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

Good day, everybody. My name is Jason Gerberry. I cover specialty pharma and smid-cap biotech at Bank of America. I'm pleased to be introducing our next company presenter, Exelixis, and CEO, Mike Morrissey. Mike, we've done Vegas and Napa now digitally. We're very -- very sterile, very benign, anti-climactic. But glad to have you nonetheless.

Well, I'm in my same living room with my same clothes on and my same tennis shoes. So yes, Groundhog Day all over again. But it's great to be here, and thanks again for the invite. And again, before we begin, I'll just remind everybody that I'll be making forward-looking statements today. So please see our SEC filings for a description of the risks that we face in our business.

Yes. Okay. Great. So look, for everybody listening, this is the first one of these I've done in Napa digital, but we're trying to make this very interactive. I don't see any questions in the queue, but I certainly -- submit questions through Veracast. I'll be -- I'll ask them.

I'm going to focus first on liver cancer because it's one of the 2020 events that we are focusing on in our research. And so very curious to get your perspective on the frontline HCC opportunity, Mike. And so on COSMIC-312, can you just discuss how enrollment is progressing? Do you think the company is still on track to have an interim update in the second half of 2020?

Yes. So thanks for the question. Look, it's great to focus on 312 and HCC as we and others, we've been really focused myopically on 9ER and prostate and other stuff. So this is a really important indication. This is an opportunity for, I think, everybody in biopharma to focus on a very important disease that impacts literally millions of people globally every year. So we're excited to be in this space. We got -- we had great data at ASCO GU with cabo and IO earlier in the year, and we're certainly working very hard on 312. So in terms of what's happening there, I can tell you that we have hit the 740-patient randomized milestone recently. So that's great news. And certainly, I think great kudos go to the team for working through really all the complexities, not only of running a large global pivotal trial but doing it in the time of COVID, where you've got regional differences of sites turning on and off and all the challenges that go with that, while the global health care system tries to manage that pandemic. We will continue enrolling for the next month or so. There's a little -- a little imbalance in the single-agent cabo arm due to the amendment that we put in relatively early, but had an impact on enrollment. So we'll finish that really over the next month or so and then move forward from there. So in terms of what's happening with the interim, again, we gave initial guidance in January of 2020 that we -- as things went on, kept on track. We'd have top line data potentially at the end of the year. I'll remind everybody, this is all event-based, both for PFS and OS. We're midyear right now. So I wouldn't to say one thing or the other relative to what's going to happen at the end of the year, but we're certainly very focused on making sure we can get this trial done as quickly as possible based upon the events that have to accrue for both progression and then survival. So stay tuned there.

Okay. Now I believe that this trial, the enrollment criteria, are pretty much virtually the same as the IMbrave150 trial. I'm not sure if there are any differences worth flagging, but I ask just because anecdotally, we've heard that the 150 trial excluded patients with poor liver function or risk of bleeding, which may represent potentially as much as 2/3 of the eligible patient population. And so I'm just wondering the representativeness of that data given that Avastin could have a bleeding risk in these patients. I think it was 30% all grades. And I would think that, that could represent an opportunity for a shorter-acting anti-angiogenic being on board with the IO agent. So just curious to get your perspectives on trial enrollment, how it's similar or different from IMbrave150 and the issue that we flagged there.

Yes. So at a very kind of top line summary, I think 312 and IMbrave are relatively similarly designed trials. We have a cabo-only arm, they didn't do that, to have a single-agent arm for either bev or atezo. We're enrolling more patients than they are, but certainly using a lot of the same investigators, all the same regions of the world. I would say the one -- I wouldn't call it a big difference, but the one nuance when you bring up bleeding as an important consideration to think about here, is that these patients with their inherent liver disease are at risk for bleeding. And that's normally around the idea that as the disease progresses, liver cancer patients can develop either esophageal or gastric varices, which are prone to bleed. And in some cases, when these ruptures occur, it can be life-threatening. In the IMbrave trial, they did endoscopy on every patient to understand that factor in real time. We're not doing that. So we're looking at very detailed medical histories, like we did with CELESTIAL, and weeding out patients that way based upon their medical history and their history for bleeding. So I'll remind you that in the CELESTIAL trial, 800 patients, second line, so later-line therapy. We had 1 grade 5 bleed. In the IMbrave trial, with the extra screening, they had 5. So -- but again, it's part of the disease. Obviously, you want to be able to maximize efficacy and safety. So I have to look very carefully at this component of the patient pathology and comorbidities as you go forward here. So very important.

