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

Good afternoon, everybody, and thank you once again for joining us for the 41st Annual Health Care Conference at Cowen. I'm Yaron Werber, one of the biotech analysts. And it's a great pleasure to moderate the fireside chat with Mike Morrissey, Chief Executive Officer of Exelixis. Mike, good to see you, as always.

Good morning, good afternoon. It's kind of interesting. The last live fireside I did was 1 year ago with you in Boston at the Cowen conference before COVID blew up the world. So it's ironic that we're back doing this virtual, but I'm pretty sure next year we'll be live. So good to see you, Yaron.

Yes. I wouldn't be surprised if Biogen was probably right next to us and we were able to dodge the whole situation.

Well, that was the worry, right? Common, speak up?

So really appreciate that. I mean, obviously, a year has passed, a lot of things have happened. So lots to catch up on, a lot of pipeline progress. Obviously, a lot going on in terms of new labels, new approval.

So let's start there and actually talk about you recently gave guidance to exit 2022 with about a $1.5 billion run rate in RCC. Can you walk us through what led to that thinking and giving that guidance? And how is ASCO GU change your thinking, if at all?

Yes, for sure. Great place to start. Before I begin, I'll remind you and the listeners 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. So we chose the 2022 exit run rate on purpose, obviously. It gives us time for the compounding of patients who enter the first-line setting with cabo/nivo to kind of lap the calendar, if you will, again, with a 20-plus month duration of response and along PFS 16, 17 months, we thought that was important. So we gave guidance this year for net product revenue to be about $1 billion. But we think that the 2022 run rate is a good benchmark to both work towards and, honestly, judge us by in terms of our performance based upon having that compounding factor in play. So we did a little bit south of $750 million this year. It was a vast majority of second-line RCC. So we think we'll keep that intact as we go forward. And we see the doubling, if you will, of that -- the $1.5 billion by the end of '22 in that run rate to really be driven by the frontline uptake that we're already seeing so far. So really excited about the approval, the label, the data for those of you that listened to our investor briefing on Saturday at the end of ASCO GU. I think you heard all the KOLs talk about the quality of that data, the totality of that data, doubling of PFS, doubling of response rates, certainly a very strong survival signal. The quality of life data really being supported by the great tolerability that we saw. So we feel really good about that, right? It's a very achievable goal. In some ways, I think it's a little bit understated. I think we can go beyond that. It's obviously a very competitive space, but renal always has been, right? When we came on in the market in 2016, we weren't the only TKI there, only molecule there. There were 11, 12 other compounds. We didn't get a lot of credit for that, a lot of promise for that, but we stuck to our guns and delivered. And certainly having the data that we had in midyear made that happen. So we're really excited about that. We're not worried about data that came out at ASCO GU. We understand that data pretty well. We were planning on clear being positive. So -- but the team is out there and getting -- making a lot of progress talking to HCP. So we're excited about being where we are right now.

And so the -- I mean, you're obviously banking on, I imagine, taking share and also an expansion of the checkpoint, TKI first line, right, away from cabo -- I'm sorry, from -- away from IO-IO in first line. And then that's going to build into COSMIC-313. But what gives you comfort that the checkpoint TKI share will go up and presumably IO-IO will go down in first line?

Yes. So again, our view is we'll take market share from all different segments of that first-line RCC setting. Certainly, if you look at trends, the key via brand impact Rx data, which is public, you've been seeing the trends somewhat changed over the last couple of months where IO-IO seems to be vectoring down, IO-TKI seems to be vectoring up. Single-agent TKI monotherapy is either flat to maybe going down a little bit. So I think that's the trend. If that continues, we'll see. I mean my crystal ball is as good as yours. But certainly, you've got a lot of very strong believers in supporters in IO-IO. And we all -- we've heard that narrative. I think the vast majority of prescribers, certainly, in the community, look at the value of an IO-TKI, especially when like cabo/nivo with all the great efficacy data and quality of life improvement relative to sunitinib and can see how that can help their patients immediately in a user-friendly sort of way, right? So again, we'll see, but we're certainly -- we're not targeting any one segment of that market, we're looking to address every eligible patient every single day. And we think we have the data and the team that can do that really well.

Yes. So COSMIC-313, we got an update at this meeting on the actual cabo/nivo -- I'm sorry, ipi/nivo plus or minus cabo in -- it was mostly first line. There was a little bit of second line. And the data, patients only got 4 cycles of ipi. So tolerability actually looked okay. It wasn't -- it didn't contribute a lot in terms of tox. It looked like it contributed a little bit, but it was harder based on how the data was presented to really tease out the triple versus the doublet. But when you looked at the data in totality and compared it to frontline, it was already better than the front line data and included second line patients. And a lot of the -- again, KOLs are mixed about the value of the triple and can it really be tolerated. So what really gives you comfort that this is really the way to go in the future?

