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

Good morning. Please welcome Michael Morrissey.

All right, welcome, everybody. Thanks for joining us today at our R&D Day, where we're going to focus on science and strategy of building the Exelixis pipeline of the future. I want to thank everybody for joining us today here in New York. I want to welcome everybody joining us online as well. It should be a good couple of hours. Before I start, before we start, I want to thank the team front row. We've spent the last month or so prepping for this. So I want to thank everybody for their efforts in refining the slides and the messaging and everything else has been a real team effort. I want to shout out to various board members who are here today. Thanks for your support and interest in coming here today as well. I want to welcome and acknowledge Dr. Monty Pal, from City of Hope. He's our long-time colleague and collaborator with cabo and zanza. He's at least today entered a really folk hero status at Exelixis. He was in the clinic yesterday, took the red eye last night, and he actually beat me to the gym today. So pretty amazing. I got there at about 6, and there's Monty on the treadmill. So impressive. But Monty, thanks again for joining us and looking forward to hearing from you about cabo and zanza and your insights there. And I also want to thank the full Exelixis team Susan Hubbard and her team, Deborah Leong helped us kind of put together the messaging and the themes for the actual presentation today and the Wilson Dow team as well. So -- so we'll focus today on science and strategy, how we're building the pipeline. You'll hear less about cabo, but you'll hear a lot about how we use the cabo lens to focus our efforts in building the pipeline in both discovery and development as we move to commercial, clearly, that's the main theme. The business of Exelixis we'll talk about at JPMorgan in January and I'll talk about guidance, talk about goals, those kinds of things, but we're going to separate those 2 so we can really laser in on what we're doing with the pipeline. So look, our job quite simply is to improve standard of care for patients with cancer. Everybody at the company spends literally every hour of every day focused on that. That's all we're here for. If we're successful in doing that, and we have been with cabo, we can move the needle for patients and we'll build value in the company and build value for shareholders. It's a really simple game in those terms, but it's a hard business. We all know that. And that's why we're here, right? So we're excited to be here. Thanks for joining us, and let's get rolling here. Okay. So let's go back one. Okay, here we go. So thank you. In fact, one more. All right. Fantastic. So okay, great. So the morning will be -- I'll say a few more words of introduction. I'll pass it over to PJ, he'll talk about how we integrate our commercial insights into the R&D strategy. Really differentiator for us. Most companies in our -- kind of at our size, our scope don't have that. I think it's a really important part. He'll pass the podium over to Dana Aftab who will be doing the majority of the heavy lifting today in terms of our discovery efforts around the pipeline, take a quick break, we'll come back, talk about zanza. He'll introduce -- Dana, will introduce Monty, And then Amy will wrap up talking about how we're doing development for the pipeline as well. And then we'll take Q&A again after a quick break. Okay. See if this works this time. There we go. Okay. We will be making forward-looking statements today. So please see our SEC filings for a description of the risks that we face in our business. All right. So let me start here and put a stake in the ground by saying that in 2024, Exelixis is positioned to be a global leader in biotech oncology R&D. That's our focus, that's our drive. That's not aspirational. I think we've already made a lot of progress there, and we're moving in that direction in a stronger focus, a stronger move every single day. But that's where we're at, and that's where we think we can be and need to be to have the maximal impact that we can have for patients with cancer. Obviously, everything is built on the cabo foundation, #1 TKI VEGFR targeting TKI and RCC $2.2 billion in revenue for the last 4 quarters. #1, TKI in terms of its ability to either as a monotherapy or combined with checkpoint inhibitors -- we've learned a lot from cabo from that story, that journey, both good and bad, right? It's never a straight shot from starting out to reaching the goal line. It's a very, very tough business. lots of stops and starts. We accept the idea that attrition is an important part of the process, but we've learned from all of that, and that cabo lens has really focused us on how to build a pipeline that can do this again and again for patients. I would say at Exelixis, we're not linear thinkers. We think on a log scale. And I think what you'll hear today is that the pipeline that we've got currently can really reach 10x the number of patients that we're currently helping with cabozantinib. And that's probably 5,000 to 10,000 patients a week. So we're excited about those opportunities scientifically, clinically, commercially, right? Again, if we're successful in generating differentiated data with our molecules, differentiated assets with our R&D efforts, we can move the needle for patients on a much larger scale than we have today. Now doing that is hard, right? And I think one of the hallmarks of Exelixis has been that we have not been shy about investing in large global randomized pivotal trials because that's the only place at the only time that you can actually prove clinical differentiation. We've embraced that approach. We've ran numerous up to 17 pivotal trials with cabo that are ongoing. You'll hear more about our success rates in a few minutes. But that's how you generate differentiating data. You can't pick the winners with a few PRs in Phase I. You can't generate a waterfall or some swim lanes and actually prove what you've got is going to move the needle for patients. Now a lot of companies do that to raise money, and that's kind of where they end. That's just the beginning point for us. We're in the game of running these trials to be able to generate the right data at the right time and the right indications that can move the needle for patients. And that's the main focus. So that conviction is a big part of what we do. We're disciplined in how we do that. We have a very keen eye for good data. We had talked about at the company for years, if not decades, that data drives the process. We're a data-driven organization, and we live and die by that data. And our compounds do as well. We have over the years. And in fact, in the last few years, stopped a number of programs because the data just doesn't look good enough to pursue. At the same time, when we see that breakout data and those hints of activity that drive us forward, we're not afraid to put our pedal on the gas and push forward to accelerate our progress because patients need the drugs that we're making and the combinations that we're pursuing and the ability that we have to operate that. So look, it's a tough business. We all know that. Our goal is to maximize success, not minimize failure and as an organization down to every single person. We have the urgency, the focus, the intensity to make a difference for patients. Okay. So how we create value. I think this is a good slide, which kind of summarizes that within some -- I think, some 4 different graphics, right? Obviously, cabo is a starting point for us in terms of the financial foundation that we've got in terms of, again, that cabo lens that helps us understand how to navigate the science, the clinical processes, the competition, the ability to actually collaborate with the competition to bring the best outcomes to patients, right? We're myopically focused on solid tumor oncology. We don't do pain, we don't do inflammation. We don't do weight loss. We focus only in oncology. And oncology is still the #1 investable area across the biopharma spectra in terms of actual dollars spent, trials run, patients accrued, et cetera. So it's a huge opportunity for us to make a difference. And we've done that already, right? We tried to punch above our weight whenever possible. but the competition, again, can also be a very, very potent source of collaboration, which I think we've done well, and we'll talk about that in a second. But our robust pipeline is here to be able to move that needle, whether they be novel molecules, novel bispecifics or ADCs or looking at next-gen in terms of zanza XB002, et cetera. So Dana and Amy will, I think, frame that really well. Again, we're a big small company. We have the speed, the agility, the intensity of a small biotech with the size and the scope of a big pharma that can really look at a much bigger scale than most companies in our spectra of opportunities. So we're excited about that. We have the ability to really maximize supercharge our culture to go to the next level. And I think we're doing that right now across the wide range of activities that we're pursuing. And you'll hear about that today. You'll feel the energy and the intensity from the team. And I think that's just indicative of how we roll every single day. All right. So cabo, I won't belabor the point here. You all know the molecule well. You'll hear less about this today because to be quite frank, while it's certainly generating the revenue that we need to invest in the company, new indications, potentially CRPC as well as net that could drive growth in the out years. Our main focus is building a pipeline, right? Doing what we did with cabo again and again and again across different molecules. But again, it's the learnings of cabo that are so important, right? The idea that in 2016, in 2020 and 2024, me-too drugs with equivocal activity of standard of care, don't cut it. They don't cut it for patients, they don't cut it for shareholders and they don't cut it in the marketplace that we compete with commercially. So we're always striving to go one level up to move the needle for patients because if we can do that, and that's hard, no doubt, but if we can do that, we could be very successful clinically and commercially. I mean cabo is a great story when you kind of go 1 layer deep. We talked about our conviction that we have to run the right trials. We have to make the right investments in terms of running pivotal trials, we've done that, I think, on a very broad level, right? If you look at what we've done by ourselves in collaboration with various biopharmas, cooperative groups, we have either completed or are about to complete 17 different pivotal trials, right? Of the 14 that have competed, 10 have worked, Primary endpoint was positive. What's interesting in the VEGFR TKI space, we weren't the first here. We weren't the second here. In fact, there were 76 molecules that target VEGF as a TKI that entered the clinic. 76, right? 11 of those were approved. cabo was #8, but because we focused on generating differentiating data, we were able to, again, move the bar for patients and generate what is now the #1 TKI in the RCC space. So again, the learning here is that being okay is not good enough. You have to be better. You have to be better for patients. And if you're successful in doing that, then you can make a business out of that as you go forward. All right. We've done that, I think, in a smart way, both clinically as well as economically I'm really, really proud of the collaborations that we've done with our commercial partners, Ipsen and Takeda as well as our clinical partners like BMS and Genentech to be able to essentially combine and conquer, right? Combination approaches in oncology have been the mainstay for not just years but decades. So that's kind of a no-brainer. But the idea where you can take I would say, different kinds of MOAs, different kinds of molecules and combine them, asking key scientific questions that then lead to key clinical outcomes is just super important in doing that in a way where we can share expenses, we can co-fund together, we can drop the individual investments and then do more with our dollars. So that's been very successful with cabo between 9ER and the contact trials, we paid about $0.25 on the dollar for those trials. And we are absolutely trying to do that again with zanza with XB002 with every molecule as it evolves in our pipeline. We want to be able to combine and work with our -- with our competition, if you will, to move the needle for patients, do it in a cost -- in an economically feasible sort of way so that we're successful both clinically, commercially as well as financially. All right. talked about commercial. I won't belabor the point here. It's a real, real, really important part of the Exelixis story in terms of how we do R&D. That commercial lens is critical in helping us pick the winners, if you will, from the standpoint of key indications. When I look back at the 40 or so approvals and launches since 2016, there's a clear reason why cabo is #1 for biotech in that space. It's because we had the right insights in terms of competition, in terms of the marketplace in terms of what standard of care look like and how prescribers would respond to the next-gen molecule with better data. So PJ and his team, Deborah, they are intimately integrated into how we actually define our strategy, how we look at potential compounds and products because end of the day, they're going to be the ones who have to carry the ball on moving that forward in terms of that commercial realm. So great to see them here and part of the overall approach. And I would certainly recommend that at a break or at lunch today, spend some time with them to kind of pick their brand about how we're doing it as well as what you hear today. All right. So here's our pipeline. You're going to hear about most of these compounds today from the team. I won't belabor the point now, except to say that, again, bottom left, compounds that we stopped this year, including XL102, our CDK7 inhibitor, interesting targets, good molecule, simply didn't meet the bar for success that we wanted to see in the Phase I trial. So we made the right move and stopped it, right? We are the ultimate prioritizers. We've been doing this scientifically for decades. This is a good example where it just didn't meet our standards for success, but we have so many options in the pipeline to move forward that are as, if not more exciting. So it was a very easy decision for us to make. So you'll hear more about that as we go forward today. So stay tuned. So I will end here. My time is up. Again, 4 key pillars I talked about that you'll hear about, I think, today in a very much better away from the team. Again, we're all in on development. We're not shy about investing. We make smart investments. We run pivotal trials, that generates the data we need to be able to move forward, both clinically and commercially. We prioritize programs exceedingly well. Data drives the process that Exelixis always has, always will. We collaborate well internally and externally. That's been a real hallmark of our success over the last 5 years and the goal every single day is to improve standard of care for patients with cancer. So welcome again. Glad you're here should be a fun morning. And with that, I'm going to pass the podium off to PJ to talk about the commercial impact on R&D. So thank you.

