treatment landscape. How do you think about competition in the future and, like, you know, kind of other, you know, emerging therapies, you know, in that space and kind of the moat that you've created around cabo and the durability of that moat?
Yeah, there's, there's really kind of two key drivers there or two key dynamics there. You know, the, the first is ultimately the data are what drive utilization. And, you know, just talked about the, the 44-month update, for METEOR, and, you know, that's really kind of the anchor that really drives a lot of the utilization. Beyond that, it's about being innovative. I think if you look at, our two most recent studies in RCC, first, COSMIC-313, that's the triplet combination of cabo plus Ipi/nivo, and we compared that against Ipi/nivo. We've guided to the next interim overall survival to occur later this year, so we'll have an update there. And then CONTACT-03, that was in a second-line population, looking at the combination of atezolizumab plus cabo versus cabo.
So Exelixis, as a company, has really been on the kind of forefront of clinical science in RCC, and, you know, it really makes me proud as an employee to hear kind of that messaging from the podium at a conference like ESMO in a plenary, to really have our company, you know, small mid-cap biotechs really called out as being kind of at the forefront of clinical science there. And that's something that we're going to continue to do, and as we look forward to zanza, you know, certainly we have the opportunities to look at novel doublets, novel triplets. You know, that's the sort of thing that we'd expect to do going forward.
Understood. Gotcha. Yeah, so back to, to CONTACT-02, you know, obviously maybe just kind of, you know, put the results, you know, or kind of the, the opportunity into context of where, you know, where you can really focus in, you know, in prostate and kind of, I don't know, quantify that opportunity for us, whether it be patients or kind of like how you think about the market size there.
Yeah. So this is an opportunity we're quite excited about. Certainly, Exelixis, as a company, and cabo has kind of a long and deep history in the prostate space. But ultimately, what you know, the positive top-line data on PFS show is that this is a chance to have the first novel combination of a TKI plus a checkpoint inhibitor in this really high unmet need population. So, right now, the current standard of care in this setting is really a second NHT or novel hormonal therapy, and unfortunately, kind of that second NHT isn't particularly effective. One of the reasons it's the most commonly used agent, though, is the alternative is generally chemotherapy.
If you think about the kind of later-line prostate population, they tend to be older men, they tend to be relatively frail, and so this ability to delay progression onto chemotherapy is actually a pretty substantial benefit. So when we looked at kind of that opportunity set, what CONTACT-02 was, was really a chance to ask kind of the cleanest question in a really high-end unmet need population. The question being: In a patient population with visceral metastases or extra-pelvic adenopathies at baseline, does the combination of a TKI plus a checkpoint inhibitor, you know, improve disease? And on a RECIST 1.1 basis, on a progression basis, that answer, you know, was, was positive, and-
Mm-hmm.
you know, we're excited to share the data at a point in the future. And so that's kind of really how we see it, is there's, you know, some specific nuances about CONTACT-02 that we think really kind of clarify the opportunity set.
Mm-hmm.
It's just a high unmet need, given the fact that, you know, frankly, the second NHT isn't particularly effective.
Got it. I mean, I know there's some questions of whether you can get approved on PFS or you need overall survival. I mean, any kind of updated thinking there? And I guess, you know, obviously, the data just came out, so, like, obviously, there's probably going to be some, you know, discussions with the FDA on that, but, any timing on that?
Yeah. So as we talked about in the press release, we look forward to having discussions with regulators at the appropriate time. Regarding kind of the PFS overall survival dynamic, you know, as a reminder, in the press release, we did highlight the fact that we did see, you know, differences in overall survival.
Yep.
Data is still obviously immature-
Yeah
Didn't hit the threshold for statistical significance, but looking forward to kind of updates on that in the future. As it relates to kind of PFS as to an approvable endpoint, again, I'd kind of point back to, you know, while there are examples of agents that have been approved on PFS, you know, really kind of that specific population of the visceral mets plus extra-pelvic adenopathies, I think, is. It, it really gives the cleanest test of whether or not the drug's having an effect. If you think about kind of a lot of other agents that look at more of an all-comers population, one of the challenges with a bone-predominant or, you know, bone-only disease is that measuring progression can be complicated, you know?
Mm-hmm.
Whether it's the challenges with bone scan or, you know, PSA, it's a challenging measure to understand how someone's disease is actually progressing. And so by focusing on this specific population, that's why, you know, it gives the cleanest test-
Okay
Of the hypothesis of does the combination have an effect or not?