Yes. So I think -- look, if we were to think about a scenario where 312 looks similar to IMbrave -- I asked the question because I'm curious how you'd envision the market evolving. Do you think that this is a market where there's a traditional first mover advantage? Do you see the marketplace becoming segmented? One of the arguments I've heard is, well, maybe some doctors may prefer to IV infuse drugs. Another counterargument I've heard is, well, you raised the issue about bleeding and maybe some doctors may prefer to just have a shorter-acting anti-angiogenic. So if the bleeding occurs, I can get the patient off of their anti-angiogenic versus Avastin. It stays in them for 3 weeks.

So the question is how the market evolved?

How you would envision, yes, the market evolving in a scenario where you had comparable data?

Yes. Look, and it's all about the data, right? And I think that's the ultimate arbiter here on what will happen, what can happen, what we've seen happen with us and others. You roll in there with better data, differentiating data that you can reinforce with prescribers in terms of the benefits you bring that others may not have. You can use that first-mover advantages are often important when there is equivocal data, but data is rarely all equivocal across the board. So again, I don't -- I don't have a crystal ball. I can't tell you how 312 is going to stack up IMbrave and the other trials that are ongoing here. This is a very competitive space. That's a bit of a statement of the obvious. It's oncology, everything is competitive, right? There's no single player in any indication that has that indication for themselves. It's just there's just too much going on here. So we are comfortable in either scenario, whether we have -- whether we're second with different data, with better data, some way we'll be able to make that happen. Or in a situation where even if we have equivalent data, there's enough differences in terms of how cabo can be used here to be able to help patients, again, pending the appropriate data, where we think we can make a lot of progress. So we've got good name recognition here. Everybody understands the activity of CELESTIAL -- of cabo and CELESTIAL, which, having that initial survival signal gives us a lot of traction to then move into the frontline setting once we have the appropriate level of data. So a long-winded way of saying we need data, and we need to be able to see how that looks. But with our commercial infrastructure and the momentum that we've got right now, we feel pretty good about that in any upside scenario, for sure.

Got it. And so obviously, part of the HCC story, I think, with investors, it's a show-me story. I think this idea of a market paradigm shift, let's say, the U.S., for example, where about 30% of eligible patients by our estimates are getting systemic therapy. With Roche rolling out IMbrave150, how quickly can you see that evolution occurring to systemic treatment rates increasing, potentially even doubling? Is that something you think could happen in 4 to 5 years? Is it a decade? In terms of your experience as you look at market analogs, you look at the quality of data that's being generated in the space, just curious to get your perspectives on that.

Well, I think if you look at the introduction of novel IO therapies into frontline indications, think about lung, think about melanoma, think about renal, those have all been very rapid. And the uptake has been swift. And the impact on, I think, both patients and on the standard of care opportunities has been as robust as you would possibly hope for. It's a little bit more complicated here because of what has historically been a lack of really kind of game-changing compounds with data that either medical oncologists or hepatologists or interventional radiologists can kind of focus on together in that segment, the patient triaging approach, but really unify that in a way that makes sense. I would propose, I guess I would argue, that I think IMbrave does that for the first time, where you have a very effective, very tolerable regimen that clearly surpassed the long -- maybe long-term standard of care that left a lot of people wanting for more, right? So I think this is the catalyst, if you will, that we've been talking about for the last 2 or 3 years now that can really unify the market, unify the standard of care across 3 different medical disciplines in a way that brings the most benefit to patients and is going to -- I think it will be the catalyst for seeing more patients go down the medical oncology work stream and bring them more benefit because they're getting these drugs, and they're getting introduced to the systemic -- the systemic oncology setting faster, earlier in their disease, when they're healthier, they're more robust, they haven't lost more liver function, they're amenable to these kinds of therapies. So yes, I think this is the catalyst and the next steps follow on from there, whether it be what we're doing or what others are doing in the frontline setting, how you then even go a step further and maybe think about triplets, how you can more effectively triage patients when they progress on frontline IO. It's that evolution. But there's -- again, globally, there's 800,000 new patients diagnosed with liver cancer every year. And that's a large number of patients that need better therapies today. So I'm super excited about this, and I think it will be the first step in what I hope to be a bit of a renaissance in liver cancer treatment paradigms going forward.