Morning, everybody. Great to be here with you today. Thanks for coming. Excited to have the opportunity to talk about, as Mike articulated, how commercial integrates across the value chain in our R&D and also talk about cabo and a little bit of the potential near-term markets that we're going into. So I'll start with cabo, then move on from there. So cabo, as Mike mentioned and as you're well aware, has been a very successful molecule. We've helped a tremendous amount of patients with cabo across tumor types -- our trailing 12-month revenue was $2.2 billion globally. And importantly, we're the #1 TKI in RCC. We're the #1 combination in terms of I-O/TKI in first-line RCC. We're also the #1 TKI in second-line HCC. We've done this despite in the case of the first-line combination setting despite not being first to market and in a very competitive market. And I think kind of the lens that Mike described is really thinking about what kind of data can differentiate and what can move us forward in the market has been the key to our success there. So really, that's the kind of the focus and a critical part of how we think about our R&D strategy going forward is what's going to have commercial impact and what can we ultimately differentiate with in the marketplace to help as many patients as possible. So really, that's how we're thinking about it. When we come up with target product profiles that we're contemplating, for pivotal studies. That's the way we think about it. So now I'll go into the potential new opportunities for Cabo briefly to talk about that for a bit. As you know, we have a positive study in metastatic castrate-resistant prostate cancer, contact 2. cabo in combination with atezo, we're certainly excited about that. I look forward to presenting the data. So I'll talk a little bit about the market in metastatic castrate-resistant prostate cancer. It's a very large market. As you can see here, there's 33,000 drug-treatable patients in the first line, 26,000 in the second line and so forth. What's important to note here, though, is all these patients are still progressing and ultimately succumbing to their disease. So these patients need more ultimately. And we're -- we're certainly hopeful, and we would be very excited to have the option to bring cabo to them. One thing that calls this point out is the overall survival rate in this setting is only 15 -- the 5-year survival rate in the setting is only 15%. So that's an area that we can improve. So important to note kind of the landscape you see here, NHTs have moved up in the treatment landscape in recent years. So not only are they moving forward in the metastatic setting, but they've moved forward into the castrate sensitive setting as well. So what does that mean? That means that approximately 50% of patients coming into sort of the first-line setting in metastatic castrate-resistant prostate cancer have already received an NHT. And in the second line of MCRPC pretty much all patients will have already received an NHT. So when you think about the CONTACT 2 trial design, that's important to keep in mind. As I mentioned, this is still a significant unmet medical need in this setting. When we do our market research, we talk to physicians like Dr. Powell here and others, many physicians don't really want and certainly not patients to use chemotherapy, particularly if they could delay it. So I think when we think about the potential for and what we're hearing in our research for cabo and atezo, it's 2 new mechanisms of action in prostate cancer, which is exciting, certainly immunotherapy broadly used in oncology to have the opportunity to use it more broadly in prostate cancer will be important for patients, exciting for physicians and to have a TKI to complement that will be great because that's -- they need more options ultimately to move forward with. So now I'll talk a little bit more about neuroendocrine tumors. So this is the other potential new market for cabo. You'll hear about this more from Amy with regards to the data that was presented at ESMO recently, very well received. The cabinet study. Neuroendocrine tumors are -- it's a heterogeneous population. It arises from a variety of tissues throughout the body, certainly GI tissues, pancreatic neuroendocrine tumors are the worst prognosis. They arise in the lung as well as other tissues in the body. The incidence has been increasing recently, and that's driven by better detection and certainly better diagnosis as well in the setting. So it's a large market opportunity. As we look at it, the second line plus neuroendocrine tumor drug-treatable setting is about 8,000 patients a year. Now as I mentioned, this is a relatively indolent disease. So when we think about that patient population, they're surviving for some time as well. Most of them are going to see 3, 4 lines of therapy or beyond. So there's actually 5x the amount of prevalent patients there or greater than 5x than the 8,000. So it's a large market, and these are a lot of patients that need better therapies. It's a significant unmet need. You can see here in the treatment landscape in addition to the somatostatin analogs and radioligand therapy, there are just really a couple of older therapies, everolimus, sunitinib, so really a big unmet medical need here and an opportunity to move the ball forward for patients and really help them in their journey. So kind of that provides a nice backdrop of how we're thinking about our strategy overall. I mentioned we have to raise the bar in the standard of care for cancer patients in solid tumor oncology. -- and that will continue to be our focus. Mike talked about that. What we want to do really is strengthen our leadership position in RCC and in GU tumors. We want to continue to strengthen our position in gastrointestinal tumors. And beyond that, we really want to go into some new disease areas in terms of lung and head and neck as well as breast and GYN tumors. Our pipeline is really well positioned to do this. Certainly, zanzalintinib, which you'll hear a lot about today from all the other presenters provides a really great option to take forward in RCC as well as other tumors. XL309, very exciting, potentially in ovarian and breast cancer and certainly our tissue factor franchise will have broad applicability in many malignancies in these disease areas. So from here, just I'll talk a little bit about how we got there, how we think about it and really the lens we put on these opportunities. So this is a simple 2x2 matrix, really talking about the factors that we think about when we think about these opportunities, both from a commercial perspective, a value perspective, but really, that all stems from the ability to improve standard of care for these patients, and those will travel together. So on the vertical axis, we really think about what's the impact for patients. We think about that in terms of the market size, potential duration of therapy. But obviously, we want to think ahead about the competitive landscape, not only what it is today, but what it's going to be down the road. So we can make those decisions as best possible for the future of our pipeline. And then we obviously want to think about maximizing the chances of success, and that's where really the integrated approach across research, across development and collaborative approach comes into play. And ideally, we'll prioritize opportunities in the upper right of this quadrant. So high patient impact and greatest chance of success, critical to our thinking. And if you see that, with the disease areas we've laid out here, lots of patients available for us to address going forward. And I want to kind of show you how our pipeline is really positioned well in these disease areas. So you see the 4 disease areas I described here and our pipeline sort of laddering up to those areas. The variety of modalities we have here that we can use to target these disease areas is critical in terms of TKI, synthetic lethality, ADCs and immuno-oncology assets. So it really gives us the opportunity to pick and choose the best ways, follow the data to take these opportunities forward. Mike referred to the fact that we collaborate with our competitors. So certainly, opportunities to collaborate externally. It would be critical to continue to build with novel combinations moving forward. Critical on oncology more and more as we go forward. And certainly, what's very exciting to us as well as the pipeline matures is the potential to combine our assets internally, which is really exciting for us as we really think about novel ways to help patient populations and improve standard of care. So now I want to do sort of a double-click just in some of our more advanced assets that we're very excited about, zanza. I've got a slide on this here. You'll hear about it from everybody, certainly from Dr. Pal with the data he presented recently at IKCS. So a very exciting molecule for us. Cabo has been fantastic. We think zanza is better and can differentiate really well in the marketplace with improved safety and tolerability, potentially improve combinability and the ability to kind of not only do more in the areas where cabo is approved, but beyond that into more markets and more tumors to help more patients, very exciting. We've got 3 ongoing Phase III studies now for zanza. You'll hear more about that from Amy, but from a commercial perspective, these are very exciting third-line colorectal cancer, 30,000 patient opportunity. It's an area of significant unmet medical need. These patients really need better therapies to help them. It's a very severe disease. Non-clear cell renal cell cancer. This is very exciting because, obviously, RCC is an area we know really well. And this is the first randomized study that we're hoping to address this specific part of the population. So very exciting there as well and an opportunity to get zanza into RCC. Our recently announced STELLAR-305 study in head and neck cancer, again, very exciting opportunity, significant unmet medical need. Head and neck cancer, as we've thought about it in the company, it's a tumor type that's a relatively similar size in terms of patient population to RCC. Many years ago, and I worked on RCC even before I was at Exelixis, as Dr. Pal and I were reminiscing about this morning. A long time ago, RCC wasn't viewed as sort of a commercial market, if you will, that it was today. We kind of think of head and neck cancer as having a similar opportunity, improving standard of care, getting combinations going multiple lines of therapy. It's a market that we believe can really grow, so exciting for us. I'll move on now to our tissue factor franchise. Again, you'll hear about this from Dana and Amy. We're very excited about this. Having a product in XBO2 that we believe is differentiated. From TIVDAK. We want to push it forward rapidly in development. It's a molecule that can address a variety of solid tumors. You see many of them listed here. So very excited about the broad potential of this molecule -- and what's really neat about the tissue factor franchise is we're able to add a different war load -- warhead, excuse me, instead of the MMAE warhead, a topoisomerase inhibitor and that gives us the ability to target other malignancies that differentially expressed that really exciting there is colorectal cancer. So the ability to target a broad swath of solid tumors with this franchise is exciting. And finally, our newest clinical asset, XL309 is a best-in-class USP1 inhibitor. A lot of excitement about this molecule. First and foremost, there's a real clear path where this can go forward in terms of development. It could work perhaps after PARP inhibitors. PARP inhibitors, as you well know, is a very large commercial opportunity. In the U.S., sales for PARPs are about $1.6 billion. On the global scale, that's $3.6 billion annual PARP sales. So the ability to help patients beyond PARP inhibitors, when they do inevitably fail or even moving forward, combining with PARP inhibitors is exciting and then potentially expanding that market beyond the PARP inhibitor market either with combining with other novel agents or chemotherapy. Certainly, some of our pipeline agents could be very exciting here, and you'll hear more about that from Dana. So what does this all look like sort of moving forward for Exelixis in terms of our vision. Today, we're cabozantinib commercially on the market. Very successful in 3 different tumor types and 6 indications. In the relative near term, we have the potential to add 2 new indications in terms of prostate cancer and neuroendocrine tumors for cabo as well as bring zanza to patients in colorectal cancer, head and neck cancer as well as kidney cancer. And then beyond that, we have a balanced and broad portfolio that we see impacting patients in the 4 disease areas I mentioned, and that's a -- it's a balance of modalities of large molecule, small molecule. So we're very excited about that. And sort of to put some numbers around that, today, we have the opportunity to help approximately 50,000 patients with cabo in our approved indications. As we add on to that, as I mentioned, with prostate, with NET, with the pipeline going forward, we get to the point where we can really be impacting 13 different tumor types and over 10x the patients that we're impacting today. So we're incredibly excited about this. It's gratifying to be at Exelixis at a time where we really have the ability to live our mission and take it to the next level and help patients with cancer. So very excited about this going forward, and I'm excited to now bring up Dana Aftab to talk about the scientific aspects of what he and his team are working on at the company.