Gotcha. Okay. Yeah, and then do you, do you want to talk about CABINET and just kind of, you know, I mean, obviously another data set that was positive, but, you know, just how, again, that could help, you know, maybe drive future growth, but just kind of walk through that, that-
Yeah.
-part.
So another population that we really view as kind of a high unmet need disease setting. So if you think about CABINET, it's actually two separate subpopulations, kind of the pancreatic NET and the extra-pancreatic NET. They're kind of the two studies and, you know, the study was stopped early due to efficacy across both. And so it really speaks to not only kind of the benefit of cabo, but also the unmet need there.
Mm-hmm.
You know, obviously, don't want to get ahead of the data when we ultimately present it, but I think it just further kind of adds to the idea that cabo is a very effective agent across a wide range of tumor types. And as we look at the future and think about things like zanza, it adds to kind of our conviction set about what can we do with a cabo-like molecule, like a next-gen cabo-
Yeah
To really kind of help patients, you know, live longer.
Great. Well, you led me right into my next question. So in terms of, like, all these, you know, data sets that you've generated, thinking about Zanza, so, I mean, do you think looking at all these, these should be de-risking for Zanza ? And I guess, you know, how tight is that in, in terms of, like, understanding, you know, how de-risked Zanza could be based on the cabo data?
Yeah. So just for those in the audience who are less familiar, so zanzalintinib or z anza is our next generation VEGF-R targeting TKI. In the same vein that, you know, cabo was really developed to improve upon known liabilities or challenges with earlier generations of VEGF-R targeting TKIs, mainly, you know, the biological hypothesis around the addition of MET inhibition, AXL, MER, TAM kinases, et cetera. Zanza was really designed to improve upon one of kind of the core liabilities of cabo, which is its relatively long half-life. So cabozantinib has around a four-day half-life, and one of the challenges that presents clinically is that all patients who are on VEGF-targeting TKIs tend to develop adverse events and have to be dose-reduced, dose-managed, et cetera.
But with a long half-life, the accumulation you actually see can be on the order of 10 days, you know, 2 weeks, et cetera. And so from purely a patient management perspective and adverse event management perspective, that's complicated. And so oftentimes, frankly, the inertia needed to get a patient back onto drug after a relatively long washout period can be challenging. And so coming back to the comment I made earlier, you know, patients don't benefit from drugs they're not on. Really, the idea was to introduce a metabolic liability into the core kind of cabozantinib scaffold, where we're retaining the kinase inhibition profile of cabo, but improving the kind of PK properties. And the data that we've presented so far clearly show that we've been successful at that. So zanza's half-life is a little under 24 hours, and so greatly improved.
And so one of the reasons, one of the kind of big, big advantages we think we have, is that we can use all of our learnings from the cabo experience, given the similarities in the kinase inhibition profile, to inform where we're going to go in the future. So if you look at the studies that we've announced so far, STELLAR-303, that's a late-line colorectal study, STELLAR-304, non-clear cell RCC, and then STELLAR-305, which we recently announced in head and neck cancer. All of those are greatly informed by the cabo experience, where we can point to highly encouraging data sets from cabo plus checkpoint inhibitors, or even cabo alone, to help inform, you know, what are the populations that could potentially benefit? What are the advantages of cabo or zanza plus IO have relative to kind of current standards?
And so we think that's, you know, really a value add for us because it lets us not only de-risk development, but also accelerate development as well. So that's something that, we're particularly excited about.
Got it. Will you walk through kind of the development plan for zanzalintinib and the ongoing trials and, you know, I guess, you know, some of the updates that we might get in the near term?
Yeah. So, I guess in terms of the development plan overall, one of the things that historically we've framed Zanza is about the breadth of development. You know, coming back to this, to our discussion earlier around using cabo to inform that, both as a monotherapy as well as in combination, cabo is shown to have, you know, real tumor responses, per RECIST 1.1 responses in about 20 different tumor types, tumor settings, tumor indications. And so that shows the potential breadth of development. And so then when we think about Zanza going forward, it's about the opportunity set of not only where haven't we gone with cabo, that maybe we could with a more user-friendly, potentially better tolerated version of it, but what are the sorts of novel doublets and triplets that we can improve upon the cabo profile?