And so not to put words in your mouth, I just want to summarize. So would you think the frontline lung and how quickly the IO was adopted in frontline lung is a good analog then for perhaps the market transformation and the move away from TACE?

Frontline lung, I would say, is one example that I use that -- yes, I wouldn't want you to put words in my mouth either. I think that was a market that already existed in terms of medical oncology. So it's obviously different. I think the challenge here now is going to be the tension between interventional radiology and medical oncology and asking the question operationally, how that's going to -- what's best for the patients from a standard-of-care perspective, right? And I think that tension will evaporate quickly when the data from IMbrave and the next wave of trials that will read out over the next year or so starts to take hold. So is it going to be instantaneous like it was with IO in lung? Probably not. But will it be faster than 4 or 5 years? I sure hope so. I mean if it takes the medical community that long to make that transition, I think we're probably not serving our patients as well as we could. So...

Okay. One issue that I wanted to touch upon -- and if the audience doesn't ask me questions, I'm going to keep -- I have a few more on this liver cancer topic, which is of interest us. So I guess in second line, ipi/nivo showed some very impressive efficacy survival data. And as it pertains to IMbrave patients who were on the TKI monotherapy, didn't have the opportunity to cross over to ipi/nivo, whereas with both your trial and Merck's IO/TKI trial, that is a possibility to cross over for lack of a better term, but they can get ipi/nivo to second line. I think they showed 22 months OS. So I'm curious if that creates a risk, a challenge in terms of the -- your COSMIC-312 showing as compelling of an OS benefit as, say, we saw in IMbrave. Curious to your thoughts there.

Yes. Well, I'm smiling here because 12 months ago or 18 months ago, that same question was asked about 9ER, right? How are you going to possibly show a survival advantage in renal when you've got ipi/nivo and axi/pem and everything else out there that's going to salvage the sunitinib arm, right? So this is the -- it's always an issue. Any global randomized OS survival -- overall survival study has that -- have that inherent risk that new therapies, new regimens can potentially salvage the control arm and kind of have an impact on the stats. That rarely happens. And certainly, we had this issue over and over and over again relative to renal that, obviously, both -- for both METEOR and for 9ER, where that didn't come to pass. So I mean, obviously, you want the best thing for your patients. Obviously, the more treatments they get, the better. And that normally has a -- I would say it's differentially impacted by the reasons where you enrolled. And obviously, we're all enrolling all of our pivotal trials globally because that's kind of where the patients are, and to be able to get to the kind of numbers we're talking about in a relatively kind of reasonable time frame, you've got to enroll globally. And a lot of those therapies aren't available. So unfortunately, right, in terms of -- either in terms of their regulatory issues or their -- the reimbursement issues. So to make another long story short, this is a topic that comes up for every survival trial, and it's something that we just deal with. And we understand and we take steps to minimize that complicating what we're doing in terms of the readouts that we want to see.

Is there any aspect of the geographical distribution of the study that adds to that comfort, knowing that ipi/nivo may not be able to salvage the sunitinib arm, for example, in terms of enrollment within Asian countries that perhaps where ipi/nivo is not made available yet?

I would say in general, the fact that global pivotal trials within the -- throughout the industry are run globally and are run in Asia, they're run in South America, they're run in Europe, I mean, everywhere, right, every -- each of those regions or subregions has different parameters around how they view subsequent therapies and what's available and what's not. So you're looking at a very broad population, and you're looking at a very broad set of issues in terms of what's just general availability. So bottom line is that I don't think that's something certainly that we worry about. I'm just -- I don't obsess about it because I understand, based upon the -- just the empirical data that we've seen ourselves and others have seen, this has not been a complicating issue in the past.

Okay. Okay. Now just a couple more on this. So Merck's Keytruda-Lenvima combination appears positioned to be a strong competitor in the space and arguably has some advantages in the eyes of KOLs that we've talked to, at least in liver cancer on the basis of their frontline REFLECT study data. In your mind, what are some of the key battlegrounds within the IO/TKI segment, if you are to ultimately be competitive with Merck, like you are, in a lot of other end markets?