All right. Thanks, PJ, and welcome, everyone. It's a pleasure to be here. So for the next hour, I'll be telling you about our plans to build out our pipeline with molecules designed to generate differentiating data in the clinic. I'll start by describing our overall strategy at a fairly high level. And then I'll go into some detail around the various programs that are vectoring towards IND filings in the next year or two. Okay. So when we restarted our discovery operations, our goal was to generate growth opportunities by expanding the pipeline with compounds that have the potential for differentiation in the clinic. We also realized, as Mike said before, and PJ as well through the lens of cabozantinib, we realized we had an incredible resource in the wealth of knowledge we gained through our experience with cabozantinib. And we understand that this knowledge and experience really give us a unique perspective that we can apply for each and every discovery program on how to discover and develop next-generation drugs that have high potential for true differentiation in the clinic and that can really move the needle for patients and for the standard of care. So traditionally, our expertise was in small molecule discovery and development. So that was an obvious choice for us to focus on. But we also chose to add an antibody-based biotherapeutics, namely monoclonal antibodies, bispecifics and antibody-drug conjugates. Because of the clear progress those types of agents were having on impacting the standard of care for patients, especially the patients that we want to serve. So our experience with cabo taught us that one of the most important principles underlying the discovery and development of a successful high-impact differentiating drug is to focus deeply on the tumor biology when designing our drug candidates and only prioritizing programs that we feel have the best potential to benefit the areas of highest unmet need. So that is the lens of cabozantinib that we apply in all aspects of our business, especially in Discovery. Okay. So as PJ mentioned, we have 4 key areas of therapeutic focus, which is shown on the left part of this slide. And we believe these represent the greatest unmet need in oncology. This comes from a deep analysis and collaboration with our commercial partners with PJ and his team. And we apply this strategy in the earliest areas of our research and development to ensure that everything we do in discovery is really focused on generating the biggest impact on these patients. So in order to deliver differentiating therapies for these patients, we're taking an approach that we refer to as biology centric modality-agnostic R&D. So really what this means is that we focus, first and foremost, on the biology, selecting targets with potential for high impact and then selecting the right modality among small molecules or antibody-based biotherapeutics, so that we're really bringing the best drug for those patients to address that target biology. So we're really centering around the target biology and giving ourselves the option of choice between modalities. So by taking this approach, not only are we not reliant on any single platform for success, but we also gain obvious breadth in terms of the types of biology and patient populations that we can address. So to drive differentiation in the clinic, we need to differentiate our compounds in the lab first, which means that wherever possible, we try to learn from our competitors, making their compounds in our own lab and profiling them alongside our own compounds to really understand their weaknesses and issues and ensure that we are bringing the best drug possible to patients so that we're improving on any other competition. We also, with each modality, use contemporary technologies to help improve both the quality of our development candidates and the speed at which we're discovering them. and able to really bring the best agents forward as quickly as possible and in the most efficient way possible. But not everything can be predicted based on experience or even the best computational or AI platforms. So empirical research is still an absolute requirement to test and refine hypotheses, which is why we've built out our own labs to conduct our own high-impact research. Okay. So in order to apply antibody-based biotherapeutics in our discovery programs, we made the decision several years ago to do this through a collaborative platform approach. We now have a sizable collaborative network that gives us access to a range of platforms, including antibodies, ADCs, including next-generation site-specific linkers and payloads bispecific technology as well as conditional binding technology. All of these feed into our discovery process for discovering biotherapeutics, but we also have a small but very focused team of highly talented scientists who conduct high-impact experiments in our own labs, while also overseeing and leading the collaborative efforts of our partners. So we're really happy with this collaborative approach, not only because it's highly scalable and efficient, but also because it gives us a lot of optionality, meaning we are not relying on any single platform for success. When it comes to small molecule discovery, our success in the past was certainly aided by our adoption of technologies, including high throughput processes that enabled us to rapidly advance and evolve our discovery programs. When we rebuilt Discovery, our small molecule discovery group, we really -- we looked back at those technologies and adopted most of them, but we also included some new technologies, and I'm going to spend a little bit of time on that. So we continue to rely on structural enablement to help drive rational drug design, but we've adopted some newer technologies now for structural biology, including cryogenic electron microscopy, which is especially useful in solving structures of large multi-subunit protein complexes. We've also adopted rapid library construction, which is a multiplex high-throughput chemistry approach that allows us to evolve scaffolds to improve potency, selectivity and other properties such as solubility and other pharmaceutical properties. So it took us a little bit of time to build these capabilities, but they're now fully operational and contributing meaningfully to our programs. And I'll be pointing out some instances of this in the slides coming up. Now our focus in the past as well as now is on high throughput processes that allow us to optimize molecules really in a massively parallel fashion against many parameters simultaneously to develop the best possible drug. Because really, at the end of the day, it's really the compound with the best combination of both on target, PK/PD activity. selectivity, safety, drug-like properties, et cetera, that will wear the crown of best-in-class. That approach was certainly reinforced for us with our success with cabozantinib, and we're now carrying that forward in all the projects we pursue in Discovery. So the output resulting from our Discovery strategy so far is summed up on this slide. With an eye toward addressing the large number of patients that we've set out to address, we've built out our pipeline with compounds that really have the potential to show differentiating activity in the clinic. Our pipeline spans the range of small molecules in biotherapeutics and covers key focus areas of ADCs and immuno-oncology for the biotherapeutics and synthetic lethality and tumor microenvironment targeting for the small molecules. So clearly, we've made solid progress in discovery. And this progress has translated to really a deep IND pipeline with 3 INDs per year now anticipated in 2024 and 2025 across the range of modalities. So our target for IND flow is actually 2 per year, but we always aim higher to account for some attrition, which can happen for a number of reasons. But thus far, we're on track to exceed our goals. We're excited about this pipeline because we believe each 1 of these compounds has excellent potential to show differentiation in the clinic and to benefit a large number of patients. large numbers of patients. So let's get into some of the key features and the data behind some of these molecules. So I'm going to be focusing first on the biotherapeutics pipeline because we have more compounds in that part of the pipeline. And I like the organization of talking about one modality first and then the second. But there's no other reason behind that. So don't try to read into it. So first is the biotherapeutics. And this next slide shows a high-level view of the 4 compounds that are now in our pipeline progressing towards IND filings with 2 targeted for 2024 and then another 2 for 2025. Both years, we have an ADC going forward. And then next year, we have our bispecific XB628. And next year, we have -- the following year 2025, we have a new monoclonal antibody, which we call XB064. So I'm going to be spending time on each 1 of these programs, but I just want to give you a brief summary on these right now. So XB010 is a 5T4 targeting antibody drug conjugate with a microtubule targeting payload with a DAR of 2. We're excited about the target of this ADC because it really fits well into our strategy for picking the highest impact targets that will also impact a large number of patients. We feel that this is an ideal ADC target because of its high level of expression in tumors and limited expression in normal tissues. So this is an exciting compound. It's going to be the first IND we file in 2024 fingers cross. XB628 will be the next biotherapeutic to file IND on. This is a bispecific targeting PD-L1 and NKG2A. This compound is exciting because it's an immuno-oncology-focused agent that targets both sides of the immuno-oncology question, both adaptive and innate immunity. So it builds in a lot of extra power for an IO approach. Also because both of these are -- both of these binders are on the same molecule, it gives us a chance to use an engager-type mechanism with an antigen expressed on tumor cells and an antigen expressed on immune cells, it gives us the ability to co-localize those cells and give the immune cells a better chance to kill the tumor cells. So those are the 2 INDs for 2024. In 2025, we have another ADC. P.J. already mentioned this 1 earlier, XB371 is our attempt to expand our tissue factor targeting franchise with an ADC that carries a differentiating payload from the first molecule we brought forward into the clinic, XB002. XB002 has a microtubule targeting payload, while XB371 has a topoisomerase targeting payload. This gives us the ability to really broaden our reach and target a tumor antigen that really has very broad expression, but not all of those tumors are sensitive to a microtubule targeting payloads. So this really helps us broaden our reach with our tissue factor franchise. And then finally, XB064. As I mentioned, this is our newest entrant into our pipeline for biotherapeutics. We are really excited about this molecule because it targets a novel immune checkpoint pathway, which is driven by ILT2. This is a checkpoint pathway that's expressed, again across the range of immune cells, both adaptive immune cells, CD8-positive T cells as well as NK cells and macrophages. So a lot of exciting potential for a new immune checkpoint inhibitor with XB064. Okay. So now I'm going to go into a little bit more detail on all of these programs. So first of all, XB010. And again, as I mentioned, we're excited about this program because of the target. The target fits within our strategy of deep understanding of tumor biology and the targets that we're picking because of that tumor biology. 5T4 really represents what we consider to be an ideal ADC target in that it has high levels of expression in a range of tumor types, which fit really well into our commercial strategy and our R&D strategy and very limited expression on normal tissues. So that really sort of fits the phenotype of an ideal ADC antigen. So as I mentioned, it has broad tumor expression. Really what we're interested in here for our primary focus is breast cancer, lung cancer and head and neck cancer, but there's a range of other tumors that express the antigen highly. Importantly, as you can see in that heat map on the right, these tumors are, for the most part, supremely sensitive to antitubulin inhibitors, which is why we picked the payload MMAE for this antibody drug conjugate. So we launched a fairly extensive discovery program against this target and what came out of that program, we call XB010. This is a compound that has first-in-class potential for a 5T4 targeted ADC. It's a high-affinity 5T4 targeting monoclonal antibody that uses site-specific conjugation and achieves the DAR of 2 with its payload, a drug antibody ratio of 2. It uses proprietary linker technology with the linker that requires 2 tandem cleavage events with enzymes that are primarily expressed only inside tumor cells. So it gives us a very clean antibody drug conjugate approach with very little leakage of the payload in the systemic circulation. And then finally, I'll be showing you a little bit of data that speak to this. The compound shows very high potency in a range of preclinical xenograft models. So again, we see very high potential for differentiating clinical activity with this type of approach, targeting 5T4. So let's get into a little bit of data with this compound. So as I mentioned, earlier, one of our approaches in -- that we've taken all of our discovery programs is to understand the competition and ensure that the molecules we bring to the clinic are better than anything else that's come before it. So XB101 is not the first antibody to target 5T4, a number have gone into the clinic and then stopped for various reasons. So we definitely sought to improve on those molecules. We feel that XB010 is really the best approach that's been taken so far for this target. Using next-generation linker payload technology, which again gives us a very clean ADC that delivers the payload very efficiently to the tumor cells, we think will be a differentiating approach for this target. As you can see here, we synthesized the -- really what was the front runner in this space for quite a long time. This is a DAR4 MMAF antibody-drug conjugate that Pfizer was developing. And you can see here dose per dose, XB010 outperforms that molecule in preclinical xenograft models, both breast cancer models that have high levels of expression of 5T4. In some cases, we see very deep and durable tumor regressions with XB010. So we've seen efficacy across a wide range of xenograft models now, including breast cancer, lung cancer, in endometrial cancer. And as you can see at the top there, and as you might recall from what I said previously, this compound is solidly on track for filing an IND next year. So we're really excited about the potential of this molecule to get in the clinic quickly and bring some benefit to patients. Okay. So let's go to the next program, XB628. As I mentioned, this is a bispecific antibody targeting 2 different IO targets, PD-L1 and NKG2A. And this program really started because of our excitement around the target NKG2A. This is an NK expressing target that plays a big role in innate immunity and blocking innate immune responses from tumors. So the molecule that came out of our discovery program is shown here, XB628. It has very high affinity binders for both PD-L1 and NKG2A, formatted into a bispecific antibody. Again, it shows simultaneous inhibition of both adaptive and innate immune checkpoints. And also because of its it's bispecific nature, it acts as an NK cell engager. co-localizing NK cells with tumor cells to really improve the accuracy and efficiency of cell kill. I'll show you some data that relate to this. It's very potent for killing tumor cells in in-vitro models. And then this is the -- since this is a bispecific, we utilized 1 of our key technology platforms to build this molecule. It's the B body platform that we have partnered with Invenra are excited about this platform because of its promise for high yield and efficient purification for manufacturing. And I have to say that the promise of that has definitely delivered in our early manufacturing. We're well underway now for GMP manufacturing and the benefits of this platform have really played out for us. So we're really happy to be using this technology in our discovery efforts. So we really do have potential for a first-in-class therapy with this agent. There are no other bispecifics as advanced as XP628 targeting both of these pathways simultaneously. So again, a lot of excitement for us with this first-in-class potential for biotherapeutic. Okay. So why are we excited about NKG2A. So as I mentioned, it's innate immune checkpoint that's expressed on NK cells, but it's also expressed on CD8-positive T cells and macrophages. And part of the reason why we're excited about this is not just because of the known biology in driving negative signaling to these important immune cells. But there's also some clinical validation of this pathway. A monoclonal antibody was generated against this target monalizumab. It has being studied in the clinic. And when it was combined with the other binding partner, PD-L1 with durvalumab in the Phase II CO study, they saw in this trial significant improvement in various outcomes with the combination compared to just targeting PD-L1 alone. Now this is a small Phase II study. It was not powered to test any significant statistical hypotheses. But when we look at the confidence intervals, we see non-overlap of those confidence intervals, which really speaks to the potential for impact with this type of approach of targeting both of these pathways together. But remember, this was a combination of 2 monoclonal antibodies that separately and individually target these pathways. We decided that it would be a much more powerful approach to target both of these pathways, but in the same molecule. You're inhibiting both of these immune checkpoints simultaneously, but now you're also building in an engager approach to improve the accuracy and efficiency of cell kill. So that's really highlighted in this next slide. So the cartoon on the left, shows the various potential mechanisms that we're engaging with targeting either PD-L1 on the left or NKG2A on the right. But when we inhibit these while also engaging both these cell types and bringing them together to co-localize. There's really potential for higher and better activity. And we see that in the lab. So -- the graph on the top right shows just a co-localization experiment, where we mix tumor cells that express the antigen HLA-E, which binds to ILT2 with ILT2-expressing NK cells. And then we looked at a range of various treatments in this assay, including a range of different controls and then single agents that target either binder or either antigen separately, either durvalumab targeting PD-L1 or monalizumab, targeting NKG2A. We also looked at the combination of those 2 antibodies together. None of those treatments resulted in co-localization of the NK cells with the tumor cells. Only XB628, which is the blue line resulted in co-localization of the cells. So that shows that the compound is actually doing its job of -- as an engager to co-localize the cells, but is it also improving cell kill. And the answer to that question is a definite yes. In the bottom right corner, you're now seeing the results of an NK mediated tumor cell kill experiment, where again, we looked at the same controls. Monalizumab on its own is the orange line that has a little bit of activity, combining Mona and durva together shows a marginal improvement over monalizumab alone. But now look at the effect with XB628. It's orders of magnitude better in terms of potency and magnitude of effect than those 2 antibodies combined. So we really think that this is excellent preclinical validation of our hypothesis at the start that building in these 2 mechanisms of action into a single molecule bringing in an engager-type mechanism really builds in an added benefit and an added punch of antitumor activity. So again, this compound is well underway through its preclinical development activities and manufacturing, and we're on track for filing an IND in 2024. So again, a lot of excitement to see Vicki's team -- Amy's team take this molecule on and develop it and generate some early data in the clinic. Okay. So now let's go on to our next ADC, XB371. So again, this is a tissue factor targeting antibody that helps us expand our franchise targeting tissue factor. So as PJ mentioned earlier, we're really excited about XB002 for its potential for improved tolerability and the reason for this is in -- really in the antibody. The antibody for XB002 is a next-generation approach. It's unique in that it only targets an epitope on tissue factor that does not interfere with tissue factors normal function, which is to initiate the extrinsic coagulation pathway. So one of the issues that we noted with TIVDAK when we were starting to analyze this space was that there's a high incidence of bleeding events. And that's primarily because that antibody is known to block the ability of tissue factor to bind to Factor VII and initiate the extrinsic pathway. So we're excited about XB002 because of the antibody being next-generation and differentiating. That has a high potential for improved tolerability versus TIVDAK, which really gives us the ability to really drive exposure, drive antitumor activity. with the agent. But again, as PJ mentioned, there's a little bit of limitation there, in that, tissue factor is expressed broadly across a wide range of tumor types. But the payload on XB002 doesn't really address all of those patients. So in this case, really colorectal is the primary tumor type that we felt was left out of the game, but also small cell lung cancer is another potential tumor type that's underserved by a microtubule targeting payload. So we felt that by bringing in the same antibody, but now with a new payload with a different mechanism of action targeting topoisomerase 1 really gives us a broadened reach. We get access to the colorectal and small cell lung cancer patients. But also, we gain access to -- there's a lot of overlap with some of the indications that XB002 should be serving as well. So it gives us a lot of options with this franchise to potentially have benefit in areas where XB002 might not have had benefit. So it gives us a lot of options to really bring a franchise forward for tissue factor. So again, as I mentioned, both molecules, XB002 and 371 use the same antibody, which is a next-generation approach. In that it does not have the same bleeding risk that TIVDAK has. XB002 is a next-generation approach with its linker and payload. It's a novel modified or a statin. It's more hydrophilic than traditional MMAE-based drug linkers and really shows improved potential for properties that lead to improved stability in the bloodstream for lower free payload in the circulation. With XB371, we're really taking the -- taking the next step towards another next generation with another type of linker payload technology. The XB371 uses site-specific linker payload technology. It binds to 4 sites on the antibody and it's a branched linker that has 2 payloads per linker. So we're getting a DAR of 8 with this compound with the topoisomerase targeting payload. And as I mentioned, the IND filing is targeted for 2025. So let's show a little bit of data on this compound now. So these are some in vivo data in xenograft models, both of them are pancreatic cancer models, BxPC-3 on the left, HPAF-II on the right, both of these highly expressed tissue factor, and in fact, we looked at XB002 in these models. And at doses -- similar to what we tested with XB002, we're seeing complete regression or very robust antitumor activity in these models. Again, at a dose of 10 milligrams per kilogram compared to what we saw with XB002 in these models, we're seeing very similar efficacy. So we really feel that we're balanced with these molecules. They both carry the right mixture of payload and potency to deliver antitumor activity. So -- this compound is well underway now for -- through its preclinical development. It's shown excellent tolerability and dose range finding studies and non-GLP tox and we're on track for filing an IND in 2025. All right. Now the last biotherapeutic in our pipeline that I'm going to talk about today is XB064. And again, as I mentioned, the excitement for us around this target is around a novel immune checkpoint pathway, which is ILT2. So ILT2 is discovered not very long ago as a potential prognostic indicator in a range of different tumor types. So on the right, on the top right is a study that looked at a population of gastric cancer patients. where it was shown that expression -- high levels of expression of ILT2 in the tumor was a poor prognostic indicator with shorter overall survival compared to patients who had lower expression of ILT2 and then similarly, with the ligand that binds to ILT2, HLA-G, it was found that in breast cancer patients who had high levels of expression of HLA-G. They also showed a poor prognostic indication for survival compared to lower expression of HLA-G. So one of the reasons why this pathway is interesting to us is that it's a broad pathway. It's expressed across a range of both adaptive and innate immune cells. It's expressed in CD8+ T cells, including an important subset of T cells known as TEMRA cells. These are important memory cells that become exhausted and can become reactivated and drive an immune response against the tumor. ILT2 signals are very strong negative signal to these cells. It's also expressed on NK cells and macrophages and transmit very potent, don't eat me, don't kill me signals to these cells. The ligand HLA-G has been shown to be highly expressed in a range of tumor types, including one that's near and dear to our heart, renal cell -- clear cell renal cell carcinoma. So this immune checkpoint really has high potential to serve some patient populations that are very important to us in a very broad fashion. So the molecule that came out of our discovery program for ILT2 inhibitors as shown here, XB064. This is a very high affinity antibody that binds to ILT2 with a high degree of selectivity. We feel that it has best-in-class potency in terms of its activity in preclinical models, and I'll show you a little bit of data that speak to this. Again, as I mentioned, it's an immune checkpoint that's expressed on both innate and adaptive immune cells. So like XB628, it has the potential to impact multiple important antitumor immunity pathways. There's some data indicating that there's potential association for resistance with PD-L1 pathway targeting. So there's a lot of potential for combination of this molecule with PD-1 or PD-L1 pathway inhibitors or perhaps after treatment with PD-1, PD-L1 inhibitors. I've mentioned already that it's -- that the ligand is highly expressed in clear cell renal cell carcinoma. And we see opportunities to combine really beyond PD-1, PD-L1 pathway targeting inhibitors. And in fact, with this molecule, we really feel that we can build an IO franchise. When you look at this cartoon on the left, this shows all of the various pathways that we currently have in our pipeline, both for XB064 and XB628, but also including the molecule that we currently have an option on from SIRPa ADU-1805, which targets SIRP-alpha, which is another important antitumor immunity IO pathway. So -- all of these pathways have been shown to be utilized by tumor cells to evade the immune system. So there's potential to hit all of these in patients to really drive the next step in immune checkpoint inhibitor therapy. Now it may be possible to combine all of these in the clinic. We haven't done it preclinically yet, but we've done some combinations and at least in the preclinical safety studies for cytokine release and others, so far, we're seeing really good activity. So there's a lot of potential for this in the future. So now let's look at some data with XB064. So as I mentioned, earlier, we're always synthesizing our competitor molecules to understand what their activity is, what their issues are and how we can build a better molecule. Because, again, as Mike said, at the end of the day, me-too molecules won't move the needle for patients. We really have to bring the best possible drugs to the clinic to really create differentiating data and move the needle for patients. So that's what we believe we've done with XB064. The 2 leading competitors in the clinic come from NGM and Beyond. That's the NGM707 molecule, which is not supremely selective for ILT2, it binds both ILT2 and ILT4 and then the Beyond-22 compound, which is selective for ILT2. When we profiled our compound next to these side by side in the same assay looking at a key functional output from NK cells, namely cytokine release, looking at interferon gamma secretion from NK cells. You can see that XB064 really outperforms those molecules with a significant improvement in both potency and magnitude of effect. As you can see in the table on the right, the IC50s there really tell the story. We're at least fivefold more potent than the NGM compound. But then when we look at the next step, which is the ability to improve NK-mediated cell kill when we mix the NK cells with tumor cells, we see really the next step in improvement in potency. We see about a tenfold improvement in the nearest competitor. Now we know with immunotherapy agents like this, monoclonal antibodies that bind a single target, having the highest level of potency possible really drives efficacy because getting the molecule into the tumor and have it stick to the target and remain there, really drives the deepest level of antitumor activity and the deepest level of efficacy. So we think this is an important advance to really move the needle and differentiate the compound in the clinic based on the competition. So data that I'm not showing you today, but that we have just recently basically replicates these types of data, but in T cells as well. So as I mentioned, this target is also expressed on T cells and drives anti-immune signaling and T cells. And we basically are able to replicate these results in a T cell assay as well. So we're really excited about the potential for this compound. All right. So that's it for the biotherapeutics pipeline. Now let's move into the small molecules. And again, at a high level, like I did with the biotherapeutics, I'll show you the sort of high-level view of the molecules that are progressing towards IND filings and then I'll go into a bit more detail on each one of them. So we have compounds now lined up for IND filings both in 2024 and 2025. I'm also going to tell you about a molecule that is already in the clinic and where the IND was already filed this year, and that's for USP1. We're excited about these molecules for a number of reasons. Again, they fit very well into our strategy. Each one, we feel the target biology is very compelling and by addressing that target biology well will really move the needle for patients. And we also have really a significant opportunity with each one of these compounds to show differentiation in the clinic. So I'll be telling you about XB309, our inhibitor of USP1. So XL309 was in-licensed this year from a deal we did with Insilico, but we also performed a tremendous amount of work in our own labs of this molecule. And some of that work really speaks to the differentiation potential for this compound. So that's one of the reasons why I'm showing you some data on that compound today. So just to remind you, USP1 really comes from a synthetic lethality hypothesis around BRCA mutation. BRCA1 mutation is really the synthetic lethal background for inhibition of USP1. The potential superiority with this compound really comes in its drug-like properties. We see potential superiority for safety, toxicology and drug-drug interaction potential. So we're really excited about this compound. It's in the clinic now, and hopefully, we'll be seeing data from it soon. The next IND we're going to file from the small molecule platform is next year for XL495. I mentioned this molecule on a recent conference call. It's the most recent entrant into our pipeline from our small molecule platform. This is a very potent inhibitor of PKMYT1, which again is a synthetic lethal target. -- which is synthetically lethal with a number of different genetic backgrounds that ultimately result in either cyclin E, upregulation or DNA damage and replication stress. And I'll get into some detail around that. Again, as with every program, we really engineered this molecule specifically to be best-in-class and have the potential to differentiate in the clinic. And that really comes from, we believe, it's improved selectivity and pharmacokinetics based on the most advanced competitor currently. And then the last compound I'm going to tell you about still is not in our pipeline. It's a pre-development compound, EXEL-7871. We expect to nominate this as a development candidate or a compound from this program as a development candidate, very soon, sometime in 2024. This comes again from our synthetic lethality approach. PLK4 inhibition has been shown to be synthetically lethal. In a background of TRIM-37 amplification. That sounds a little bit esoteric, but it really is focused around the ability of cells to duplicate centrals and nucleate microtubules to form a mitotic spindle and go through mitosis. So I'll go into to some detail about that in just a moment. So this compound was optimized very quickly through some of our use of the technologies that I showed you earlier, namely structural biology and rapid library construction. In a matter of just a few short weeks, we are able to really rapidly advance this program and come up with compounds like EXEL-7871 that really have very high potency and selectivity for the target and that are generating really compelling data preclinically. Okay. So let's get into some of these individual programs. So first, XL309. So again, this compound comes from a target that we have a lot of excitement about, which is USP1. USP1 is a ubiquitin ligase that is an important regulator of the ability of cells to repair single-strand DNA breaks. Now, this is a similar mechanism that's engaged by PARP inhibitors. PARP also is important for another mechanism of single-strand break repair. Both of these pathways become synthetically lethal under a background of BRCA mutation. When cells have mutant BRCA, they no longer can repair their DNA through homologous recombination, and they become highly reliant on the single-strand break repair pathways. If you inhibit those pathways under the background of BRCA mutation, the cells really no longer have any mechanisms to repair their DNA, no efficient mechanisms and it results in very rapid cell death. Synthetic lethal targeting or targets are really attractive to us for a number of reasons. One, there's always a built-in biomarker to select patients with. And second, there's usually a built-in safety mechanism, in that, normal cells in normal tissues don't have the genetic anomaly in their background. So they are spared the cell kill effects of the agent. And only the tumor cells will succumb to the mechanism due to the synthetic lethality. So again, we're really excited about this program because of its potential to bring benefit to the patients we're interested in serving, which include, as PJ mentioned, the entire universe of patients who are currently treated with PARP inhibitors. But as you'll see, we also have a lot of opportunity to go beyond those patients, especially when we combine with other agents that also caused damage to DNA. We're now really putting a lot of stress on the DNA repair machinery and by inhibiting single-strand repair, it really gives us an additional opportunity to kill those cancer cells. So let's get into some data. So again, this was a program that we in-licensed, but we also generated tremendous amount of data in our own labs on this molecule and compared it to the leading molecule that's been in the clinic for a couple of years now, which is from KSQ. One of the reasons we're excited about this molecule is because it really does show best-in-class potential and a high potential to differentiate in the clinic. Why? Well, it's really again around selectivity, drug-like properties and safety. When you look at the ability of a compound to not cause an interaction with other drugs, that's very important with cancer chemo or with cancer therapy and general targeted therapy or chemotherapy because these drugs are metabolized by cytochrome P450 enzymes. And if you go in with a cytochrome P450 inducer, or inhibitor, you can really cause tremendous interactions with the combination drugs. So the best drugs are the ones that are not going to cause these interactions and create complications for the patients or the treating physicians. So when we looked at the KSQ compound, we found that it was a potent inhibitor of some key cytochrome P450 drug-metabolizing enzymes. But importantly, it was also a potent inducer of the most important cytochrome P450 CYP3A4. So that's the most promiscuous drug metabolizing enzyme and inducing that enzyme really creates a lot of complications and problems. So just creating a molecule that doesn't hit that gives you important differentiation from the KSQ compound. And we feel that with XL309, we have that. But we also have some other benefits. In some drug-like properties, we see much better solubility with XL309 compared to the KSQ compound and also better safety. When we compare the drugs side-by-side in preclinical toxicology models in rats, we found that we were able to achieve about threefold more exposure. Basically, a threefold higher therapeutic index with XL309 compared to the KSQ compound. And then when we compared the compounds in the Eurofins safety screen panel, we also saw that the KSQ compound was hitting quite a few targets that indicate a potential safety risk, and so we ensured that our molecule was not hitting those. And you can -- as you can see on this heat map, we think we really have a significant advantage with XL309. So again, our goal is to bring a better molecule -- actually, the best molecule to the clinic to bring the most benefit for patients, and we really think that's achieved with XL309. We also generated some key structural information with this compound. So this shows a cryo-EM structure of Fanconi anemia DNA complex with FANCI and FANCD2, which is the DNA clamp, with DNA, USP1 -- the USP1 activating factor as well as ubiquitin. And as you can see in the catalytically competent complex, this is a really interesting mechanism that USP uses to deubiquitinate proteins. It essentially coordinates a very important histidine His593. It coordinates with an aspartic acid with a key nitrogen on the emitizole ring of that histidine. That positions the other nitrogen to abstract a hydrogen from the catalytic cystine CYS-90, which can then perform its attack on breaking the peptide bond on ubiquitin and releasing it from, in this case, FANKD2, but also from PCNA. In the co-complex with XL309, we now learn a lot about the mechanism for inhibition. The molecule binds to a cryptic pocket on USP1. Positions the loop that contains aspartic acid 751 away from the histidine. It can no longer coordinate on the nitrogen of the emitisole ring. It now binds with a different aspartic acid. And now that emitazole can no longer allow the histidine to conduct its attack on ubiquitin. So understanding this mechanism of action is important because it gives us a lot of insight on potential mechanisms of resistance and also how to generate the next-generation molecule that will avoid those types of resistance pathways. So let's go into just a little bit of data with this molecule, a little bit of in vivo data. As I mentioned, we think this compound has best-in-class potential based on our comparisons with the leading front runner. But we also see in vivo the same sort of activity. We see very high potency and deep responses in a range of different xenograft models, including both BRCA mutant models as well as BRCA wild type models. And especially in combination with the PARP inhibitors, you can see in the upper right graph, when we combine really a non-efficacious single-agent dose of XL309 with really a moderately efficacious dose of olaparib. You can see really beautiful synergy with that combination with deep regression and lasting efficacy long after dosing is stopped in that model. So really high potential for bringing benefit to patients with this molecule in combination and potentially in combination with PARP inhibitors, potentially with other agents as well. So the next compound I'm going to talk about is XL3495. So this came from our PKMYT-1 inhibitor program. We're really excited about PKMYT-1 because this is another great synthetic lethality compound. It's synthetic lethal in the background of an unstable genome. Now that's not the typical way to define synthetic lethality. We typically define synthetic lethality through a genetic marker. But I like defining it this way because genomic instability is really a hallmark of cancer. And there are a lot of ways that you can artificially create that with various treatments. So from a strict synthetic lethality standpoint, it's known to be synthetically lethal with increased Cyclin E levels. Now the reason for this is that Cyclin E is known to create unstable genomes when it's over expressed. This is because it is an important regulator of G1 S phase checkpoint modulated by CDK2. Cyclin E activates CDK2. So when it's upregulated either through amplification or through other means, it really is like hitting the gas pedal on that kinase, driving cells through the checkpoint prematurely into S phase and through S phase, which results in the accumulation of DNA damage. So PKMYT1 normally functions in the other key cell cycle checkpoint, the G2M checkpoint modulated by CDK1. It normally phosphorylates and inhibits CDK1, allowing the brakes to be placed on the cell cycle so that the cells can repair their DNA before they go into mitosis because when cells go into mitosis with a lot of DNA damage, they undergo a process called mitotic catastrophe, which results in cell death. So in the case of unstable genomes, inhibiting PKMYT1 is catastrophic for tumor cells. And again, there's a great synthetic lethality concept here. You can genetically select patients that either have amplifications in the gene encoding cycle, which is CCNE1. But also, there are some other genetic factors that also result in unstable genomes. FBXW7 goes for ubiquitin ligase that downregulates cyclin Es. So when it's mutated, Cyclin E levels accumulate, and then protein phosphatase 2a or the gene encoding it. Also, when there are mutations or deletions in that gene cells also undergo replication stress because of its feeding into various cell cycle pathways. So this represents just those genetic anomalies represent a really attractive patient population. Cyclin E is amplified in a wide range of gynecologic tumors, including uterine sarcomas, ovarian cancers and endometrial cancer as well as others. The ubiquitin ligase, FBXW7 is mutated also in a bunch of -- sorry, in a large group of uterine sarcomas, endometrial cancers. But importantly, also in about 15% of colorectal cancer. So that really represents quite a large patient population and one that we really are very interested in serving. Okay. So let's get into some data with this compound. So this is another program that benefited from our structural approach and structural enablement. And you can see a crystal structure of PKMYT-1 with the inhibitor bound, the active site. We used our structural data to really help drive selectivity and potency in this program, and that program fit in XL495. The lead clinical comparator right now is from repair, RP-6306, also known as Lunresertib or [resertib]. As you can see, the -- we identified some issues with that compound, namely in selectivity and also potentially in pharmacokinetics with a predicted half-life in humans of 2 hours. That was a prediction that we made based on our own in vitro assays, and that prediction turned out to be quite accurate in that repair recently presented some of their data on this compound and showed that in the clinic, they're getting a half-life of 2.8 hours. So their recommended Phase II dose is a BID dose. Whereas our recommended Phase II dose or our recommended dose or a predicted dose in humans is a once-daily dose. Now if you look at the potency again, this kind of gets back to one of the concepts I talked about at the very beginning. The potency of our molecule is not an in vitro assays is not higher than the repair compound. In fact, some people may look at that and say, "Oh, you lost potency, isn't that a detriment. But as I mentioned earlier, it's really not around any one specific assay. We look at a range of factors to really understand that we're bringing the best possible drug to patients. It's really at the end of the day, it's all about PK/PD, making sure that you have complete coverage of the target to deliver the right kind of efficacy for patients. So at our predicted dose in humans, we're predicting 100% target coverage at our predicted dose, whereas the predictions we've made for the repair compounds show that it's not quite reaching 100% at their recommended Phase II dose. So we think that, that's really going to be the benefit of this molecule and its hook for generating differentiating data. So we also think that part of this -- part of the improved -- the improvement for this molecule is around selectivity. We saw that the competitor compound was targeting about 30 kinases from the broad kinome panel. So we definitely sought to improve that profile and we achieved that with XL495 with about 35% reduction in off-target kinases that are inhibited at 100x the concentration that is required for inhibiting the primary target in cells. So again, we think this is an advancement and has the potential for best-in-class in the clinic. So let's look at a little bit of data on this compound now in efficacy studies. So in various preclinical models that show either amplified Cyclin E or increased Cyclin E levels due to mutation of the E3 ligase, FBXW7, we see really phenomenal potency with this drug. In some cases, complete tumor stasis or even some regression. And importantly, when we compare it to the maximum tolerated dose of the repair compound, we're seeing better efficacy. Repair themselves have not been able to go over 15 mg BID with their compound. It's not tolerated at doses any higher than that, whereas ours can go quite higher than that. Now if you look at the absolute dose of our compound compared to the repair compound, you'll say, "Oh, but it looks really a lot less potent. It's dosed at 150 mg BID. Why is that?" Well, that's really because of the pharmacokinetics in mice. Our predicted pharmacokinetics in mice for this compound are completely in line with what we're seeing. It's much more highly labile in mice than in humans. So again, our predicted dose in humans indicates that we will be able to dose once daily with this compound at a very reasonable dose. So importantly, with this molecule, as I mentioned earlier, synthetic lethality is typically viewed as a background -- a genetic background that creates the synthetic lethality. But we're also looking at the opportunity to create the background chemically by combining with drugs that cause DNA damage. And there, we see very high potential for combinations with this drug. XL495, combined with chemotherapies that are known to damage DNA, namely cisplatin or 5-FU. Even combinations with XL309, our USP1 inhibitor or with Olaparib, another single-strand DNA repair inhibitor, both of which can create damage to DNA show really supreme synergy in in-vitro assays. When we test some of these combinations in mice and in vivo efficacy studies, we really see beautiful efficacy. So the combination with gemcitabine in a colorectal model with increased Cyclin E levels due to FBXW7 mutation. We see very potent synergy. Both molecules or both treatments alone have no activity. We're going in with very low doses. But when we combine those doses, we see complete tumor growth inhibition with the combination. And then similarly with irinotecan, a topoisomerase 1 inhibitor. We're also seeing really nice combinatorial efficacy with this molecule. We're excited about that for obvious reasons. There are a number of ADCs both approved as well as in our own pipeline with topoisomerase payloads. So the potential for combining this molecule with ADCs with that type of payload or with topoisomerase 1 chemotherapy, for example, in colorectal cancer patients really gives us a lot of opportunity for moving the needle for patients with this program. So finally, the last compound I'm going to tell you about comes from our PLK4 inhibitor program. I've gone a little bit long, so I'm going to try to get through this as quickly as possible. But again, we're really excited about this program because it's another synthetic lethality hypothesis that has to do with the cell cycle. So PLK4 is a well-known cell cycle kinase that controls central duplication during S Phase. Centrioles are critical for mitosis. They form -- they help form the mitotic spindle. They're the nucleation points for microtubules so they can pull the chromosomes apart, during the process of mitosis. And without centriole duplication, cells can still undergo mitosis, but it's very inefficient. And they do that by nucleating the microtubules on the -- what's called the percentriolar material itself. And actually, PLK4 is one of the components of pericentriolar material. So when you inhibit PLK4 in a normal cell, it really doesn't have an impact or that much of an impact. Even though they're not duplicating their centrioles, they're still forming spindles, mitotic spindles, and they're still going through mitosis. However, under the genetic background of TRIM37 amplification, it's a completely different story. TRIM37 is an important ubiquitin ligase that actually ubiquitinatea number of different components of the pericentriolar material. In some cases, it changes how they interact with the centrioles and in other cases, it actually down-regulates them. And it down-regulates a very important component known as SEP-1192. In that case, the cells no longer have a point of nucleation for the microtubules. They cannot separate their chromosomes and they undergo mitotic catastrophe and they die. So TRIM37, again, is an important synthetic lethal genetic anomaly that's actually been described in a range of different patient populations. The biggest ones are in breast and lung cancer patients, and it represents probably 50,000 or 60,000 patients. But another one is neuroblastoma. Neuroblastoma is a very small indication. There are only about 1,000 patients in the U.S. every year, but 80% of those patients show high levels of amplification of this gene. So that represents a significant opportunity to bring benefit to an underserved patient population. So just a bit of data from this program. I think I might have mentioned earlier, this program really started or gained speed once we fully built out all of our platforms. So it really benefited tremendously from structural enablement. We had the crystal structures in place before the chemists even started synthesizing molecules. And then they started synthesizing molecules with our rapid library construction approach. So they were evolving scaffolds very rapidly. And within the course of just a few weeks, they made the advance shown here on this slide, going from compound A to compound C. They're able to improve potency by logs of activity from greater than 20 micromolar down to about 75 nanomolar. So that really shows the power of these types of technologies, which we're now using on every new small molecule program going forward. So the only competitor that we really had access to structures for at the time that we are engaging this program early on was from ORIC, a patent from ORIC. They still don't have a compound in the clinic, but we synthesized a range of compounds from their patent and pick the best ones. And this actually shows the profile of the best ones. There are significant deficiencies in selectivity with this compound. And the important point to really focus on is the differential activity in cells in TRIM37 amplified cells versus nonamplified cells because, again, that underlies the synthetic lethality hypothesis. The wider distance you can get between those IC50s will translate to the best safety profile of your compound. And you can see that we achieved about a tenfold improvement in what they -- what we saw with their compound. That translates to beautiful efficacy in a TRIM37 amplified neuroblastoma xenograft model with our compound where we see deep regression with a very low dose of our compound. Now the other comparator here is repair. They announced recently that they are going to be bringing a compound forward to the clinic. We're still generating data with some of the -- what we feel are the best compounds from their patent. But from their own data, what we found in their presentation was that their differential for TRIM37 amplified versus non-amplified. They don't give the IC50s, but they show the curves and our estimation on the curve is that there's only about a fivefold difference. So similar to what issue with the ARA comparator. So that may be a liability with their compound. We won't know that until we actually get the data ourselves and compare it side by side. But so far, we think our compounds have a solid chance at showing differentiating data in the clinic. And so far, we're still in the early days. We have some pre-DCs, but we're still evolving these and refining the data. So stay tuned for more on this compound. We expect to declare the DC soon in 2024 and file the IND in 2025. Okay. So just to sum everything up, I really hope I provided you a clearer view on how our investment in R&D has really paid off in terms of generating exciting new development compounds that have the potential to move the needle for patients and really differentiate in the clinic. So just due to the time constraints today, I was not able to go into the earlier programs, but behind these, we have a whole really host of both biotherapeutics and small molecule discovery programs that are at earlier stages of development. And hopefully, you'll be hearing about those in the near future. But again, we've kind of hit steady state now. We're on track for filing 2 to 3 INDs a year for the next couple of years. And then we'll be in a steady state of 2 a year for the foreseeable future until we generate the next big blockbuster and can fuel more growth and more discovery. Okay. So thank you for your attention. And now we have a break. So enjoy your break. [Break]