You know, in sensitive tumor types like RCC, are there novel triplets? Are there novel combinations? Are there— Is there the potential to move earlier line into earlier lines, given the, you know, potential differences in adverse events that we've seen so far? So that, that's kind of informing, you know, how we think about it from a very broad perspective. Specifically to the ongoing studies, so STELLAR-001, STELLAR-002 are kind of our two phase I, phase Ib studies that are looking at both Zanza monotherapy in dose escalation, dose expansion, as well as Zanza in combination with nivo, you know, other checkpoints, nivo/rela , to really give us an idea of where we can go kind of in the future. STELLAR -303, that's our first pivotal study there.
So that's a study in late-line colorectal cancer, looking at that against regorafenib. And so the original idea behind both 303 and 304 was these were kind of the two low-hanging fruits, so to speak, of the Zanza development portfolio. Rego, unfortunately, isn't particularly effective, and so there's a real unmet need there in a pretty significant population. We've seen some pretty compelling data looking at Zanza in combination with atezolizumab and COSMIC-021, or cabo in combination with atezolizumab and 021, and then cabo in combination with durvalumab and the IST, the CAMILLA study. And we've seen some pretty compelling response rates, and that kind of drove interest in 303. We recently announced a slight change to the protocol there, focusing on the non-liver met population, and that's really informed by kind of an emerging.
Consensus, so to speak, in the clinical community, that specifically in colorectal cancer, patients with liver mets tend not to do well on checkpoint inhibitors. LEAP-17 is kind of the most contemporaneous, large randomized data set to essentially confirm that. And so we looked at that data and really, in consultation with, you know, ad boards, all of our experts, advisors, and really said, "You know, to really increase and maximize the probability of success of that study, what can we do?" And it's really to first, on a hierarchical basis, look at the non-liver met population and then see if some of the differences between zanza and Lenvima, which was the TKI in the LEAP-17, can that drive improvement in the overall population as well? So that's colorectal, non-clear cell, that's the 304 study.
That really builds upon all of the experience of cabo and cabonivo in that population. And then 305 builds upon some data that was presented at ASCO of last year, which showed a 54% response rate of cabo plus pembro, and that's in contrast to a sub-20-20% response rate for pembro alone in that setting. And so, you know, conceptually similar, if you think about why the initial idea behind, you know, cabo plus a checkpoint in RCC, was really to provide kind of that initial, you know, debulking of the tumor to drive some of the immunogenicity to add to the checkpoint. And so given that, you know, pembro alone isn't particularly effective from a response rate or PFS perspective in head and neck, we really think there's a compelling opportunity there.
So that kind of is a little bit of a flavor of our late-stage programs.
Yeah. No, that's great. I mean, how do the development timelines for Zanza kind of like, you know, mash up with, you know, kind of cabo in the, you know, probably multiple, you know, patent expiry scenarios? You know, basically, like, again, in terms of, like, if, if it were to go, you know, just or expire or, go generic in, like, 2026, will Zanza be approved or, you know, do you think it'll be on market by then? Like, yeah, just trying to match up in terms of the different scenarios that could play out out there.
Yeah. So it's, you know, still too early to kind of comment on, you know, when a lot of these data sets will read out. Really, kind of our focus going forward is on kind of, you know, development of cabo, development of zanza, development of XB002, and continuing to kind of invest in those franchises, and then frankly, execute commercially on cabo. You know, we do have, you know, an upcoming trial for our ANDA with cabo in October. We continue to be highly confident in that, and, you know, it's probably the single biggest question we get today.
Single biggest question we get from, from most investors, and, you know, frankly, Ben, probably the biggest headwind or overhang on the stock now for a, for a couple of years. And so, you know, we're going to continue to kind of operate and defend our cabo franchise while beginning to... or while continuing to execute on our earlier stage portfolio. It's kind of, you know-
Yeah
... our strategic path forward.
Got it. Helpful. And then, yeah, maybe if you want to provide some background on XB002 and kind of your thoughts there on that program and, you know, ultimately, you know, there is already, you know, an approved drug, similar, you know, MOA. So just kind of, you know, where's the differentiation? There's also others that are in development. So again, like, you know, where do you guys think you kind of are different?