Yes. Well, I think it's -- look, our job is to continually drive the improvement in standard of care. And I think that's one message I would send today to people interested in oncology, but also interested in our story is that oncology is the most heavily invested in therapeutic area across biopharma. It is the most competitive from the standpoint of different -- large and small players trying to move the needle. Our view is the only way you're going to make a difference is by improving standard of care, coming in late with me-too type data. So what, right? It doesn't really help patients. It doesn't really help HCPs do their jobs better to improve the quality of life and the survival of patients with these metastatic diseases. So our job is to push the envelope and ask some really hard questions, make those investments that give us a chance to basically raise standard of care. So the triplet mix, in that regard, really fits in well and it's the first of many that you'll see going forward, where we take best-in-class IO-IO, only in class, at least up until now, and what we think is best-in-class IO-TKI and ask the question, if you combine them, can you, again, raise the bar, right? So a simple question. Dr. Apollo's data across these rare GU tumor types is certainly very supportive of that. The theoretical idea that you can add cabo to ipi/nivo and not only raise the tail potentially, but address all the patients early on who progressed on ipi/nivo, which is a pretty high segment as well as looking at, say, cabo/nivo and ask the question, if you give a little bit more in terms of the IO enhancer, can you play the same game, too? So it's really fundamentally looking to mix and match and hopefully add the best of both. But it's a pivotal trial. You've got to run the experiment. You've got to open the envelope one day and see the p-value and see the adverse events and the overall tolerability, but that's the game that we're in. And we're not going to continue to push top line revenue growth and really the value of the company if we're not really aggressive going forward, so we are.

And you got to -- the plan is to finish enrollment this year and potentially data SEC next year?

Yes. I would say enrollment is essentially done. We're just finishing up the extra few patients that we added relative to making the -- powering around the survival a little bit more appropriate based upon the final, if you will, 48 month ipi/nivo survival follow-up where they actually got to a real medium. So that's gone really well. I'm really pleased about how well 313 enrolled, 312 enrolled during the middle of this global pandemic. We really had a lot of strong support globally, which I think is just indicative of the interest in the approach in the molecule and the overall idea that we want to raise standard of care. So yes.

The final study, Mike, was 690?

Final study was, I believe, final end was in the 840 range.

840.

Yes.

And the initial enrollment was just shy -- was -- remind us the original number. It was 760?

It was in the 650 range. Yes. Yes. We had a couple of hundred more patients just to basically improve the powering on survival, which obviously is pretty important readout in this case, Yaron. So...

Got it. Okay. Can we talk about COSMIC-312 in first line HCC? And just remind us, so that's fully enrolled. I think you -- we're expecting data in the first half of this year, if I remember correctly. And that one also was -- enrollment was expanded. Could you just remind us the number, and it was also designed so you can file globally and have a China specific, sort of, component to that.

Yes, exactly. So that kind of evolved over time. We started the trial and we added patients after we started the trial, but got feedback from the FDA that they wanted to see more patients in the single-agent cabo arm to help define the contribution of components better, which was easily done. So we made that amendment, I forget if it was PI2 or PI3 in terms of adding another 100 or so patients to be able to kind of equalize the arms for both cabo and sorafenib. So obviously, contribution of components for any doublet is critical. And the FDA and all global regulatory agencies want to understand that if you're adding a second molecule in, what's the impact on both efficacy, safety, tolerability, et cetera. So that was easy. Yes, trial has been fully enrolled in terms of the ITT population for, gosh, months now. I think we finished that back in August. We had hoped to have top line data at the end of last year. Events came in slower than modeled. People like to debate what that means. I'll stay out of that narrative. But yes, we're on track to have top line data in the first half of this year. Similar situation to the 2019, 2020 narrative with 9ER, right? Bev, atezo with IMbrave has survival advantage relative to sorafenib. So obviously, that's, again, raised the bar in terms of what you need to have a competitive offering. So obviously, we want to be able to win on both PFS and survival to be able to compete there. But it's a very similar trial to IMbrave. We're just swapping out cabo for bev. So with cabo's activity with the celestial data, we feel pretty good about that. So -- but again, we got to run the experiment, got to get the p-value, which should happen sometime first half of the year.