Please welcome back Dana Aftab.

All right. Welcome back from the break. So before Monty goes on, I just wanted to spend a few moments talking about Zanzalintinib. Now as we said at the beginning, Mike talked about the cabo lens, I talked about that a little bit as well. It's very important to us in terms of how we proceed with all aspects of our R&D. Now no place would benefit more from the cabo lens than cabo itself. So when we really first restarted our discovery efforts, the very first program that we focused on was cabo. No one knew that molecule better than us. We understood the target biology supremely well and had a deep understanding of it and also a deep conviction that, that was really an important driver, a critical driver of its differentiating activity in the clinic. Targeting that spectrum of kinases, namely MET, VEGF receptors, AXL, RET and a few others really addressed the target biology of renal cell carcinoma better than any other VEGFR TKI that had come before it. So we knew that molecule well had a deep understanding of the target biology and a deep conviction that, that target profile was really among the best to address these patient populations that we're serving with cabozantinib. But we also knew what we would have improved if we could have turned back the clock and made additional changes to the molecule. So that really was the program that we undertook that resulted in zanzalintinib. We understood that with a next-generation TKI that has the background of cabo. An improvement would be on our ability to manage adverse events. As many of you may know, cabozantinib has unusual pharmacokinetics. It has a 5-day half-life, approximately 99 hours, which makes managing adverse events challenging for treating physicians. If patients have a dose omin toxicity, hand-foot syndrome, fatigue, something else and they need to dose reduce, that can be challenging because it takes a long time to wash that drug out, reestablish sort of a safe profile for the patient either at a lower dose or having to go on a dose holiday to get them out of trouble and then reestablish treatment. In that case, we might even -- we might miss an opportunity in that their tumor might progress during that period. So we set out to specifically craft a molecule with an identical target profile to cabozantinib but with improved pharmacokinetics to really help manage adverse events. And we really believe we achieved that with cabozantinib -- I'm sorry, with zanzalintinib. Both molecules are highly potent against their intended targets of MET, VEGFR2, AXL and MER. And then when we compare the compounds in a very large kinase profile or kinase panel, side-by-side, we see very similar activities. So nothing really pops up as a big difference between those compounds. And that was really our intention, which is to achieve a very similar target profile. But we also changed the molecule in ways that we knew it impact half-life or should impact half-life. And we did -- and we definitely achieved that. And in the clinic, now we're seeing a much more manageable half-life of about a day. So -- and importantly, when we profile these compounds, again, side-by-side in very similar tumor models or identical tumor models, in this case, we're looking at the CT26 model, which is sensitive to PD-1 inhibition. We see very similar potency for single-agent activity and very similar combined efficacy when we combine with a PD-1 inhibitor. So we believe we really achieved the goal with this compound. Now I just want to point out that impacting half-life is not an actual biological parameter that is a simple parameter. It results from a complex series of parameters that we refer to as ADME. You actually have to alter the ADME profile of the compound to change its half-life and its pharmacokinetics. And as I'm sure most of you know, ADME refers to absorption, distribution, metabolism and excretion. So when you change those properties, you change a lot of other potential aspects of the compound that relate to PK/PD, meaning the ability of the tumor to get to its target and exert its effects on the target because that doesn't just rely on the potency against that target. It relies on a lot of factors: protein binding in the plasma, protein binding in the tissue, protein binding in the tumor, which is the intended tissue of targeting. Protein binding in normal tissues, which might be unintended tissues of targeting. So as you'll hear from Monty, in just a moment, we're really seeing differentiated activity with zanzalintinib. And we believe that the altered PK/PD profiles as they relate to the tumor versus other tissues, may really underline the differences we're seeing in the risk benefit profile of these two agents. And as Amy -- or as PJ said, we really believe that zanzalintinib is really the best-in-class molecule. It will really outperform cabozatinib when we are able to generate the data to prove that. So to tell you more about zanzalintinib, I'd like to introduce you to Monty Pal. Monty is a professor in the Department of Medical Oncology and a well-known medical oncologist and physician in the area of GU oncology. He's at the City of Hope Cancer Center in L.A. And Monty has been a longtime collaborator of the company. He actually was one of the investigators on the very first trial that investigated cabozantinib in patients with clear cell renal carcinoma, the drug-drug interaction trial we often referred to as XL184009. So he was one of the first physicians to treat patients with kidney cancer with cabozantinib. So he's really one of the experts, and we're very pleased to have him here to tell you about zanzalintinib today. So without further ado, Monty Pal.