Yes. Again, as I said before, this is an area with a lot of competition. The Merck-Eisai trials, amongst others, are in the mix right now. There's other IO trials going in the frontline setting, too. The -- yes, the situation is just -- it's going to come down to the data, right, relative to what we have versus what others have in terms of efficacy, safety, tolerability, et cetera, right? So it's something that everybody -- anybody in this business, whether you're on the company side or on the investor side, has to have some degree of comfort or understanding that nobody is -- no one is alone here, right? And we're certainly not, they're not -- we have very -- I think we have very competitive data. You mentioned the ipi/nivo survival data being 20-plus months. Well, cabo/nivo had the same kind of numbers for OS from the overall cohort that we had at ASCO GI. So we're confident in our data. We've got a 20% to 30% response rate. We've got good PFS data with cabo/IO in first- and second-line HCC. So yes, I mean, we can all quibble about small Phase Ib data sets and try and dissect what they mean and don't mean and who's got the edge and what is it all, who's got what, where, when. But end of the day, you've got to run the big trial. You've got to randomize. Certainly, running a randomized trial is the great equalizer here, and we're doing that. And as I mentioned before, we hit our first big enrollment milestone with hitting the 740 recently. And so I feel like we've got good momentum here. And we'll get the trial done, and the data will speak for itself. So we'll see.

Okay. Well, one thing that I at least noticed from COSMIC-312 is that you obviously have the cabo therapy mono arm, and Lenvima did not show superiority over sorafenib single agent. And so one thing that we've heard from KOLs is that this could be a differentiator for you guys if your monotherapy arm looks more -- shows superiority over sorafenib. So I don't know if you can speak to the strategy. You guys, I think, are the only ones that have a single-agent arm. And do you think that could be a competitive differentiator for you?

Sure. Look, I think it's possible. I mean that arm was put in there based upon feedback from the agency around having single-agent cabo in the frontline setting, helping define the contribution of components in this indication in this line of therapy. So plain and simple, it's -- 185 or so patients will be enrolled here. So as you would imagine, the powering, showing any kind of superiority head to head against sorafenib here is going to be pretty tight, right? So again, it's more there to fulfill a regulatory requirement than to show superiority. So I don't want people to get too focused on something that wasn't designed to do. But again, we talk to the agency a lot, we get feedback from them and we try to incorporate it to both satisfy their requirements and make a better trial, which we did here. So...

Okay. Well, that's my last question on HCC. I do have one from the field. So I'll ask that. It's now a better time than any. So XL092, the question simply is, well, what should we expect from that? I presume that means the question means this year. You presumably will have some preclinical data that you can share and perhaps also disclose moving into the clinic and a program and what that program would look like would be my own inference, but perhaps I'll let you take it from there.

Yes. So XL092 is our next-gen cabo. As we've talked about previously, we have spent a decade or more studying, investigating the activity of cabozantinib in the clinical setting by itself, in combination, different doses. Really broad development program. We have learned a lot from that, and we applied those learnings to making what we believe is a better next-gen molecule. That was based on a number of very, I think, very straightforward but elegant hypothesis about how one could make a better cabo. And we did that preclinically based upon all the data we generated and certainly now moved that into the clinic about a year ago to start asking some of those same questions in the clinical setting. We have committed to presenting data for that molecule to a major international meeting. The abstract has been submitted. It will be in the fall if it's accepted, which I certainly expect it will, and we'll get the data out at that time. I won't comment on all the details, the who, what, when, where and how much. That will come out with the abstract and the presentation. But again, the goal here is to be very descriptive and pretty direct about what we've got and why we did it and what we're seeing. And it may include clinical data as well. Stay tuned on that. But again, we're very excited about this. We're very -- again, the opportunity here to be able to look broadly across different indications, different combinations, doublets and triplets in the vast open white space where cabo has not played yet, we think, is very attractive and can be, I think, a very important part of our business, both for helping patients and building value in the company. So on top of that, we will have probably 3 more presentations in the fall at various meetings about other pre-IND assets for things like CDK7, potentially the ICON-2 assets, other stuff as well that we're excited about from our earlier-stage pipeline to give people the sense of the kind of work we're doing in drug discovery through this broad network of both internal and external researchers. So it's a good time for us to kind of, again, put a stake in the ground and talk about kind of the next wave of molecules that are moving into the clinic.

So it would be some form of Phase I data in the fall when you say may include clinical data. So at a minimum, it's more like PK/PD type of data. Just trying to...