Right. Let's focus -- let's shift and I'm going to focus to a little bit of your ADC, your new pipeline. You've done a series of deals over the last 6 months. Can you give us a little bit of a sense on the overall strategy, both on with XB002, your tissue factor ADC and then your new deal with Catalent, NBE and most recently Adagene?

Yes. Yes. It's really exciting. We've been a hard-core discovery shop for decades and primarily focused on small molecules. ADC technology is, I think, a really good match of and really fits our strength in terms of really, really intense hard-core understanding of tumor biology, all the different pathways and networks and cell types that go into driving the tumor microenvironment modulating that as well, along with the mix of biology, if you will, biologics and chemistry, right? So we think it's a really great place to start building expertise and depth. And as we did with kinases 15 years ago, when we go into an area, we go in heavy. It's not -- you can't dabble on these kinds of things. It doesn't get you anywhere long term. So we're doing a lot in-house in terms of understanding the biological framework and potentially new warheads, new linkers that we think we could bring to bear here. But then you're working with leading companies that have great platforms that we can mix and match and almost play the a la carte game with relative to what's best for a given tumor type, what's best for a given binder, in terms of then how we deliver these payloads. So yes, the ICON molecule XB002 that we licensed in at the end of last year, really exciting, we think, is potentially best-in-class molecule targeting tissue factor. They did a great job of designing a noncompetitive binder with tissue factor in terms of the interaction between Tissue Factor and Factor VII. They've got, arguably, better next-gen linker warhead from Zymeworks and some data which supports that preclinically. So that's, I think, kind of a good example of how we want to maneuver here going forward. The Catalent and NBE deals have gotten off to a great start. It's amazing to me how fast those guys move and how fast we can take ideas about binders and then work those into their systems to be able to get ADCs out the back end. So we have already seen some pretty interesting results that is driving our -- I would say, our focus in urgency and getting those things moving forward towards development candidates as we go forward. So it's a little bit longer process on the back end in terms of getting master cell banks made and those kinds of things to be able to go into full scale production. But we're really excited about this interlap -- this interaction, this intersection between chemistry and biology in a way that we've been able to exploit in the past and think we can do it now with a brand-new modality as well. So stay tuned. There are lots going on.

Yes. And when you think of Tissue Factor, which cells do you think are particularly good targeted by Tissue Factor, right, Tissue Factor VIII you see?

So we talked about that on the last few calls. We think it's pretty broadly expressed as an address, if you will, around certainly cervical cancers where there's the most data in terms of the first-generation molecules. But if you look at -- you look broadly at expression. You think about lung cancer, you think about ovarian cancer, bladder cancer, head and neck, pancreatic. So again, we've done, I think, really well. And I think the COSMIC-021 study is a good example of where we have taken hints of activity or hints of kind of biology, expression data and then being able to really design these broad basket-type trials where we can then maneuver quickly, add patients where we see activity, think about accelerated approval. So everything we're doing right now is really built to do that. So the learnings from 021, the ability to basically leapfrog and really save time and focus your activities and your investments is the way to go. So with 002, with 102, the CDK7 inhibitor, things that move into development in the future will play that same game: get to the MTD, start thinking about combination approaches and then do these basket studies where we can really -- with our clinical development expertise and depth and just the power that we built over the last 4 or 5 years be able to rapidly expand the data that we have, that gives us more optionality than how we strike for approval as early as possible, right?

And so the differentiation versus tisotumab, or TV, is really related to the Tissue 7 activity and potential hemorrhage?

I think it's both arguably a better binder and arguably a better warhead. So the question -- and I don't want to get too far ahead of things, but can you push the dose? Can you -- because you're not worried about blocking the interaction with the coagulation cascade, can you push the dose? And with a better -- arguably better payload, can you actually see better cell killing? So the question is, can you just -- can you really get better activity than what you would normally see with kind of baseline first-gen molecules? So we have to prove that, obviously. So not saying it's going to happen, but I think that's the potential. And again, if you're going to be second, you've got to be able to differentiate from the first. And that's been our view from literally day 1 with cabo, right? We're not always going to be first, but we have to be best if that's the case.

And to file the IND, just remind us in terms of CMC, what are the last steps that you need to do?

Yes. That's -- all that release assay validation is basically done. We will be filing the IND imminently. And then once we hear back from the agency, we'll get ready to go in demand.