Well, I'm really so delighted to be here today. And Dana, that was a very kind introduction. It got me thinking about my history with Exelixis and the development initially of cabozantinib, which I knew coming right out of my fellowship as XL184. And it was just an amazing opportunity to treat some of the first patients with cabozantinib with renal cell carcinoma. And Capital, it was an amazing experience. I was just on the phone about a month ago with the patient of mine who moved from Los Angeles to Texas. He was actually on cabozantanib or XL184 at the time for a total of 11 years. And mind you, this is at a point in time at which the median survival for renal cell carcinoma, we projected at a year to 2 years. So it really underscores the transformative impact, I think, that you at Exelixis have had in renal cell carcinoma, and I applaud you for all the amazing work that you've done there and multiple other malignancies. And it really makes me quite excited, enthusiastic about zanzalintinib. And as I'll highlight for you today, I think that, that really represents a new potential pillar of therapy in advanced renal cell carcinoma. Today, I'm going to be sharing with you slides that I presented at the International Kidney Cancer Symposium, which is a meeting of kidney cancer experts worldwide. This year, it was held in Nashville, Tennessee in October. And I can tell you that the presentation was really met with a lot of enthusiasm from my colleagues in the field. On the next slide here, I've given some background that I think actually Dana and PJ and Mike have already really eloquently spoken to. What's fundamental about renal cell carcinoma is that it's a VEGF-driven disease. But what's really unique about cabozantinib and now zanzalintinib is this added RTK activity, specifically against targets like MET and the TAM kinases. And I think this really lends itself not just to inhibition of tumor growth by angiogenesis, but also inhibition of immune suppression in the tumor microenvironment. So I think there are several mechanisms at play here that really made cabozantinib and now zanzalintinib stand out in the pack. But beyond that -- and again, this is just to reiterate what Dana just said, I think there's unique properties associated with zanzalintinib such as the short half-life. And I really want to emphasize both the efficacy in this presentation and the safety of zanzalintinib because many of the principles, I think, that Dana outlined really translate to what I view to be both superior safety and efficacy that we see with the compound. So we started this assessment of zanzalintinib with a dose escalation, and that included a total of 49 patients. We really did a careful exploration, and I was on a lot of these calls of dose of zanzalintinib ranging from 10 milligrams to 140 milligrams and ultimately settled in on the dose of 100 milligrams that you see utilized in studies to date. We did an expansion cohort that I'll be focusing on today in clear cell renal cell carcinoma. This is a total of 32 patients. And bear in mind, this was a heavily pretreated population. All patients in this cohort had 1 to 3 prior lines of therapy. The primary endpoint in this experience was objective response rate in PFS at the 6-month mark. Of course, as I mentioned, we're also going to delve deeply into safety in my presentation today here. Now there are some elements of the demographics in the study, they're characteristic of renal cell carcinoma. So for instance, you see that the median age of this population was 65. Renal cell carton has mill predominance. So you see that in this case, about 3/4 of patients were males, you might expect. But there are certain elements of this demographic table that allude to the fact that in this cohort, we were dealing with patients who had a very aggressive phenotype of disease. So for instance, 38% of patients in this experience had liver metastases. And again, that's really a harbinger of a poor prognosis in this particular setting. I'll also emphasize that 1/3 of patients had bony metastatic disease. And again, this correlates with inferior outcomes by and large, for patients with renal cell carcinoma. And again, we allowed 1 to 3 prior therapies, but most patients are really at that upper end of the spectrum. You can see here that 41% of patients had received 3 or more prior lines of treatment prior to entering into this experience. I think as you look at the prior therapies, it really reflects a very contemporary experience. You can see here that the vast majority of patients, all but 1 had received prior checkpoint inhibitors. So this really reflects the current treatment landscape of the disease. And beyond that, many patients in this experience -- and this is something I'll allude to, had received both VEGF TKIs, and more specifically, cabozantinib. So bear in mind that in this experience, 80% of patients have gotten prior TKI. And within that, 53% of patients had received prior cabozantinib therapy. There's other elements in this demographics table that I think warrants some mention too. If we look at the best response to the prior therapies that patients had received. Here, you only see 9% of patients that had a partial response to their immediate prior line of treatment. And that's notable because sometimes in these early Phase I experiences, you'll have patients that are serial responders. Not so in this study, you can see that best response of stable disease was achieved by 50% of patients in the prior line of therapy, and about 1/3 of patients really blew through their prior line of treatment with primary progressive disease. So again, I think the amalgam of these characteristics that I've outlined really alluded to the fact that patients in this experience had an aggressive phenotype. Despite that, what we see here, I thought was really quite remarkable. The response rate was 38% in this cohort. We had 12 PRs, the disease control rate, 88%. And if you look at the waterfall plot here, you see coated in blue those patients that had prior cabozantinib therapy. So if you're looking at the various patients indicated here, blue indicates prior cabo utilization. On the right-hand side in the bar graph, you'll see highlighted the specific outcomes for patients that have received prior cabozantinib therapy. And recall that this represented 53% of the patient population. And if you look at those patients that had gotten prior cabozantinib, this is really what really demonstrated to me that this drug has some unique activity here, zanzalintinib resulted in a response rate of 24%. The disease control rate here, 94%. So I thought that was quite remarkable. And really, if you look at the right here, the response rate of 57% for zanzalintinib in patients that had not received prior cabozantinib I think really represents a high watermark amongst many of the salvage therapy options that we have available. If you're wondering about the 6 patients who had no prior TKI exposure, the response rate there was 50%, 3 patients were partial responders. And importantly, what's also highlighted at the box in the bottom right there is the fact that 3 of 4 prior cabo exposed patients actually had discontinued prior cabozantinib on account of disease progression. We're going to expand this bar plot here. Going back up just 1 slide, please. Okay. Yes, there we go. There should be a build slide there. We can come back to that. These spider plots here, I think, really nicely articulate the fact that we have many patients who have had protracted benefit from therapy with zanzalintinib. On the left-hand side, you see patients who are cabozantinib naive, so that amounts to 14 patients. And within that cohort, the median duration of response was 7.4 months. If we look at cabozantinib exposed patients, 17 patients in this experience. That median duration of response hasn't yet been reached within this cohort. And the arrows that you see there indicate that many patients are having ongoing responses. So I certainly look forward to updating this data set with zanzalintinib for you in the near future. If we look at a median follow-up of 8.3 months, 50% of patients were still continue on treatment, that alludes to the fact that, again, I expect this data to only get better with time as we continue to update it. And this is an important characteristic as we look at other agents within this landscape. 75% of patients incurred responses at their first post baseline tumor assessment. And that's important because as we think of other agents that we can potentially utilize in the salvage setting, they're quite slow to evoke that response. But what we really need both in this setting and frankly, in the upfront setting for any patient that has substantial tumor burden, that has symptomatic disease is a really rapid response. And so I think that second bullet point really alludes to that. I'll just also point out, in the third bullet point here, as of September 6, 6 of the ongoing patients have been on zanzalintinib longer than their most recent prior therapy. So we're starting to see this cohort of patients emerge within this data set, who have actually been on zanzalintinib longer than their immediate prior therapy, which I'm always encouraged by. Okay. So it looks like we've got that build slide back up on the screen now, which is great. So the one bar graphic that I didn't get to was shown here on the far right side, and that includes patients that had received a prior VEGF, TKI. So recall that constitutes 80% of patients that were in this experience. And if you look at those patients inclusive of cabozantinib, the response rate to zanzalintinib was 35% with a disease control rate of 92%. If we exclude cabozantinib there, you have every single patient within this particular subset driving clinical benefit with a response rate of 63%. So again, as I look across the landscape of monotherapy data sets that I've seen in the salvage setting for renal cell carcinoma, this really stands out to me as being the high watermark. What I didn't do in my presentation at the International Kidney Cancer Symposium that I've got an opportunity to do today is to highlight for you some really unique responses that we saw. And I think this underscores the potential activity and ultimate placement of the drug. So this is -- these are 2 patients that I'll present to you who have received zanzalintinib monotherapy in the context of this trial. The first is a 65-year-old gentleman with metastatic renal cell carcinoma that felt recurrence 8 years after his initial nephrectomy. He had metastases to both the lung and to the bone. In terms of his treatment history, he's received what has really sort of demonstrated to be a very active regimen with cabozantinib and belzutifan in the frontline setting. He did have a best response of a partial response in this case, but discontinued therapy after a year on account of toxicity. This patient then received nivolumab and progressed after a mere 3 months on the drug. So this patient then started on zanzalintinib at 100 milligrams and actually had a confirmed partial response at the second post-baseline scan and his bone lesions at this point had completely resolved. So just to really get into the details, and I apologize to those of you on the webcast, you won't see my laser pointer here. But if you look at the baseline assessment here, you can see this large tumor that approximates the cardiac border over here in the left lung. And you can see that over the course of time, and this is by the second baseline assessment over here, we have a 53% reduction in the size of the lesion. You can see that better here. And then as we approach week 33 on therapy, this lesion has completely disappeared. This is not something I'm used to seeing in the salvage setting of renal cell carcinoma. So I think this is incredibly compelling, especially given consideration to this patient's prior therapies. This patient did develop some new brain lesions at week 33. Notably, he didn't require any dose reductions during the course of therapy. This is another impressive case that I wanted to point out. This is a 67-year-old gentleman with metastatic clear cell RCC to the right adrenal gland. He had received cabo nivo for a year with the best response of stabilization of disease, and then received 10 months of pembrolizumab with an investigational agent with the best response of PD. So keep in mind, throughout this patient's treatment trajectory, he has not yet achieved a response to therapy. He started on zanzalintinib at 100 milligrams daily. And I'll outline for you here again. Apologies to those on the webcast who don't see my laser pointer. This patient developed a partial response at the time of the second scan. You can see the liver border here. The adrenal lesion right here that's outlined in yellow. And you can see that it's really shrunk down substantially by the time of that second response assessment. Those patients subsequently developed some progression. But as has been my experience was really able to tolerate therapy with zanzalintinib for a very protracted period of time. This patient remained on therapy for 77 months. So you can see that, that duration exceeds their duration of therapy on initial treatment with cabozantinib with belzutifan also exceeds their duration of therapy with pembrolizumab with the investigational agent. So I thought that was very, very impressive. And I mentioned from the outset that I really wanted to emphasize a discussion of safety alongside our discussion of efficacy, which I also think really stands out. If we look at the safety data from this cohort of patients, the 32 treated with zanzalintinib here, we have a median exposure of 6.4 months. Now what really sort of struck me as I was reviewing this data is that if we look at Grade 4 treatment-related adverse events, these are the ones that we as investigators, really chalk up to the drug, you see that no grade 4 events, no grade 5 events were incurred with zanzalintinib. The treatment-related adverse events, those could be ascribed to other factors, perhaps concomitant medications, underlying conditions, the disease itself. But seeing that there were no events related to zanzalintinib was incredibly encouraging. And again, I think, marks a distinction in terms of toxicity compared to other agents in this space. 50% of patients ultimately warranted dose reductions, 69% had dose holds. But the other important characteristic that I draw your attention to here is that only 9% of patients required discontinuation due to adverse events. And again, I think that represents a relatively low number in this particular context. I'm going to dive much deeper into the toxicity data here within this slide. And I wanted to underscore for you how different this is amongst other therapies that we currently have available for renal cell carci. Now we've done this by offering a just a position of the data that we have in this cohort for zanzalintinib against the accumulated data we have for cabozantinib over time. And I think this is a fair just position. If we look to some of the characteristics that really complicate my day-to-day use of TKIs in my clinical practice. Diarrhea, hand-foot syndrome, some of these really stand at the top of the list. If we look first at diarrhea, at first blush, it might seem that the rates of diarrhea are fairly comparable, 69% for zanzalintinib versus 60% for cabozantinib. But look at that rate of grade 3 or greater diarrhea, this is really what gets us into trouble, what prompts dose reductions in the clinic. You can see that's 3% for zanzalintinib and contrast 11% for cabozatinib. I highlighted in green some of the other toxicities that again, really get in the way of dose administration for me and the clinics with most TKIs. And again, these occur in a very limited fashion with zanzalintinib by comparison. The rate of fatigue in this experience, 19% was zanzalintinib versus 53% with cabozantinib, no grade 3/4 toxicity. The next two, I think, are very critical. AST and ALT increases. You can see here that there's no Grade 3, 4 AST/ALT increases that we observed with zanzalintinib within this experience versus observation of many in the context of cabozantinib. And the reason I think this is particularly important is it really lends itself to combinability with immunotherapy. If I think to initial experiences with, for instance, sunitinib with nivolumab or pazopanib with nivolumab, some of our initial forays into combining these drugs, it was really liver function test abnormalities that got us into trouble. In contrast, I think that this is -- this profile here is going to lend itself quite well to those combos. And I have a slide that addresses that in a bit more detail later. Hand-foot syndrome, listed here is palmerplant,or erythrodysesthesia, or PPE, is an incredibly challenging toxicity that we incur with virtually all TKIs. The rate here at 9% was zanzalintinib versus 39% with cabozantinib. So a huge discordance there, and again, 0 in the way of Grade 3, 4 toxicities. And this is real. I've used this drug a fair amount in my clinical practice and zanzalintinib really stands out to me in terms of the tox profile. I've certainly had an experience very similar to the data that I presented here with far fewer in the way of dose reductions and discontinuations on account of toxicity. I wanted to highlight PPE polymer plant or rather dysesthesia, as we more commonly refer to it hand-foot syndrome for you. Again, this is side effect of that is really sort of classic for many TKIs. This lends itself to a really painful and debilitated swelling on the ponds of the hands or souls of the feet. We have several attempts that supportive care modalities to address this, but nothing that's truly effective outside of dose reduction or dose holds as alluded to in the slide. On the right-hand side, you can see two panels that outline the manifestations of these toxicities in patients. And it really gets in the way of day-to-day activities whether it's writing a bike, driving a car or simply using a typewriter, a keyboard. So all of these day-to-day activities really get truly impacted by this. And it's really encouraging to see that those rates are mitigated substantially with zanzalintinib. In this next slide here, I alluded to the potential for zanzalintinib in a combination strategy. In the STELLAR-001 experience, we had the opportunity to review anzolitinib, not just as monotherapy but also in combination with atezolizumab and avelumab. And I think this is a fair juxtaposition of the STELLAR-001 regimens against CheckMate 9ER, KEYNOTE-426 and clear, at least to get some initial assemblance of what the tox profile of these combinations might look like. And what we see here really resembles my experience with zanzalintinib monotherapy versus TKI monotherapies, and that this combination seems like it's going to be safer and better tolerated than some of the others that we see in this particular space. So I'll first start at the very bottom. I referred to the propensity of this regimen to succeed because of the low rates of AST, ALT increases. That's exactly what we're seeing manifesting here. If you look at, for instance, KEYNOTE-426, you see a 7% and 13% rate of grade 3, 4 AST, ALT increases versus just under 2% and under 3% with zanzalintinib. The rates of diarrhea, similarly, far lower than what we see across all the other TKI I-O combinations. The same holds true for fatigue, hand-foot syndrome and all the other elements that I think have really challenged the utilization of TKI I-O combinations in the clinic. I've put a couple of bullet points here to summarize this really sort of broad array of data that I've covered here. I think it's perhaps best summarized with the first bullet point, which is the single agent zanzalintinib. Really demonstrates incredibly promising antitumor activity in patients with heavily pretreated advanced clear cell renal cell carcinoma. This response rate of 38% is impressive in and of itself. But when you really consider those elements that I highlighted in the demography of this population, this is a very aggressive phenotype of patients that we were treating in this experience. I think it makes it all the more impressive. The second bullet point also important. You can see here that antitumor activity was observed with zanzalintinib, not just in the context of patients that were TKI-naive or pretreated with immunotherapy alone, but also in the context of patients that have received prior cabozantinib therapy. And that suggests and really dovetails on some of the points that Dana made around the uniqueness of this agent and the approach that we're taking with it. This objective response rate of 57% in cabo-naive patients, again, I think really represents a high watermark amongst monotherapy strategies that I've seen in the salvage setting for renal cell carcinoma. That response rate of 24% in patients previously treated with cabozan, it also tells me that we're offering patients something different with this approach. Zanzalintinib also appears to be very well tolerated even in VEGFR-TKI pretreated patients. You can see that we had low rates of discontinuation, 9% in this experience, which I think is incredibly encouraging. And these rates of hand-foot syndrome, AST/ALT increases all the things, again, that are getting a way of my day-to-day practice with currently available TKIs, we really mitigate with zanzalintinib in my opinion. And again, in that last set of slides there, we demonstrated that full dose zanzalintinib can be effectively combined with immune checkpoint inhibitors, including atezolizumab and other checkpoint inhibitors but I think, a very reasonable safety profile. And again, some of those properties of zanzalintinib, I think, are going to really lend itself to success and combination approaches as well as monotherapy. So I'm going to conclude my remarks there, and it's really my honor to introduce Dr. Amy Peterson, the Chief Medical Officer of Exelixis, who's going to be discussing with us further the development strategy. Amy?