Yes. And again, I don't -- yes. I don't want to go into the details of what we're going to have because we're still looking through that right now. But if we choose to have -- we have the opportunity to put clinical data in with this presentation. If we choose to do that, you'll see that when we both have the abstract and have the presentation. So...

Got it. Okay. [Operator Instructions] Maybe just shifting gears to the lung data that were presented, this was a data set that was presented subsequent to the last time you and I talked at our conference in Las Vegas. You guys obviously have subsequently initiated a Phase III with Roche for atezo/cabo in second-line lung. In terms of the strategy here, trial design, I think there are some other -- 2 other IO/TKIs already in Phase III. Do you -- are there any important differences that you'd flag? Do you feel like, ultimately, pretty consistent in terms of the approach from a trial design perspective?

Yes. I think that there's various slight nuances that are different going against docetaxel post IO, post chemo. It makes the most sense. That's the big bolus of patients when they progress. They usually have seen both of those in some shape, manner or form. So straightforward design, minimal bells and whistles. The ASCO data set from cohort 7, I thought, was very strong. I think people really understood the patient population that we worked with in terms of having patients be a mix of second and third line, mix -- an equal mix of resistant and refractory. It's a pretty late-stage population with a 27% response rate, good overall tolerability and safety. So we're excited about that. And this is one of the -- I think one of the more straightforward questions we have to ask clinically. So great to be working with Roche and their depth in lung. And they'll run the trial, and they've got a global network of investigators that they've worked with in the past -- both the recent past with atezo, but kind of broadly speaking, with their other assets. So it's a great place to play, for sure.

I know companies are, and I'm sure you are as well, highly reluctant to talk about FDA interactions. But have you had discussions with the FDA about studies in this setting? And I ask just because just thinking about the idea that the patient who had some response to IO and this idea that perhaps they were deemed a progressor and maybe weren't and in a single-arm trial, are they having a benefit from their prior IO therapy? Or are they having a benefit from the atezo/cabo regimen? It's obviously a challenge that we hear from physicians in terms of analyzing small data sets, especially single-arm data sets in the lung space. So just kind of curious, any feedback that you've gotten from regulators? I think at the end of the day, when you have a comparator arm in Phase III that solves for everything. But there's also a discussion in the backdrop. Do you upsize this trial to maybe 125 patients and have the optionality for accelerated filing?

Yes. So there's about 5 or 6 questions in there. So I don't want to go out of school and talk about what -- our discussions with the agency. I can tell you that we have, I think, designed the appropriate Phase Ib trial cohorts with 7 as well as a single-agent cabo cohort to be able to understand the activity of the doublet versus the single-agent cabo. Your commentary around pseudo progression, again, I don't want to debate that with you. We've certainly talked to a lot of -- a large number of investigators ourselves about that, and everybody has a different view on that. And it's kind of in the area of anecdotal information more than really hard data, collected data that -- again, in a randomized sense that you could -- at least I can believe in, right? So I'm not too worried about that. That topic has, I think, in some ways, come and gone. And again, it's out there, but it's less of a concern relative to what we're seeing. And we'll answer the question appropriately in terms of having a control arm to be able to compare cabo/atezo to docetaxel, which is certainly one of the molecules that is used routinely as these patients progress from either IO chemo or either as a combination or sequentially. So again, the data will speak for itself. In terms of what we're doing in COSMIC-021, I think I've been pretty consistent that lung is probably a step or half step behind where prostate is, where we have a very good kind of view on line of sight for a potential Subpart H filing with prostate. We need more data with lung to be able to understand the totality of data with cabo/atezo versus single-agent cabo because we've seen some activity with cabo by itself in lung in the past. So I don't want to oversell what we might do with that from a regulatory point of view. I don't want to discount that either. I want to get more data. I want to look at all the data. I want to understand the nuances of that data with the appropriate number of patients. And we can upsize both cohort 7 with an additional 50 patients, and we can also add more patients to cabo if we're seeing some level of activity there. So I want to do this in a rigorous fashion. I'm going to get all the data. And as we analyze that data and we finalize our plans, then we'll be sure to share that with you and the rest of the investment community.