Okay. So I was going to shift over to CDK7. And then XB102 or XL102, and the data you showed at the triple meeting was specifically looking at mono. I think it was combo with docetaxel as well in ovarian, triple-negative and metastatic prostate. Again, it sounds like you're moving into an advanced solid tumor sort of COSMIC-like COSMIC-021 like study. Which tumors do you think are particularly -- would potentially be good candidates for CDK7 modulation?

Yes. I think it's a really exciting area based upon -- it's upstream from CDK4/6, the role that CDK7 plays in terms of activating various, say, nuclear hormone receptors like the ER, estrogen receptor, et cetera. So we're thinking, again, it makes sense to go after ovarian cancer, go after a variety of breast cancer subtypes, triple-negative cancer, ER-positive breast cancer. We're really -- we have a pretty good deep bench around metastatic CRPC prostate cancer, and we think this could play a role here in combination with an NHT as well. So -- but again, same exact idea, get in demand, and we've been in demand now for a couple of months and certainly very pleased to see the progress in terms of enrollment there. Get to the MTD, start doing combination work as soon as we get some doses with the single agent, get to the MTD there and then start looking at these very, mechanistically driven tumor types and really ask the right question at the right time in terms of can we get some quick hits and some quick wins based upon our understanding of the biology and the role this target might play across the board. So it's pretty simple pharmacology. The question is, is it real? Does it work? And we have a probe not to ask those questions clinically.

Is there -- I mean, when you look at other CDK7 drugs out there, any thoughts on where you can differentiate? Or how wide is the therapeutic window of the CDK7 or how to best develop it?

Yes and no. I mean some of the earlier first-generation molecules had pretty significant liabilities with both potency and, I would say, oral bioavailability. So it's hard to say that. I guess I wouldn't put too much credence on that, too. You go back to cabo and the VEGF game, we had ideas that we had approved clinically, which we did and certainly worked out really well. So look, we understand the different levers we have to play with. And the idea that what you see in a mouse may not translate to demand, but that's our job to get that data as quickly as possible. And then when it looks good, just crank, right?

Okay, terrific. 092, I was going to maybe -- this is a program that we've talked about several times. I think we've done actually a webinar with you about 2 or 3 months ago, in terms of -- this is a program that also, as you said, you're cranking toward an MTD in Phase I then moving to combo. At what point do you move into a pivotal? Because the next step is to figure out MTD as a mono, do the checkpoint inhibitor combos, get to the right dose and then move into an expansion cohort that could even be pivotal-enabling. And you're looking at white spaces where maybe cabo is not used where checkpoint inhibitors are not used, but there's a lot of white space. So how do you get to -- the challenge is, how do you move quickly, get to some kind of a proof of concept, so you don't have to run massive studies at risk as you move into a pivotal? So how do you sort of balance that -- the 2 sides of that?

Yes. So it's actually -- in my mind, it's a pretty simple approach. You want to -- we've validated the initial hypotheses already in the clinical setting around how we designed what we think is a better next-gen cabo with a shorter half-life, more user friendly. You've heard the narrative, I won't repeat it again. But all that is pretty much lockdown. Now we need to validate the activity clinically, both as a single agent and in combination. We know where to look, right? We know how to do that, and we know how to check those boxes, I think, pretty quickly, based upon the activity profile of cabo. So we don't need a lot of data there to feel good about how it's vectoring based upon what we've seen with cabo to date and that benchmark relative to how we could frame 092 as an active -- a clinically-active molecule. The white space concept, I think, is a really important one. And it's one that we have done so much work with cabo over the years, but really only taken a couple of indications forward with the depth and breadth that you'd like to see to be able to build a franchise molecule that can drive top line revenue growth, right? And so we have -- we're sitting on all this extra data with cabo that we really haven't pursued yet. So all those learnings transfer immediately over to 092 once we feel good about the validation that we've got in this early clinical work, and we can really build upon that, right? And there's so much more going on when you think about what's happening with other -- other IO, if you will, add-ons on top of PD-1, PD-L1, other approaches where you could see a molecule like 092 with its pleiotropic activities across all the important targets and pathways that drive tumor cell growth, tumor survival, tumor angiogenesis, the immune system. So it's really -- especially one that's packaged better, if you will, from a usability point of view in the clinical setting as you titrate those doses. So again, it's -- we've done an incredible amount of work, and we're really excited about going to the next level. You talk about starting pivotal trials. I'm interested in starting waves of pivotal trials going forward, because I think that's where we now use the clock to our advantage. We have these -- we certainly have the financial depth and the free cash flow to be able to invest heavily here while we're working on the ADCs, while we're doing new molecules. And for us, it's all about driving that opportunity to really build top line growth across a portfolio of assets. So we've been -- we've been successful with cabo, and that was the tactic, get cabo over the goal line, have that be the driver of the business. But now we have to expand, and that just takes a little bit of time. But the team is growing and it's very excited, and we're just cranking right now, right? So I'm super jazzed.