All right. Well, thank you, Dr. Pal. I appreciate the accurate introduction of my name. Just kidding, Dana. I've only been at Exelixis for about 3 months, but I actually have been intimately involved in clinical development, clinical trial execution for decades, really both from a company-sponsored standpoint participating in sponsored studies at the University of Chicago where I did my training, as well as designing and enrolling into investigator-initiated studies. I've been at a variety of companies, large companies like Genentech Roche, small companies like Medivation, CytomX and growing companies like Beijing. I think from each experience, I have learned quite a lot, and I'm very excited to actually bring those experiences and learnings here to Exelixis to help develop what I think is a really exciting broad pipeline and to realize really the potential for patients. So I'm thrilled to be able to speak to you today about our development strategy and clinical pipeline. And if you're wondering why we're all looking down, it's because our slides are here so that I figured you'd rather see our front than our back, but I will try and point to the slide as much as possible. So what's our special sauce for development at Exelixis? Well, it starts with cabo. You've heard that a lot. This is our flagship molecule, multiple approvals, hopefully, more to come soon. We're applying the lessons that we learn from cabo to better develop other drugs like zanzalintinib. Because zanza mimics the target profile of cabo, while curtailing the half-life, you've heard that again and again and again, it could have even better activity in indications where cabo is active. So we -- that -- what that means is we can and should and are obligated really to look into all the settings where cabo has been successful but not only the pivotal studies. Also, there are a lot of earlier studies with cabo, and we need to consider how best to accelerate zanza in some of those indications that we hadn't yet pursued with cabo. We have forged successful and broad clinical collaborations with investigators but also with companies. And we can leverage that for zanza to allow for cost sharing, which can result in financial derisking and the ability then to translate those cost savings into even more development. For the pipeline, as you heard from Dana, we are really tuned into best-in-class. This allows for differentiation from -- in the clinic and an opportunity to do really even more for our patients, right? Maximize our success, minimize without focusing on minimizing the failures. We have a lot to do, though, and one has to be thoughtful and strategic. In my opinion, there are really three competing strategies to balance while laying out early development of any given asset. That is the probability of success includes technical and regulatory success. Time or speed to market and value. Most early-stage companies will focus on probability of success, right? Make sure you get an approval. Do it quickly, do it cheaply. Smaller companies without revenue typically have one shot on goal to get the asset approved. But at Exelixis, not only does cabo provide a road map to help us consider zanza development, but it provides cash flow so that we don't have to focus only on PTS and time. We can actually take some smart risks, cast a wider net. Finally, we have to ensure that we maintain a sense of urgency that is synonymous with smaller companies and can get lost with big co. To improve our long-term value creation, we need to be efficient with our decisions. We need to scale appropriately and be thoughtful with our clinical trial spend. We need to do this, right, because we have a lot to work on not only now with 3 molecules in the clinic, but 3 more coming next year, 3 more after that, as you heard from Dana. So of the pipeline agents that are currently in the clinic, zanza is the most advanced. Three pivotal studies approaching nearly 1,000 patients treated, and we're looking forward to sharing data from the clinical programs as it matures. XB002, our tissue factor ADC modified orostatin payload, similar target, though, to Tibdec, but the actual epitope that's recognized by XB002 is different. All rationally designed to hopefully offer a differentiated toxicity profile that minimizes neuropathy and bleeding risks without compromising activity. And just a month ago, we announced the in-licensing of XL309, our USP1 inhibitor from in silico and we intend to investigate the best-in-class potential of this asset, not only to deepen and potentially prolong the responses seen with PARP inhibition in sensitive tumors, but to also broaden the eligible patient population. So I'll start first with zanzalintinib, our third generation TKI. So as you've been told and as we have shown, zanza has a very similar target profile to cabo and has potent activity against some key targets not only involved in angiogenesis, but cell proliferation, metastasis, immunosuppression, the latter of which we believe really lends it to be a partner of choice with IO-based therapy. We believe it has best-in-class potential and indeed are seeing responses in the patients who progress on early generation TKIs, including cabo, which Monty just shared with you from our STELLAR-001 expansion cohort. And with the half-life of a day, zanza does get to study state quickly and if dose de-escalation is needed or dose reductions are needed to treat adverse events, it can actually get to -- the serum concentrations can drop rather quickly for more efficient relief from these adverse events. So all of these characteristics culminate into zanza. I'll use the analogy that Dana used wearing the crown of best-in-class, not only in kidney cancer, but in other indications and importantly, in combinations where this target profile has potential. So speaking to the combination potential of zanza, the development opportunity is vast. Tolerability profile and immune permissive features of zanza lend it to be a partner of choice in combination with IO. We have explored zanza in combination with multiple checkpoint inhibitors and have initiated pivotal studies with three of them. But we really believe this to be the tip of the iceberg. We're combining zanza with additional IO combinations. So doublets, triplets, we're looking at lag CTLA4 and TIGIT. We just announced our collaboration with Arcus to combine zanza with their HIF2 inhibitor in STELLAR-009. We're also interested in combining zanza with our own pipeline. So for example, can Zanza be combined with XB002, our ADC to Disubactor. And as we better understand the emerging toxicity profile and tolerability, we can also investigate whether or not combinations with chemotherapy could be done with zanza to unlock additional opportunities. Many of the combinations that I just mentioned are being evaluated in these 4 Phase I/II studies with zanza. And these -- it's through these studies that we're really able to interrogate a vast array of tumors. You can see them listed at the bottom boxes all across. And it really blankets what I would consider the spectrum of solid tumor oncology. So that we can hone in on those that make the most sense to develop further. Given the breadth of clinical data that's been generated with cabo, you can see that in the heat map over on the right. It's been generated either alone or in combination with immune checkpoint inhibition. We need to be focused and disciplined when we consider which indications and combinations to do first. There are more opportunities than we have resources to investigate. And so we need to constantly consider and balance Pts, time to market and value, with the ever-changing competitive landscape. And when considering the development of zanza, we need to take into account where do we need to generate data with Zanza versus where can we capitalize on cabo data, where are the areas of high unmet needs, and where is there a window with regard to the competitive landscape. So I'm going to spend a little bit of time on this slide. These are the 3 pivotal studies that are emblematic of how we are able to be nimble with cabo data and take smart risks, move quickly and efficiently into studies that if positive, can help a lot of patients. Meanwhile, we will generate the relevant data with zanza to support our regulatory expectations like dose and contribution of components and those supporting studies that are there for zanza, you can see those listed underneath the design graphic. In colorectal, you can also see below the zanza row, the data that has been generated using cabo. And in colorectal, many of you know, we recently modified the endpoints and sample size based on the emerging data from at least 2 supporting companies. So BOT/BAL with Agenus and [ PEM/LEN ] with Merck. This is data generated in patients without liver metastases. We increased the sample size of the study by almost 50%, so going from 600 to 874 patients. This allows us to do with the primary endpoint analysis initially in nonliver met patients, followed by the ITT. We have observed data with cabo in the similar setting with colorectal, not only in 1 study, COSMIC-021, a cohort there where we evaluated atezo cabo, but also CAMILLA, which is shown on the slide where we evaluated cabo plus Durva in 29 patients. So in CAMILLA, we observed a response rate of 28%. This benchmarks actually quite well against what [ PEM/LEN ] showed in the LEAP-017 study of 10% in the ITT population, and really even more so, to currently are more relevant, I would say, to the currently approved therapies, which are regorafenib, Lonsurf, [indiscernible], all of them showing single-digit response rates of 1%, 1.6% and 6%, respectively. The median OS from both CAMILLA and COSMIC of 9.1 in 14 months also benchmark really nicely against the 6.4 months observed with the comparator arm of 303, which is regorafenib. Of course, we're not assessing zanza in 303. We're assessing -- sorry, we're not assessing cabo. We are assessing zanza in 303. We're assessing zanza in 303, and that we think it's even better. So a positive signal in this study, either in nonliver met patients or in the ITT sets up very nicely for zanza to be advanced in earlier lines, either as monotherapy or in combinations with IO and frontline setting with chemo or potentially in the maintenance setting. The 304 study is a really important study. There have not been any randomized Phase IIIs in non-clear cell kidney cancer. Treatment recommendations are based solely on mostly monotherapy Phase II studies, including the most recently published data from the [ PEN/LEM ] KEYNOTE-B61 cohort, where they observed a response rate of 49% in previously untreated patients. We see this population as a high unmet need and an opportunity to get zanza approved in an area of an oncology where Exelixis is a leader. The response to cabo/nivo, as shown here, and presented at ASCO this past year by [ Dr. Joe Lee ], who currently works at Exelixis was an impressive 54% in frontline and 36% in second line, and the median PFS for both lines combined was 13 months. The response rate was SUTENT, which is the comparator in the study, for 304, it was 18% in its study in frontline and the median PFS was 8 months. Finally, cabo monotherapy beat SUTENT patients with papillary -- with papillary kidney cancer in the SWOG 1500 study, where cabo response rates were 23% and SUTENT was 4%. And the median PFS was 9% versus 5.4%, respectively. So we think the Pts or the probability of success of the 304 study is quite high based on this data. STELLAR-305 is our most recently initiated pivotal study. This is a Phase II/III study, and we just announced that the first site was activated just a week ago on Monday. We believe this study is yet another example of taking a smart risk, not only because of the Phase II/III design, but also given what was observed with cabo pem in a multi-institutional center, multicenter -- sorry, multicenter Phase II study that was published in Nature Med earlier this year by [ Dr. Saba ], and it was conducted in patients with inoperable recurrent metastatic head and neck squamous cell carcinoma. And in that study that's shown here on the bottom. The combination of cabo and pembro resulted in a 52% response rate, a 14.6 month median progression-free survival, and a 22.3-month median overall survival. This benchmarks well to monotherapy pembro, which is the comparator in 305, which showed in the KEYNOTE-048 study, a response rate of 19%, a median progression-free survival of 3 months and a median overall survival of 12 months. It actually also benchmarks well to the Phase II [ PEM/LEN ] study that demonstrated a response rate of 36% and a median PFS of 8 months. And we are aware that the LEAP study conducted by Merck on head and neck was negative for OS, and obviously, we'll review the data as it becomes public to assess whether or not any further changes to the 305 study are warranted. But early signals of activity here actually can help inform additional indications with zanza IO, for example, squamous non-small cell lung cancer. So collectively, these 3 pivotal studies marry the balance of Pts, speed and value and will inform additional pivotal studies that will feed the life cycle management of Zanzalintinib. In summary, we have a development approach with Zanzalintinib that I believe to be robust, that is in partnership with a variety of companies and compounds and enables flexible cost sharing and derisking. We have and will continue to be as efficient as possible with our clinical trial design, so that we can be nimble. We can move quickly into pivotal studies that will definitively address potential of Zanzalintinib to improve outcomes and allow us to make rational decisions to advance or discontinue combinations and/or indications that no longer make sense. So is there more to do? Absolutely. I look forward to doing this thoughtfully, not flat-footedly, to advance zanza into new studies and new indications and new combinations. So let's turn now to tissue factor ADC, XB002. So Dana already discussed the rationale behind creating this ADC that recognizes a different epitope, a different tissue factor epitope, and has a unique payload. With XB002 development, we can actually apply some of the principles as we are for zanza that will best inform its path forward. In order to ensure targeted delivery of the payload, we need to focus on tumors that -- where tissue factor expression is high. But of course, we can't use that as the sole criteria given not all tumors will be sensitive to the payload. And for example, and Dana pointed this out as well, you can see in the heat map over on the right that colorectal, one of the highest expressers of tissue factor, however, colorectal cancer is not particularly sensitive to an anti-tubulin payload, and this is really not one of the indications of interest of XB002. However, our tissue factor ADC with the irinotecan like payload XB371, this is a high indication of interest for us. Additionally, we can look at the competition [ that's ahead ] to inform where to take XB002. So in this case, we have TIVDAK and we can readily benchmark our data, both in terms of activity and tolerability, ultimately to identify those areas where we can and should accelerate development. So this data was presented previously, but it's important, I think, to have in mind as we consider the potential for differentiation of XB002 from TIVDAK. On the left side are the PK exposure curves for our intact ADC of XB002 at the indicated doses that were tested in the Phase I escalation of the JEWEL-101 study. On the right is the PK exposure of free payload at corresponding doses also tested in the Phase I. So note the scale on the right only goes to 10, whereas the scale on the left goes to 1,000, overlaid on both of these graphs in fluorescent blue, yes, fluorescent blue is the exposure of [ intact to deck ] on the left and free MMAE payload, which is a different payload, observed at the labeled dose of 2 mgs per kg to deck on the right, 2 mgs per kg every 3 weeks. So when you look at the same dose of XB002 of 2 mg per kg, which is in orange, and I don't have Monty's pointer, but hopefully, you can see that. The 2 mg per kg dose is more than twofold higher in intact ADC exposure than TIVDAK, while the circulating free payload is actually more than tenfold lower. And in fact, exposures of intact ADCs of intact -- sorry, the exposure of the intact ADC for XB002 at doses of 1 mg per kg or higher, are higher than the intact exposure of TIVDAK at 2 mgs per kg, and the free payload is less than 2 mgs per kg of TIVDAK at all doses tested with XB002 up to 2.5 mgs per kg. So what does this mean? It means that lower doses of XB002 can have the same tumor penetration as TIVDAK 2 mgs per kg with less off-target toxicity due to lower amounts of free circulating payload. And based on what we already know from TIVDAK, there is potential for monotherapy activity and room for improvement with the tolerability profile. With XB002, we don't expect bleeding to be a significant event, which was observed at 62% all grade, and 5% Grade 3 with TIVDAK, and given the modified payload, we also expect neuropathy to be less, where it was observed at about 42% all grade and 8% Grade 3 on the Q3 week regimen of TIVDAK and even higher on the Q2 week. For the ocular toxicity, tissue factor is expressed in the eye. So we do expect ocular toxicity. But given the minimal bystander effect and lower free payload from XB002, it does have potential for differentiation with regard to this particular toxicity. So given the breadth of tumors where tissue factor is expressed and the monotherapy activity observed with TIVDAK, we are keen to understand the monotherapy potential of XB002, and we plan to assess each of the tumors that are listed here in the middle in the expansion cohorts with XB002, there's combination potential as well. So we have a dose escalation in combination with nivolumab. And once we understand the combination dose, we intend to interrogate the tumors that are shown on the left. And finally, we're evaluating -- not finally, but in addition, we're interested in combining with our own pipeline and with zanza, and the reason for VEGFR combination with tissue factor is intriguing, and the indications where one might consider pursuing that is listed over on the right, where we like the idea of zanza's half-life being 1 day so that as toxicity is observed, we can take the Zanzalintinib off and hopefully recover the patient from that toxicity relatively quickly. Whereas if we were combining with an antibody like bevacizumab, we have to wait about several weeks, couple of months for that toxicity to resolve. So finally, we're also evaluating the optimal dose of XB002 to fulfill Project Optimus. And Jewel is really a large Phase I/II study that will fulfill and answer quite a number of questions and objectives, and we're hopefully going to be able to share the data with you soon. So turning now to XL309. As you heard from Dana, the IND is active, and sites are now screening patients, and we're particularly excited about 309 as it has the potential to be best-in-class. And this graphic is meant to capture the potential patient population, should we demonstrate that one, XL309 can recover responses in patients who have lost their response to PARP inhibition, a small but not insignificant patient population, as you heard in PJ's talk. Whether it can deepen or prolong responses in combination with PARP inhibitor, not only in patients who carry the BRCA mutation, but potentially expand that to patients whose tumors have homologous recombination deficiencies other than BRCA, and in the go big or stay home mindset, it is possible that XL309 combined with platinums, which is -- which mechanism of action is also to inhibit DNA repair. So should the data that Dana showed you with XL495 in combination with platinums and with [ 5-FUB ] replicated with XL309, potentially first not in nonclinical specimens, but then hopefully, ultimately in the clinic. This will actually open up an incredibly broad patient population as platinums and 5-FU are indicated and given in a large array of treatment of tumors. So it's early days, and we are really looking forward to embarking on the dose escalation, both as monotherapy and in combination in the Phase I. Okay. Finally, I'll end on cabo, our flagship molecule that continues to bring meaningful benefit to patients, and hopefully, we'll continue to be advanced into new indications. My first week of joining, Exelixis announced positive results from basically 3 Phase III studies, a cabo atezo met the first dual primary endpoint of PFS versus second NHT in patients with metastatic castration-resistant prostate cancer and continues for OS. This study really embodies Exelixis' willingness to take smart risks. It was conducted after the Phase III COMET study was read out negative for OS and the COMET-2 study was terminated early. Both of those studies evaluated cabo monotherapy in prostate cancer. And to date, IO has limited to no success in monotherapy -- as monotherapy in metastatic castration-resistant prostate cancer with the exception of those few patients who have MSI-high or dMMR prostate cancer. In partnership with our academic colleagues like Dr. Pal, we actually conducted a very thoughtful and systematic review of the COMET studies and identified a population of patients namely those with visceral metastases, where a benefit from cabo could be observed and where PFS is not potentially confounded by the effects of cabo on bone metastases. The CONTACT-02 patient population is unique from other studies. And dare I say, represents the worst of the worst prognosis of patients with measurable visceral disease. Patients were required to have measurable disease. They were required to have extrapelvic disease. Bone-only disease, not allowed. Low-volume localized disease, not allowed. Patients were not required to have progressed within a certain time on their first NHT, unlike the CARD study that evaluated cabazitaxel versus second NHT. Those patients were required to have progressed within 12 months on their first NHT suggesting that this study selected a patient population that is resistant to NHT, which the study ultimately went on to show that cabazitaxel was superior to second NHT in this potentially NHT resistant patient population. Finally, prior chemotherapy for metastatic castration-sensitive prostate cancer was allowed. This is different from PSMA 4, only neoadjuvant or adjuvant chemotherapy was allowed in PSMA 4, and the patient had to have completed that at least 12 months prior to study enrollment. So we have a patient population with a very poor prognosis, but we do believe that these patients could benefit from cabo therapy or cabozantinib therapy plus the addition of IO, and we're really looking forward to presenting this data early next year and discussing it further with the FDA. An unexpected bonus, from the CABINET studies, which really are in effect 2 independently powered Phase III studies evaluating cabozantinib versus placebo in either pancreatic or extrapancreatic neuroendocrine tumors, which I will say includes lung, about 20% of the patients in the EP cohort had lung neuroendocrine tumor. So really, this is a pan neuroendocrine tumor compilation of 2 studies. And I'm going to -- a picture is worth a thousand words here. So I'm going to show you the PFS curves that were presented by Dr. Jennifer Chan at ESMO to a packed audience. So as you heard from PJ, about 8,000 treated, but 15,000 newly diagnosed patients with neuroendocrine tumor a year. due to the indolent nature of this disease, though, the prevalence is about tenfold of that, at about 170,000 in the U.S. alone. So this is very different from something like glioblastoma, where the incidence is nearly the same, about 12,000 patients per year, but the prevalence is only about 13,000, and that's because all patients die within a few years of being diagnosed with GBM. That is not the case with neuroendocrine disease. There are a lot of people out there with advancing, progressing neuroendocrine tumors and no active therapies left to try. So this debilitating disease typically associated with intractable pain because it can track along the nerves. And with endocrine syndromes, that are unpredictable when they will manifest and include symptoms like flushing, diarrhea, bronchospasm and some long-term cardiac issues. It is a health threat to many of these patients, and they need new therapies. Treatment currently consists of somatostatin analogs or radio-labeled somatostatin analog, so Octreotide, not radio-labeled. Lutathera, radiolabeled. Everolimus is indicated for pancreatic net and certain nonfunctional nets, and SUTENT for pancreatic net. Chemotherapy like temozolomide is used, but really is not thought to be effective. So what are you seeing on the slide? You're seeing 2 Kaplan-Meier curves that have both wide and sustained separation. In pancreatic net on the left, treatment with cabo results in a near-quadrupling of the progression-free survival along the course of the curve with a hazard ratio of 0.27. So this is not only at the median, where the difference went from 3 to 11.4 months. 60% of patients progressed at their first assessment. That's seen in the blue line that drops down to the 40% mark literally immediately. All patients progress by 12 months, which is the median progression-free survival on cabo. In the extrapancreatic neuroendocrine tumor group on the right, cabo results, cabo treatment results in more than a doubling of the progression-free survival. And again, across the whole spectrum of the curve, not only at the median, where it went from 3.2 to 8.3 months. And again, 60% of patients progressed at their first assessment. And these are patients who really have no other therapies. So this is a poor prognosis group of patients who need additional options. The adverse event profile from cabo was really no different than what has been described before. And we, in addition to our partners, are very much looking forward to working with the Alliance who conducted the study to file this data with our respective health authorities as soon as possible. Okay. So I covered a lot of territory in a short amount of time. I hope though that through this presentation, you can appreciate how we are really well established to continue to bring value and treatment options to patients while building and expanding additional value with our pipeline. By remaining biology centric, we let the science drive development initially focusing on validated targets like cabo with zanza, TIVDAK with tissue factor, PARP inhibition with XL309. But we're not stopping there. We will continue to advance the current and coming pipeline rationally and efficiently. We will accelerate when appropriate, to pivotal studies, and we will kill quickly, when appropriate, if differentiation, high probability of success, obvious value to patients are not forthcoming in the clinical data. Exelixis has deep strong relationships with investigators. We have excellent reputation with our vendors and CROs and are capable of making very judicious decisions when necessary, while also being aggressive with our development plans when called for. Exelixis is primed to be an even bigger oncology company that it currently is, multiple drugs with multiple indications, and I'm thrilled to be here to help shape our development programs and move them forward so that we can actually continue to do what is truly our most important job. Help patients recover stronger and live longer. Thank you for your attention. I'm going to pass the floor to Mike and I look forward to your questions.