Yeah. So I guess taking a step back, I think, you know, the way we view our portfolio, not only zanza, XL092, our earlier stage pipeline, is really through the lens of the cabo experience. As I said before, cabo was developed with the understanding of validated biology that we think we can improve upon. So either novel, kind of first in class or best in class perspective. So zanza, as I said before, is kind of validated through the cabo experience and improves upon it with the, you know, the shorter half-life and subsequent potential advantages because of that. XB002 takes a validated target for an ADC tissue factor, which is a highly compelling target expressed in a wide range of tumor types, and really takes that reference product, Tivdak, from Seagen and Genmab, and says: "Okay, how can we improve upon it?
How can we differentiate it?" So if there are really three components of the ADC, it's the antibody, the linker-
Mm-hmm
... and the warhead. We think that XB002 is actually differentiated across all components of that, and that's what drives our enthusiasm about it. So first, on the antibody side, one of the challenges with tissue factor as a target, even though it's broadly expressed, Tivdak's antibody is actually competitive with factor VII. So if you think about the role of factor VII in the coagulation cascade, a competitive antibody can actually be pretty challenging from a safety and tolerability perspective, notably about bleeding risk, and unsurprisingly, they have that in their label. And so kind of XB002 targets tissue factor in a slightly different way, where it's non-competitive with factor VII, and so it's differentiated, and presumably, we, you know, wouldn't expect to see the same bleeding risks. And the data that we presented last year at the ENA Triple Meeting certainly confirm this.
Then on the linker warhead side, again, at that ENA conference last year, we presented some data to suggest that, you know, what we're really seeing is a much more stable ADC. So if you think about kind of Pharmacology 101, what do you want to see from an ADC? It's about exposure, and it's about less free drug.... because free warhead is often what drives a lot of the tolerability and adverse event issues, things like neuropathy, things like neutropenia, which can be hallmarks of kind of the MMAE class. And so again, unsurprisingly, Tivdak does see, you know, fairly high rates of both of those. And so the data we presented last year essentially show at equivalent doses of 2 mg per kg, see about twice the exposure and one-tenth the amount of free drug.
From our perspective, that's a really compelling starting place when you're looking at an ADC that targets a, antigen that is well-validated through Tivdak. And so as we see their data, we can use that as kind of a guidepost on where are we going to go forward with development, with the perspective that we think, you know, frankly, XB002 is a, a better ADC with a potentially much wider TI that allows us to look at a broader set of indications, because, as I mentioned before, tissue factor is highly expressed. Just to kind of add to that, we've recently started talking about XB371. So that's another tissue factor targeting ADC that we have, but instead of an auristatin warhead, which is what XB002 has, it has a topo warhead.
So the reason that that's important is that tissue factor is widely expressed, but not all tumor types are actually sensitive to an auristatin chemo. And so when you see things like colorectal cancer, where tissue factor is highly expressed but is not sensitive to an auristatin, that's where a topo ADC would be quite effective. So we take a very kind of empiric, and what I think is rational approach, to ADC development. It's not a one-size-fits-all approach.
Sure.
It's how can we leverage this differentiated antibody in this wide expression profile tissue factor to make sure that we're kind of covering the bases, there, so.
So when would we get the next update there in terms of, you know?
Yeah. So I guess kind of, you know, our standard playbook, so to speak, is we generally present data when it's stable, mature.
Yeah.
You know, our philosophy is, "This is what the data are," as opposed to-
Yeah
... you know, a lot of companies tend to put dribs and drabs-
Yeah
out there, and, "This is what it could be," and try and hype the stock and raise money.
Yeah.
You know, we're a profitable company. We have a strong balance sheet, so we're not out kind of trying to hype and raise money. And so, you know, the way that we think about XB002 is, you know, it's a couple of steps behind Xanza.
Okay.
So if you think about our disclosure philosophy around Zanza, last year's ESMO, we had the dose escalation-
Yeah
... data, which basically said, you know, in an all-comers population, is the drug behaving like we think it should be? And if you compared kind of the initial cabo experience to the initial zanza experience, very similar, but a lot of the PK differences were clear. We were successful in kind of doing what we set out to do. Then we looked at expansion cohorts in what were known to be, you know, cabo or sensitive tumor types, and that's the recent data that we presented, a bit earlier this year in May, around, you know, the 50% response rate in cabo-naïve patients, 30-something% response rate, kind of in an all-comers setting. So it's really kind of using that first dose escalation experience to validate were we successful in kind of doing what we were trying to do?
Mm-hmm.