A question about what are the next steps in terms of the cabo patent case to -- that way that investors will really have some kind of understanding what's the long-term visibility? This is a question we get almost on every call in Exelixis these days.

Yes. I get a few of those, too, right? So I -- yes, as we've said, we're not going to litigate that publicly. We've got great data, great patents. And the team is just outstanding, both in terms of internal expertise as well as our outside counsel. So what's publicly known is that there's no markman. So I think people can interpret that as they will. But my view is that the claims were very simply written. So it's kind of hard to narrow them further if they're already as simple as they can be. So I see that as a good sign and a kudo to our legal colleagues for -- as I always said -- as I was talking about patents back when I joined CIBA-Geigy back in the '80s, keep it simple, right? Because if you have to litigate this, you want to make sure that it's understandable and it's a very simple kind of narrow focus, which is great, right? So they did a great job. The trial is scheduled for middle of next year. So that's the next big milestone. Obviously, there's lots of stuff going on between the markman that didn't happen and the trial. So -- but I don't want to expand upon that right now. I'm sure my lawyers would not appreciate that. So I'll be somewhat reserved there. But look, it's a super important priority for us, and we're leaving no stone unturned to make sure that we do everything we can do to be some...

There was a single filer so far.

Yes.

And can you remind us who that was?

That's MSN.

Okay. All right. I think we have about 3 minutes left. Between 092 and 102, you're focusing on breast cancer now or you have a new focus on breast cancer. How do you use 092 and 002 to map up your strategy? Is it one of them is going to triple negative, the other one is going to go to ER positive? One of them moves up, the other one saw a single-relapsed refractory? Or do you do a sort of a redundant and see who looks better and take it from there?

Yes. I would say our focus on breast cancer is indicative of our interest everywhere, right? In terms of solid tumors, we have a lot going on in lung, a lot going on in GU, prostate, obviously, we're hitting that hard. Very interested in colon cancer, have something pretty interesting data there, too. So I would look at breast cancer. You talked about it, right? I don't think we're smart enough right now to be able to say based upon preclinical data and the first few patients, we're going to take this molecule this direction and take the other molecule in that direction. Our view is you need to get the right data in the right species at the right time and let the human clinical data define priority path, urgency, next steps, et cetera. So we will invest here appropriately to be able to get that key data. And then we will be, I think, very disciplined with how we then move forward in a very data-driven fashion to make sure that we're investing the appropriate money and time and human resource with what really is called for, right? So -- but that goes across the board. That goes for prostate cancer, to liver cancer, lung cancer, everything. We're going to be heavy. We have to continue, again, push the bar in terms of improving standard of care. And if that is the case, if we're successful there, then we think we can monetize that and help a lot of patients in the products.

Maybe last question. I think it's actually came in -- it's related to -- there's a few questions related to checkpoint combos with -- in the CRPC as well. Can you give us a sense on just the 3 relevant cohorts in 09 -- 021 and what just mind us on data timing and enrollment status?

Timing for -- most important metric is filing the sNDA if the data continues to look good. So we talked about doing that in 2021. That is my #1 priority this year is to get sNDAs filed for 311, 312 and 021 prostate cancer. Everything else is secondary in terms of how we publish, how we present. We did so many presentations last year. People should expect more of that when we have time and the when data is mature and everything else. But our #1 focus is filings because filings drive approval, approvals drive top line revenue growth. So it's a pretty simple.

The combo arm, I believe, is fully enrolled.

And it has been for months. And we've talked about that step there being the BIRC review of the response rates. Yes.

Which is expected mid-'21 in-house?

Yes. When -- it's here when it's here. They're doing a lot for us right now. So I don't want to comment on giving guidance on timing, but that's the key milestone because we have to look at that data relative to the investigator data and then say, "Okay, is it good enough to file on or not."

And the 2 monotherapy arms, are they fully enrolled? Or they're still on that?

They are just wrapping up their enrollment. We're enrolling the last few patients there. Yes.

Okay. Terrific. I think we're out of time, Mike. Great same views always. Thanks for the time. I'm sorry?

I'll see you next year.

I'll see you next year in person. Up on, this can be fine.

I hope so. See you. Thank you.

Thanks, everybody. Really appreciate it.