All right. So I will wrap up fast here, and we'll move on to break for Q&A. So let me close, coming full circle to where we started today, right? As we move into 2024, Exelixis is positioned to be the biotech leader in oncology R&D on a global scale. That's our goal. That's our mission. That's what everybody is working towards every single day. And we're going to do that by taking the pipeline you saw today, and like with cabo, move the needle for patients, improve standard of care for patients with cancer. That's the goal, that's the mission. We all have a laser-like focus on that single goal, okay? Quite simply, we're planning on having multiple launches in numerous indications in the back half of the 2020s based upon the pipeline work that you saw today. That's the mission, right? And we're going to do that with great science, the great team that we have, the momentum that we have from cabo and overriding commitment to patients because we build value for patients, we build value in the company and we build value for shareholders. So thanks again for coming today. Appreciate your time. I'm going to take a quick break and get set up for Q&A. We'll do that. So come back in about 5, 10 minutes, and then we'll -- after that, we'll then break for lunch. Okay. Thank you. [Break]

We have 2 mic runners on the sides. I think Susan is taking -- whoa -- that's right. Susan is taking, I think, questions from online. So happy to have a dialogue. If you want to ask a question of a specific individual, have at it. We'll all jump in afterwards, if you want to ask a general question, then I will triage that to the appropriate parties. So okay. And we can't see you because the lights are so -- still up, but I think, I see one with -- Jeff, go ahead, please. Can you bring Jeff a mic? Yes.

So Jeff Hung, Morgan Stanley. You talked today about your biology centric modality-agnostic R&D. From a modality standpoint, are there any key areas or technologies that you think you still need to complement or improve upon your existing capabilities? And then I have a follow-up.

Okay. Thanks, Jeff. Dana?

Yes. So at this point, we're really happy with where we stand with our technologies. As I mentioned, we took some time to build these out, and we're now seeing the fruit of our labors and the data we're seeing are really encouraging. So at this point, we're exactly where we want to be and where we need to be, but we're not close to new opportunities. But right now, we want to stay focused on delivering on our goals of bringing differentiated compounds to the clinic at a pace or to our pipeline at a pace that will result at around 2 INDs per year. So currently, we're well set up for that. But I didn't tell -- talk about sort of everything we're doing. So for example, in the small molecule space, I really only highlighted synthetic lethality in the tumor microenvironment because that's really where we have compounds, sort of exiting early R&D into the pipeline. But we have a lot of other areas that we're studying. I kind of hinted at that a little bit in one of the structures in the structure slide, we are looking at some dominant oncogenic drivers and some other areas. So I would just say stay tuned to see really what's -- there's more to come. And I think there's some exciting potential for the earlier programs.

And then for XL495 on -- there was a slide that you had about the combination. Did you also look at the combination of [indiscernible] with gemcitabine and irinotecan. And if so, how do they compare to the combinations you showed with XL495?

Yes. So they showed limited data, but I think the data that they showed preclinically are similar to ours. Look, the approach we took was really especially in that program is really focused on human PK. Our ability with XL495, we think, is really going to be best-in-class or the best drug going into the clinic at this point to show the potential for on-target activity, for 100% coverage of on-target activity at the projected human dose. So we think that ultimately is what's going to translate to success in the clinic and differentiating data.

Yes, go ahead.

Greg Renza from RBC Capital Markets, and thanks, Mike and team for holding this event, for welcoming us. Mike, maybe just a broader question from you. Certainly, the [ cabo Lens ] is underpinning many aspects of the presentation and Exelixis' strategy and certainly making the compelling case for the edge that it provides you and the organization. I'm just curious, on the flip side, as you aligned -- lay out that framework, where could it be wrong? Where would it actually not bear the fruits that you're anticipating? Certainly using the successes of cabo to lift you to the future. And then I'll just ask my second question as well. Similarly and related as you mentioned, getting that leverage historically, 25% of the funds for cabo were fueled by Exelixis alone, leaving the balance externally. Can you expect that luxury going forward as you invest in zanza in the pipeline?

So look, as I've talked about today, and we all know this is a really, really hard business. I would say, perseverance needs to be juxtaposed with humility in what we do and how we do it. I think we probably made every mistake known to man with cabo as we discovered that molecule, developed that molecule, took some wrong turns, took some right turns. It's a molecule that we were giving it back twice by big pharma. The first few pivotal trials didn't work, right? There was a point in time when we had probably 3x more debt than cash because of that. So we've learned a lot from the good stuff as well as the stuff that didn't work out as we had planned. And I think it's the -- you can grow pretty thick skinned in this business, if you're going to be in it for a long time and be prepared to take the turns you need to take and learn from the good stuff and the bad stuff as you go forward. So I think that's our unique experience. There's many successful ways to move the needle for patients. Our view is informed by cabo, but we're certainly -- and I think that's part of the beauty of having people who have been at Genentech, been at big pharma, other small companies because we milled all that together. And as we talked about even before the day started, having a management team that's been together for not years, but in some cases, decades, is a really strong way to build trust to learn together and to go through the good times and the bad times in a way that provides that momentum to be successful going forward. So that's my view. Others may feel differently, but they can talk to you about that at lunch. In terms of co-funding, look, that's a really important part of the process. Obviously, I think the interest in co-funding is driven by clinical success. And that was certainly the case with cabo. We certainly got a lot of momentum there early on. Once we had the METEOR data, we had CABOSUN data. We had the framework for what a best-in-class molecule like Cabozantinib could be. And I would expect a similar kind of thing with zanza, XB002, other molecules in the pipeline. It's incumbent upon us to show the activity to show the differentiation to show the benefits. The discussions we're having with a variety of biopharmas are based upon that, with certainly with zanza, and I would expect others as time goes on. But look, a really important way to share the cost, share the burden and then share the upside. And the group of people that win the most are the patients when we do that, again, combining and conquering patients benefit the most, right?

Alex on for Peter Lawson at Barclays. Can you -- just a question on your USP1 inhibitor. Can you comment a little bit on the Phase I design, if you're taking all comers initially or refining the population in any way? And if you're doing anything different in terms of strategy or scope of development versus what Insilico had planned to do.

Amy, go ahead.

Yes, sure. So the Phase I is under consideration and redesign as we think about the patient population that could potentially benefit. It's right now somewhat restricted, but we are increasing that to an all-comers population, probably not. But certainly to a patient population that is very likely to derive benefit, which could extend into HRD and beyond BRCA mutation.

It is Tuuli for Stephen Willey, Stifel. I have 3 questions on my end, but the first one is for Dr. Aftab regarding XL495. So for this molecule, if I remember correctly, you showed significantly better kinome selectivity to compare to the benchmark molecule. But it looks like it still hit 19 out of like 34 kinases. But do you think that selectivity is sufficient to ensure safe dosing in the clinic?

Yes. I just want to put that into perspective a little bit in that the screening that we do for kinome selectivity, we use a concentration in that screen that is 100x the IC50 for the intended target in a cell-based assay. So you have to understand that that's the level of differential we're looking at. Now that just really helps us formulate a hypothesis initially to understand if we really are giving the drug a better chance at showing improved tolerability because that's really what it's about when you're trying to drive selectivity of an important synthetic lethal target. You really want to hit it as selectively as possible because the potential for toxicity due to off-target activity creates a liability for the molecule, especially in the normal tissue. So the next step, though, is to then do comparative toxicology testing, and that's what we did in this program where we compared in a rat non-GLP toxicology study a very, I would say, aggressive preclinical non-GLP exploratory toxicology analysis to really understand the therapeutic index of our compound directly side by side with the repair compound. And I had the data on the slide, we truly did achieve what we believe to be a significant improvement in therapeutic index of about threefold at a minimum just in that preclinical study. So we do think that, that sort of 35% reduction in the in vitro kinome screening that we saw between the 2 compounds really does translate to a potential for improved safety in patients. And that's just one of several parameters we were focusing on to drive potential for best-in-class with this molecule.

I have just 2 more for Dr. Pal's regarding STELLAR-001 study. So in this study, are all these cabo-exposed patients actually cabo refractory? I think you showed 17 of those, right, in the bar graph. So are they all cabo refractory? And second is actually, so 75% of this prior cabo exposed patients responded to a zanza. So do you think it's -- zanza is just resensitizing these patients? Or can you please just speculate on this?

Yes, sure. And this -- I'll address the second question first. I think it's a good one in the sense that, as Dana had alluded to, I think it's the unique properties of Zanzalintinib, the tolerability that likely allows patients to sort of sustain dosing for a longer duration of time and therefore, derive both efficacy and prolonged dosing of the drug over an extended duration relative to their prior dosing with Cabozantinib. So I think that those properties really lend itself to activity of Zanzalintinib post cabo upfront. Regarding the response data for prior Cabozantinib in this context, we've mentioned that amongst those 4 patients that had prior Cabozantinib that responded, all 3 had prior progressive disease, 1 discontinued Cabozantinib other reasons. We're going back right now and trying to dig up the response data for the remainder of cabo-pretreated patients to clarify that as well.