And then confirm that in tumor types, more homogenous populations known to be sensitive. So similar to that, with XB002, last year at the triple meeting, we kind of talked about a lot of that original kind of PK exposure, free drug data, and then we've continued to escalate. We talked about on the last earnings call that we've now hit kind of. We have our phase II dose, and then let's go into expansion cohorts to really test that hypothesis around, you know, is this differentiated? What's the activity profile, and what are tumor types that are known to express tissue factors? So that's kind of the next step going forward.
No, great playbook there. Maybe last couple questions. So, you know, you got an R&D day coming up in December. People are very excited about it. I guess maybe give us a little flavor of, like, what we should expect there in terms of like, you know, where the focus will be. Will it be on some of the Zanza programs and going deeper there, as, you know, we just kind of discussed, or, you know, kind of the really early stage programs to understand, like, new targets? Like, or it could be a blend. Like, so where do you think, or what are you planning to kind of-
Yeah, so don't want to kind of preview that-
Yeah
... too much.
Yeah.
Still got a bit of time to pull all of that together. I think, kind of, you know, somewhat of a statement of the obvious is that, you know, we're really at an important kind of point for the company where we have this, you know, successful cabo franchise that's allowing us to invest in all these different areas in the future. And, you know, we have a lot to talk about, whether it's z anza, whether it's XB002, whether it's our earlier stage portfolio in terms of XB010, that's our 5T4-targeting ADC, our overall, you know, research philosophy. So it's really about kind of framing the company overall. And so, again, don't want to get too ahead of anything for the R&D day, but it's certainly going to be exciting.
Yeah. We look forward to it. Maybe just a question on, you know, kind of your cash and obviously, you know, this commitment to buy back stock. I mean, you came out with the, the share repo. So, I mean, I guess, one, you know, how active, you know, do you think you'll be in the second half of the year? And also, you know, is this indicative of other times? Like, once you kind of, you know, do the full repurchase, is this something that we could see consistently from you guys in terms of, you know, again, driving shareholder value that way?
Yeah, so, we've committed to, you know, completing the buyback by the end of the year. Chris talked about on the last earnings call a little bit, a little bit more detail around, you know, the progress we've made so far. You know, don't want to speculate in terms of, you know, what can we do in the future. But, you know, it's one of the things that I think is a strategic asset for us, is the strength of our balance sheet.
Gives us kind of the ability to do a fairly wide range of opportunity sets, whether it's, you know, buybacks like we announced earlier this year or, you know, business development, licensing transactions or, you know, frankly, even potentially something larger if there's an asset that we have a high conviction level in that we think could make sense to bring into the portfolio. So, you know, we think having a strong balance sheet is really an asset for us. We want to make sure that we're maintaining that.
Yeah. And maybe what is the ethos for you guys in terms of M&A and business development? And again, do you feel that that's a priority in terms of, like, having to do a medium to larger size deal? Or do you want to just kind of focus more on licensing and kind of, you know, smaller business development? Like, where do you guys kind of want to play right now? And what's the priority?
Yeah. So external innovation-
Yeah, yeah.
-which is pretty encompassing, across a wide range of transactions, is certainly a main priority for us, high priority for us. You know, Mike talked about it-
Mm-hmm.
-in his prepared remarks, even on the last earnings call, that, you know, it's something we're certainly focused on as a company. What that looks like, when that happens-
Mm.
How it looks like is probably more driven by our conviction level in the asset. I think from a philosophy perspective, what we're not gonna do is go out and do a deal for the sake of doing a deal. You know, we don't want to put a couple of bars on a pipeline chart just to kind of know deep down that it's probably gonna go away in a couple of years. We want to make sure that we're doing the right thing, taking the right amount of risk for the right amount of value, and understanding how do we have an advantage? Is it on the clinical side? Is it on the mechanism, cancer biology side? Is it on the commercial side?
Mm-hmm.
Is it a market that we know really well? And so, you know, that's how we approach all of these things is, you know, what's the compound? What's the indication? What's the competition? These are all things that we've done really well. We've been successful with cabo, and so that kind of informs, again, you know, the cabo lens is how we view all of this. But I think suffice it to say that, you know, external innovation, whether it's BD, whether it's M&A, whether it's a mix of anything, whether it's more of these option deals, we did a couple of those-
Mm.
at the end of last year, certainly a priority for us, and we want to again, continue to use the cabo gas to fuel the Exelixis engine, and BD external innovation is certainly a part of that.
Perfect. Well, Andrew, let's leave it there. Thanks so much for joining us.
Great.
All right. Good to see you.