Got it. So it sounds like you still believe that there's the potential pending more data. But obviously, given the sheer size of the opportunity in lung and the competitive dynamics sort of makes sense to operate in 2 tracks, right, in terms of initiating the Phase III so that you're not too far behind. But obviously, if timing wasn't as big of a consideration, maybe you'd accrue more data before jumping into a Phase III?

Well, you need a Phase III either way, right? If you believe, as we do, that a near 30% response rate in a highly pretreated, highly refractory population is a meaningful signal, and all of our advisers in the academic world are excited about that signal, then with all the caveats of single-arm, non-randomized studies that we all know about, we all talk about, so no surprise there. Then a global pivotal trial isn't nice to have, it's a must have, right. Because even if you go down the Subpart H routes, you still need to have a confirmatory trial going to have that level of traction with anybody, right? So it's just -- and in some ways, it's just almost common sense in terms of a logical progression of what we have to do. We have a very clear line of sight on what success could look like in either scenario, and we just need data to fill in the details so that we can choose path A or path B or both. So -- but I don't want to get ahead of myself here because we're just not as far in lung. Prostate, we had a very clear signal. We had a very clear understanding. We've been in that space for a long time. That was relatively straightforward. This is just a little bit more complicated. And we take our time and do it right and not overhype it, not underhype it or not undertalk about it so we discount it, but we just need more data. So plain and simple.

Okay. Yes. And I know that you guys haven't been too detailed in terms of when the next timing update for more data could come out of either prostate or lung, but presumably, you've been clear about your prostate regulatory backstop from a timing perspective. So is it just fair to assume that at the major oncology medical meetings, be it either the next ASCO or AACR or ESMO, these are the venues to be looking at for further updates from both of these programs or indications?

Yes, I wouldn't -- yes, I would not count on that. I think what we've done is we put a stake in the ground with the ASCO GU and the ASCO presentations for prostate and lung. The headline is that the combination is active and has good tolerability with low discounted rates for both. That, in my mind, is the -- we have fulfilled the obligation to investors to give them a sense of what we're talking about here. The next step is get all the data and pull it together, pick your regulatory path and go, right? So my most important priority is to get complete data sets and do filings and get approvals and then market the drug, okay? That is -- the data you'll see along the way will coincide with those milestones as we go forward, right? So...

Yes. Okay. Great. That's really helpful thinking about the pathway here. Maybe just shifting gears to CheckMate 9ER. I'm reluctant to -- well, I have a hard time creatively asking the questions in any different way. I feel like you've been asked the questions a million different ways here. But your PFS data was very intriguing, I think, in terms of the hazard ratio that you showed. And it seems like that could be a potential point of differentiation versus competitors. And as investors make cross-trial comparisons, 2 competitor data, however challenging and tricky that can be, I guess the one thing I'm wondering about as I look at the competitor Merck data, the PFS sort of bounced around between 15 and 17 months. In the more recent presentation, I think it came back to -- I think it was 15 months. So I guess just on that point, right, what should we compare it to, right? I think when you guys presented 9ER, you guys were quick to point out the right comparator for OS on hazard ratio. So just curious how to think about the PFS signal comparing it to [ some of those ] data.

Yes. I mean again, I would look at everything. I mean I can give you bits and pieces here of what I'm intrigued by. But the totality of data is the critical thing, right? Across risk groups, across PD-L1 status, it's the totality of data which drives it. And certainly, as you get -- as you can see longer-term follow-up from either IO/IO or IO/TKIs, it gets a little complicated to me. As you would imagine, my tables that my guys make for me are very extensive and have gazillion columns and the whole bit. You mentioned the 15 months kind of bouncing around a little bit. I would still focus on the delta. Again, that's the important -- how much additional benefit does the combination give over the control? And if you look at the data from either the first cutoff, which had, what, a 12.8-month follow-up or the most recent one, which had a 27-month follow-up with 426, you're talking about a 4-month delta in terms of the median PFS for the ITT. And that's just -- and that's in the ITT population. And when you dive deeper into the different risk groups, the different PD-1s -- PD-L1s, that patient population, that gets obviously much more complicated. So I look at the whole thing. Look, it's a very important data set. I would never advocate that cross-trial comparisons are going to make or break anything because there's just -- that's an inherently risky proposition. But everyone does it. I do it. We do it because you try to understand kind of what the data all means. And you try to normalize based upon the control, which luckily everybody has kept pretty constant with sunitinib here. So -- but yes, we're -- look, we're super excited about the data, where -- the efficacy, the tolerability, the low discount rates. I think there's a real opportunity for us to help patients in a very different way. And BMS had a call, had their -- one of their R&D Day updates. I think it was this week, and they talked about cabo a little bit, best-in-class combination with nivo for 9ER, et cetera. So our #1 priority is to get this file in ASAP and pending approval, be ready to be sprinting on day 1 after approval. So lots of work to do. The team is very motivated. Yes, for sure.