All right. Just -- Akash at Jefferies. So XB002, the tissue factor, I guess, MMAE, ADC, I think you had data in October of last year, you didn't have any confirmed response rates. Can you kind of confirm what's the update? I think you had initially 15 patients worth of data. Have you guys actually seen confirmed responses with that program to date? And can you talk about what's going on with Project Optimus? And what does that do for time lines for getting that into pivotal studies? I'll pause there and have couple more.

Yes. So just to correct you, the payload with XB002 is the modified auristatin. MMAE is the payload for TIVDAK. So ours is a different pay load. With regard to activity, we're looking actually forward to printing that data in 2024. We don't need and shouldn't be trickling out data. We have nice deep relationships with our investigators, and we would much prefer that they present the data at the right time. So I'm not going to be providing you any updates on that today.

Okay. And on Project -- okay...

And Project Optimus, yes. So we are investigating Project Optimus in all of our studies, right, with zanza, with XB002, we have it built into XL309, we're trying to do that early so that we don't have a post-marketing commitment.

Okay. Understood. So your team talked about kind of discipline when it came to R&D spend. What -- have you guys considered maybe giving midterm guidance on your R&D spend? I think a lot of investors and analysts didn't have the step-up that you had in 2023. And if not, are there any kind of key levers that will determine whether there's going to be an increase in R&D spend or a decrease? What are the most important contributors to R&D spend over the next 2 to 3 years?

Yes, Akash. As I mentioned in the intro, we'll talk about the business side of Exelixis at JPMorgan in January. So stay tuned for that, okay?

Okay. Perfect. And then finally, on the PKMYT1, the repair data was pretty unimpressive. I mean, you've seen data in ovarian, where you have 40% plus response rates, both with DDR agents or with ADCs, some without even a biomarker. What gives you confidence that incrementally better exposure on that target would actually lead to differentiated efficacy here?

Yes. So thanks for the question, Akash. So again, it really comes down to on-target activity. Our predictions are that we're going to achieve 100% target coverage with our molecule at our projected human dose. Similarly, our same models predict for the repair compound that it's not achieving 100% target coverage, which is why we believe they really needed to do their combination with their ATR inhibitor because in that way, they're really covering the target from a couple of different angles. The ATR inhibitor is also impacting phosphorylation of CDK1, but from the other angle, through activation of CDC25 phosphatase. So when they do that, they're getting much better coverage on the intended downstream target of CDK1, and then that translates to much better efficacy than the single agent of [indiscernible]. So we think that if you translate that directly to target coverage, we should be able to see the same level of activity with a single-agent that they're seeing with the combination. But single-agent activity in that program is interesting. What's really exciting is going beyond that and combining with drugs. Both our own drugs and our own pipeline, XL309, as you could see, we saw the highest degree of synergy with XL309 plus XL495 compared to any of the other combinations we've done so far. So -- and that was just one of many that we're seeing great potential for combination. So that's really where my excitement lies is in the combinations. I do expect to see single-agent activity but it's really in the combinations that I think we're going to be driving differentiation in the clinic.

This is Chi on for Jason Gerberry, BofA. I think I have 3 questions. Maybe the first one, starting with XB002, can you help us understand how you see XB002 and XB371 fit into your broader development picture? Are you thinking more about indication segment between the two? Are you thinking more 371, say, product life cycle management in XB002. And then a follow-up on your other programs after that.

Yes. No, I think right now, we've got XB002 that we're advancing. And the indications that we're thinking about are those that obviously express tissue factor that are sensitive to auristatin payload where we can differentiate from a lead competition for 371. How that enters the clinic, I think, will absolutely depend on the indication, the duration of response required in order to get activity -- the activity profile of the ADC. Certainly colorectal is high on the list, which is more sensitive to a tecan like payload than it is to an auristatin small cell lung cancers. Dana mentioned, we're looking at tissue factor expression there to see if that's an indication that could be completely captured, which won't be captured with XB002. So I think we'll look to see where they can complement separately. And then where they have a similar overlap. It's really about the tolerability profile, the patient population that we're going after and what is the toxicities that are tolerated in that patient population based on prior therapies.

Okay. And my second question is on USP1, maybe a 2-part question. So KSQ has been in the clinic, correct me if I'm wrong, for a couple of years. And Roche had a deal with them, presumably, they may have seen some data on it. So I'm curious, to what understanding do you have on your competitor data? And what might that may or may not have -- drive your enthusiasm for the program? And again, secondarily, what type of earnings have you learned from your competitive programs as you can leverage to expedite your Phase I development plan?

So as Mike has said, I'm trying to say, we are a big company, but we have to be small. We have to be nimble. We have to be quick. Not anticipating that it will take us 2 years to generate data. And I think we have to look at this molecule where it has activity both in patients who are PARP refractory, but also in combination with the right PARP inhibitor and potentially in combination with chemotherapy to advance it. So we're always looking at the competition. The fact that it is slow moving, I can't really speak to that, but we will not be that slow.

Okay. Great. I guess, my last question, hopefully, a quick one, is on your cabo prostate data that you aim to present in early 2024, can we expect to see OS data in that presentation? And what is your projected next OS analysis to be?

Yes, it's hard to give guidance on what we'll present before the fact. So stay tuned on that, right? So we hope to have data early in the year. So we're excited about being able to get that out there and get people look at the data. So Amy, Monty, you want to add more color commentary?

No, I think we had our top line results. We had a statistically significant PFS benefit and a trend in OS. So we'll continue to follow.

And Amy, I appreciate your comments and your talk around the uniqueness of the setting. This focus on visceral metastasis in very advanced population is something we really haven't seen specifically targeted in advanced prostate cancer before. I think to some of the slides that PJ was showing earlier with the landscape of available therapies in prostate cancer, and it might seem like a crowd in place with docetaxel and cabazitaxel and multiple NHTs. But I mean, the truth of the matter is if you look at real-world data, very few patients actually get chemotherapy. There's really a lot of redundancy amongst the NHT options that we have available. Agents that are rolling out in the radiopharmaceutical space are getting very poorly utilized because the -- I mean the rollout has just been very, very challenging. Distribution is a big issue. So there's a real place for a regimen like this that people are familiar with in other context to really serve a unique role in prostate cancer.

Jay Olson from Oppenheimer. Congrats on all the progress, and thank you for providing this impressive update today. A couple of questions. What is the time line for us to see Phase III clinical trials with zanza in clear cell RCC? And then what's the potential impact of the IRA on your clinical development plans, especially for a small molecule like zanza where you can leverage cabo earnings to initiate a broad range of pivotal trials in parallel? And then I had a follow-up question on XB002, if I could.

Okay. Just take the first part, and I'll take the second part.

Oh, okay. Yes. So we are actively enrolling these studies. We're in these steep part, if you think about enrollment, being like a hockey stick where it takes a little bit of time to get your momentum and then you are rocking and rolling. We are rocking and rolling. But with regard to the readout and the time lines for those, I'm not really at liberty to discuss any of that. It really depends on the events as they come in, they're time-based. And fortunately, in some patients, we're literally waiting for death to occur. So I just can't make a prediction on when we'll have those data.

Jay, in terms of the IRA, look, it's one important component that we consider as part of our overall strategy for how we're operating on a discovery level, on a development level, the continuum between those two. It's not the overriding driver, right? We want to have data and making priorities and decisions based upon that data. But certainly, in the small molecule world, it's currently envisioned, front-loading pivotal trials makes sense because the clock is ticking after your first approval. So a lot could happen over the next 5 years, 10 years with the IRA and litigation and those kinds of things. So we have to keep our eye on that and see how that evolves. But we're here to do the business of high-end, high-quality drug discovery. Data drives the process. We'll prioritize and focus based upon that data for good and bad, and move forward aggressively. That's the goal.

And then on XB002, can you just talk about the potential to demonstrate efficacy in patients who have lower or heterogeneous tissue factor expression levels? And also, have you seen signals of immunogenic cell death that would support synergistic combinations with immune checkpoint inhibitors?

So we have a cohort in JEWEL-101 that is going -- that is tumor agnostic, but tissue factor expression required, where we hope to have an understanding of expression levels, hopefully not only as a measurement of IHC, but other measures to help us understand what is important to figure out how much target is required on the tumor in order for it to be active. And then as regard the immuno.

Yes. So for driving -- sorry, for driving immunogenic cell death, we have not looked at that with XB002. We have looked at it with XP371 and we do see potent induction of immunogenic cell death with auristatin payload, it's a possibility, but we just don't have the data on that. And then from a tissue factor expression standpoint, we're certainly collecting the data to understand that in the clinical trial.

Right. And I mentioned this during the talk, but we are combining with nivolumab, PD-1 to understand whether or not there's potential clinical additivity with those 2 combined, which could get into the immunogenic cell data.

Paul Jeng, Guggenheim for Michael Schmidt. I have two questions. Maybe the first for Amy. Can you sort of talk about your clinical development strategy for zanza in combination with the HIF-2 alpha from Arcus? Obviously, you have a strong foothold in RCC with cabo. Where do you -- and Merck has a very extensive Phase III in that setting, where do you see the main opportunity for differentiation with the combo for zanza?

Okay. Early days, we just announced the collaboration. We're just starting the Phase I dose escalation. We think that zanza is a best-in-class. We'll see what happens with the Arcus HIF inhibitor and the combination. We don't yet the outcomes of the belzutifan [ Len ] studies, but we do believe that there is space and an opportunity for us to advance and we'll figure that out as we understand whether or not and how much differentiation there is with the zanza and the Arcus' HIF-2 inhibitor.

Great. And then my second one is perhaps for Dana. Just on the PKMYT1 program, for CCNE1 amplified tumors. Can you talk about sort of how you view that approach compared to perhaps CDK2 or [indiscernible]? How much do you see from the CDK2 programs to your program?

Well, I think those are different, right? We're looking at a different way of basically attacking the problem. And as I mentioned, synthetic lethality can be viewed through multiple lenses, right? So looking at CCNE amplification as well as some of the other genetic biomarkers that I talked about FBXW7 and protein phosphatase IIa, it gives us multiple potential genetic biomarkers to look at, and there is some validation for those with the work that [ repair ] did. They've shown some very nice biomarker data on those biomarkers. But as I said earlier, that's sort of the first step for us, really the excitement around that compound is around its potential for combining with other mechanisms that drive DNA damage and genomic instability. Those genetic biomarkers are just the first step. The next step is to look at combining with DNA damaging agents in those biomarker positive populations and also in those biomarker negative populations. We think there's a broad potential for treatment options with this compound. And that's why I'm really very excited about that compound.

Hello. This is Alex Ramsey on for Andy Hsieh at William Blair. A couple of questions on the earlier stage pipeline, maybe for Dana. So first, for XB010, the 5T4 target seems to have a bit of a rocky history. So we were just curious, based on your competitive intelligence, what sort of went wrong with these prior 5T4 modalities. And then I have another question on another program.

Yes. So that's a great question. We did spend quite a bit of time digging into that. And the primary front runner, as I mentioned, was the molecule from Pfizer. They took an interesting path in developing that compound. As I mentioned, they used MMAF, which maybe at the time they were doing the development, this wasn't widely known, but now it's widely known to be generally an inferior or statin-based payload compared to MMAE. They also used first-generation technology, which essentially relates to just sloppier ADC chemistry with sort of a dirtier profile in vivo. And then in their clinical trial, the highest enrolling disease group was colorectal cancer, a patient population that we would never enroll in our trial with -- or statin-based payload. So they claim that the trial was stopped due to lack of a path forward. They didn't highlight efficacy or anything -- or sorry, safety as a main concern. They are really just prioritizing? And then the only other really advanced clinical compound, I can't remember the name of the company that was developing it, but it was the [indiscernible] based payload, which is really an unproven technology and potentially very toxic. So that company deprioritized their compound because they're a small company, and they have to focus their resources on other things. So as I said, we're excited about that program because it really does fit the bill as an ideal ADC target. That's why a number of companies have looked at it. Only 2 really got into the clinic so far and made any inroads, and we feel that their attempts really informed us on a better approach. So we think we are taking a better approach, and we're taking another stab at it. But as we said multiple times, the proof really comes in the clinical data. So we're going to move that program quickly. And then we'll make our decision quickly. If we're seeing a good benefit risk profile with that, we'll continue to push it, or we'll have to deprioritize it because we have so many other compounds coming up in the pipeline.

Perfect. And then for EXEL-7871, the PLK4 program. We were just curious among the various PLK homologs. What is it about PLK4 specifically to cause it to bubble up as a target for this program?

Yes. So it first surfaced in 2016 as a potential synthetic lethality target in TRIM37 amplified tumors. And then there were a couple of papers published really just 2 or 3 years ago that really laid out the biology behind those. We were studying PLK4 inhibitors back in the day that when I was in the lab doing drug discovery work. And we really didn't understand the biology enough in those days to really make that work. Now, so much more is known about the biology of that compound. There's still a lot of that pathway, that kinase. The fact that the kinase itself seems to form a scaffolding function with -- as part of the other proteins that form part of the pericentriolar complex is really interesting and that the ubiquitination really affects its ability to sort of self-assemble as part of the scaffolding complex as opposed to regulating its levels, whereas ubiquitination of another scaffolding protein does regulate its levels. So there's still more coming out on the biology. But what's really clear is the synthetic lethality. In these TRIM37-amplified patients, that's really a very unique situation that leads to lethality with the PLK4 inhibitor. So clearly, there's more biology to elucidate, and that's one of the things that we -- I think we do well with Cabozantinib. We didn't fully understand how broadly impactful that combination of targets could be suspected based on our early data, but then it was really after we got that development compound out there and started really pushing the studies in the lab, not just in our own labs, but with a broad collaborative network of experts in academia and elsewhere to really probe the biology. So that's going to be no different with this program and every other program that we bring forward to the pipeline.

This is Joyce on for Yaron Werber from TD Cowen. First on XB628, the NKG2A target has maybe been somewhat of a mix target for inflammation in AML with NK cells. As you think about future development plans for this drug, which tumors maybe come to mind and might be most relevant in terms of highest overexpression?

Yes. Obviously, for us, going into tumors where we've seen responses in activity with PD-L1, PD-1 targeted agents is an obvious choice. When those patients progress, there's often signs that NKG2A is a potential avenue to bring more benefit to those patients, driving in the NK cells into those tumors, while also continuing to inhibit the PD-1 pathway is really a large part of our rationale for selecting patients. So that's really the first step for us going into tumor types where we see activity with the current sort of stable immune checkpoint inhibitors.

And then maybe just a couple for Dr. Pal. Could you just elaborate on where you the unmet need for zanza in clear cell RCC versus the current cabo experience? And do you think zanza would replace cabo because it's safety profile? Or would it primarily get used in cabo experienced patients? And then in the cabo naive setting, do you think zanza offers better efficacy than what cabo offers in these patients?

Yes, sure. I think it lends itself to perhaps my previous response there. There's something unique about Zanzalintinib. We highlighted that with that juxtaposition of the toxicities of zanza versus cabo. And I do think the fact that patients are experiencing less in the way of toxicity means that they're going to have further dose intensity, and that's going to lend itself to improvement in progression-free survival and certainly allow them opportunity to achieve deeper responses. In terms of the development plans for Zanzalintinib, Amy has already highlighted the STELLAR-304 trial, which is ongoing, that's evaluating sunitinib versus Zanzalintinib plus nivolumab in that non-clear cell RCC space. I'm so grateful to Exelixis for taking that on. I mean, this is the first randomized Phase III in non-clear cell RCC that we've had, and we're greatly in need of a way to better address those particular patients in the clinic. But I don't want to speculate on further trial designs for Zanzalintinib in the future. But having said that, I certainly see a need for a TKI based for combination therapy that's better tolerated than what we have now. As I look towards triplet regimens that are currently in development, combinations of len, pembro, belzutifan, LEN or other [ axi-based ] combinations. I see that there's going to be which toxicities are going to get a way -- in the way of effective treatment of regimens within these doublets and triplets. And I think Zanzalintinib really stands uniquely poised to address that.

All right. We have hit the time. So I want to thank everybody again for joining us today. Really great questions. Again, thanks to the presenters and the staff for helping us pull this off. We're going to break for lunch. Again, thanks again, and we will continue these discussions over a sandwich. Okay. Thank you.