What about CR, though? I mean CR was 6.9. Now it's up to 8% at the most recent ASCO presentation. I would assume that perhaps there is a -- perhaps the 2019 presentation, CR may improve over time. I'm just wondering, do you think that, that -- is there a more appropriate comparator, at least on CR in terms of the data cut?

Yes. Again, like I said before, Jason, I look at all the data across all the cuts all the time. I mean I -- you can't focus on -- at least I can't focus on one individual cell in an Excel table that's got columns and say that's the important one. It's all important data, right? It's the totality of that data really defines the patient experience. CRs are important. CRs are probably more important in the academic KOL community than in the community setting where the vast majority of patients are treated. It's just a different world in terms of how the therapeutic intent differs between those 2 different kinds of prescribers. Somebody who has 100% tumor shrinkage versus 90% tumor shrinkage, is that really different? Can the imaging really define that to a certain degree? Anyway, I don't know, right? So -- but look, you'll see our data. You'll be seeing their data. We're happy with our data. We think it's really potentially best in class. And when it comes out, so you can opine upon that, as I'm sure you will, and let me know if you disagree or not.

Yes. Yes. No, I think -- look, the CR obviously comes up a lot just because of the initial success that ipi/nivo had in the frontline RCC setting and even durability of ipi/nivo relative to IO/TKI. So we as investors who talk to KOLS, we do hear it come up. But again, we probably select for more conversations with KOLs and academics than we do with community oncologists. So there is probably an inherent bias in the conversations that we have as well.

Yes. Absolutely. There are thousands and thousands of community oncologists who we talk to a lot, who have a different view on that, right? And they have a different view on the totality of that data, which is fine. I mean certainly, IO combinations have revolutionized frontline therapy in RCC. So there's plenty of room for everybody to have an impact and to get involved. And we think we have data that will allow us to capture significant market share. So -- but again, it's -- you're talking to a very select small group of KOLs when you make those small calls, right? I think you've acknowledged that. So I would agree. Yes. Yes.

Okay. It seems like you guys are pretty excited about what you're seeing for your tolerability profile as you started with lower starting dose and what you're seeing overall in terms of Grade 3, 4 AEs and dropout rates. Are we reading between the -- I guess, probably reading between the lines, I think you guys are pretty open about -- openly excited about the data that you're generating around cabo from a tolerability perspective.

And you've -- yes. And you've seen some of that data. You haven't seen it for 9ER yet, but you've seen the data from prostate that was, what, 5%, 6%. Lung that was about 3%. Bladder at ASCO was 0% dropout rate -- so for TR AEs. So the message here -- and this goes back a decade or more with how we develop cabo at the higher doses, trying to find a sensitive population that then we could refine the dose. And I think our excitement is a reflection on this 10-plus-year journey to find the optimal dose that maximizes -- in this case, in combination with IO, that maximizes efficacy. And we certainly are seeing that with the hazard ratios for PFS and survival with tolerability that is, I would say, unexpectedly and surprisingly good, right, based upon the history of what we've seen at 140 milligrams first and then 100 milligrams in GBM and then 60 milligrams in prostate with a very frail population. So yes, so I think it's been a long time coming. This is the hardest part of the business, is finding the right dose for chronic exposure, chronic benefit, chronic therapy. And I think what we're seeing with what we published and presented earlier this year, and certainly, what you'll see in the fall, is that starting with cabo at 40 is really the right place really. And really in that sweet spot, we've threaded that needle to be able to maximize clinical benefit with a good tolerability and a low discontinuation rate.

Yes. Yes. Okay. And obviously, M&A -- I'm sorry, you have to run, Mike?

No. No.

Okay. Just on the M&A front, you guys have been vocal about wanting to be a multiproduct company. I think yesterday, you made some comments about wanting to be a multiproduct company by 2025 or 5 years out. Obviously, M&A plays an important role to that. And I'm curious, the sort of team that you have in place to vet deals, is it coming from the bankers? Is it more internally generated ideas? And it seems like with all the cabo indication expansion, your challenges in growth in the near term, I think it's more a question in investors' eyes about durability in the longer term. So just wondering how you're thinking about the trade-offs of building a solid, sustainable, early to mid-stage pipeline versus late-stage assets and bringing those on board. Can you maybe just talk about that balance that you're trying to strike from an M&A perspective?

Absolutely, because that's the main focus of what we're doing right now. I mean we really have 3 different work streams at the company. And it's, again, statement of the obvious, right? Number one is to build and expand the cabo franchise, cabo 092. We think we've got something special there. We've got very, I think, great insight into the biology that is driving that and how do we maximize that now across indications, across lines of therapy, across other combinations, doublets, triplets or other modalities besides IO. So that's a really important place where we're going to invest going forward, right? Our early-stage pipeline. We've done this effectively with cabo, one of the few companies out there that's taken kind of a homegrown idea, made the molecule on the lab, developed it, arguably made a lot of mistakes along the way, but has kind of learned and grown to where we're helping literally tens of thousands of patients every day with molecules -- a molecule that we designed, we made, we developed. And we want to do that again. And I think we have the right people and the right team with the right insights, both biologically and chemically. Chemically broadly, not just small molecules, but also biologics that can do that. Thirdly, are there existing clinical assets that we can either partner with or acquire that can help us diversify our offering as we go forward, can catalyze that, can accelerate that in a way that makes sense? Certainly, from the standpoint of our -- I think our strong discipline financially, our view on -- almost a militant view on how we're going to spend our money, we want to be careful about that. It's not that we're risk-averse. But we understand the risk profile here, especially in these later-stage assets, and want to make sure that any deal we do has the ability to be essentially monetized commercially. And that's been, at least in my view, where the big risk has been in the last 4 or 5 years. It's relatively straightforward because the goal line is pretty clear right now in terms of getting drugs -- never easy, but it's straightforward, getting drugs over the goal line in terms of having good clinical data that you can file on and get approved on, whether it be to full approval or to accelerated approval, right? That's been done in oncology over and over and over again, right? The challenge is, for many of these drugs, can you actually sell them, right? Are there -- is there -- do you have enough differentiated data? Is the population big enough? Do you have enough traction to be able to actually make that investment pay off relative to the future kind of forward-running commercial opportunity. And a lot of companies, large and small, have made big investments and have basically risks in the -- in that commercial setting. So we want to avoid that for all the obvious reasons. We think we've got the right insight in how to ask those questions in terms of looking for clinically differentiating data or the ability to find that. And that's going to be a big driver. So we're going to be careful, we're going to be aggressive and we're going to make hopefully the right moves that will allow us to build all 3 pillars as we go forward.

All right. And the last unfair question.

Great. One more. Perfect.

And it's totally unfair, right? But it's the IP front. I think you only have one filer, which is peculiar. You have a ton of IP. It seems like a situation that's very ripe to sell. So I'm just curious if you can comment. Why not pursue a settlement since there is only one filer? It's a little odd for a drug that's approaching $1 billion U.S. sales to only have one filer.

Yes. So the question is what again?

This does not seem like a situation that's ripe for settlement with -- I think it's MSN, a single filer. Typically, we see 10 filers on day 1 for a small molecule drug that's approaching $1 billion. So it seems like a situation that could be potentially cleaned up pretty easily.

Yes. So you're right, that's an incredibly -- no, I'm just joking. So that's a question I obviously can't answer, right? I don't want to dive into the details of what our legal strategy is here. We have very strong IP. We have a very high degree of confidence in both composition of matter and the polymorph patents. We're very committed to maximizing the value of that as we go forward. So I'll leave it at that. And the team within legal, both internally and externally, really have a good handle on this. So this will play out over time, and you'll see the result as it unfolds.

All right. Well, great. That's extremely helpful. I appreciate you taking all the questions, fair and unfair. And appreciate all the answers, Mike.

All good. Yes. Thanks again. I really enjoy all your work and talking with you on a weekend. So all the best.

All right. Thanks, Mike, and thanks to all the investors who listened to the call today.

All right. Yes. All good. Thanks.

With that, operator, we can wrap it up. Yes. Thank you.