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

I'm, Nareg Sagherian, Managing Director of Investor Relations and Programming here at Bio. Welcome to the fireside chat with Michael Morrissey, President and Chief Executive Officer at Exelixis. Visit the Bio Boulevard to learn more about Bio's new podcast, featuring biotechnology breakthroughs, the global problems they solve and the stories of patients whose lives have been changed forever by our work. You can also download our first annual diversity and inclusion report at right-mix-matters.org. We'd like to take a moment to acknowledge the General Sponsors that support our conference. I'll also leave our disclosure statement on the screen for a moment. So without further ado, I'm going to introduce our moderator. Please help me in welcoming Asthika Goonewardene from SunTrust Robinson and Humphrey.

Excellent. Thank you, Nareg. Hi, good afternoon, everyone, and welcome to the 2 p.m. fireside chat with Mike Morrissey here. My name is Asthika Goonewardene, Senior Biotech Analyst with SunTrust. It's great pleasure to have Mike here with us today. I'm sure Exelixis is not a stranger to a lot of you in the room. I'll take a minute, Mike. I just want to give you a quick run-through of your illustrious career here. You started off with -- got your Bachelors in chemistry from University of Wisconsin, got a PhD in Chemistry from Harvard, started your career as a scientist, CIBA-Geigy to Berlex to Exelixis in February 2000 in Discovery Research, because 2007 to 2010, you were President of R&D. And then July 2010, you took over as President and CEO, and have held that spot and done, I would say, a very fantastic job since. I guess we don't need to dig too much into what Exelixis is all about, we're all familiar with the company. I cover the company personally in my role at SunTrust. It -- have a buy rating on it, price target actually went up today $35. Yes. It's one of the few profitable mid-cap biotechs in the space which is quite unique. And -- but it also has a history that spans over 25 years.

But Mike, maybe start us off with -- when you took the helm in 2010, you faced some interesting challenges. Tell us about that, please?

For sure, thanks. Everybody, thanks for coming out, and thanks for that kind intro, feeling very old right now. So I had my 20-year anniversary at Exelixis last week. So it was a bit of a fortuitous milestone. So it's about 10 biotech half lives in terms of my longevity there. But it's been a great ride. I've had just tremendous opportunity to learn and grow and apply and with the team build some pretty strong momentum right now. So -- but yes, let's talk about the 2010 time frame that transitioned where we were as a company then, where we were hopefully going. It was a moment in time that was pretty dire for a couple of reasons. In that point in time, we had just received cabo back for the second time, cabozantinib, our lead molecule. We discovered the molecule as part of a GSK collaboration, they passed on in-licensing that as part of that deal. We flipped it relatively quickly after that to BMS in end of 2008, early '09. They spent a couple of years working on it with us, and then at the dawn of the whole IO revolution, they had just bought Medarex and decided to put more money into IO in terms of their portfolio prioritization efforts, and that prompted us to get the compound back for a second time. So it was a 2-time return. A couple of weeks after that, George Scangos, our then CEO left to go to Biogen, great move for him, great move for them, and I took over the -- at the helm of Exelixis. So it was a moment in time where we needed a, I think, important frameshift. The preceding decade, we had focused a lot of our efforts on doing really high-end industrialized drug discovery and early development. And that we were following the Intel Inside model, where we hope to have each one of our compounds focused in oncology or cardiovascular or information, be part of a big pharma portfolio and then be able to monetize future value creation through the efforts of those pharmas doing full development and then commercialization with those efforts. That was a pretty labor-intensive and certainly money-intensive effort. We -- in the decade or so that we spent focusing on that from the time I joined to time I became CEO, we probably brought in about $1 billion in real, not bio bucks, but real hard cash in terms of a variety of different collaborations. The downside there is we certainly spent more than that to be able to move about 20 compounds into development, into the IND state and beyond. So that model of doing high and high risk, usually high attrition drug discovery, early development, stopped paying for itself. And there is lots of debate about how we were raising money, how we were burning cash and not really seeing -- or providing a clear picture to how we would kind of get out of that cycle. So when I took over, and I would say, about a year or so before then, it was becoming more and more clear to me that for us to be successful in building a sustainable business, and the keyword is sustainable, we had to transition from that discovery, early development shop into a full-blown development, pivotal trial and commercialization organization. And even going back to my days at Berlex back in the '90s, I think there were some pretty good examples of that speaking, which was Biogen that really turned the corner on their transition from discovery development to commercialization with the advent and the launch of Avonex, which was a competing product for what we had at Berlex with Betaseron in the MS space. So I was early in my career, it was very enlightening to see what -- just kind of speaking the obvious, what a product could do in terms of not only revolutionize what a company does, but providing the cash flows and the financial stability to be able to then build a business that you can then take forward based on what -- on the free cash you could generate. So that was 2010. We spent, I would say, about 6 months debating that at the Board level and making a lot of I think important analysis about what was involved there. But end of the day, we agreed that tactically, we would become the cabo-only company. So we stopped everything else, we let a lot of people go who were not involved in the cabozantinib development effort in an attempt to focus our money, our people, our intellectual capital on making cabo into a drug. And that -- there was lots of drama there in terms of how that went, how that happened, we had certainly some false starts. But we had a basic belief in the compounds activity, even back then, we had seen broad activity across a variety of different tumor types. We had seen early activity in thyroid cancer that looked encouraging. So we knew it was active, the question was how active could it be? Did we have the right dose? Did we have the right patient population? And did we have the right development plan to be able to monetize that quickly? And that was part of then the story going forward.

I mean it was quite a big bet to take, right, at the time. What gave you the confidence to really get behind that? And I'm also curious how did this play out with the investment community, the Board to pretty much become a cabo-only company?

Yes. So I think the basic premise was that we had enough internal data to highlight it was active, right? There's clearly activity across the board in various tumor types, even back as early as 2008, 2009. But again, as we all know, seeing activity in the first handful of patients versus doing large global randomized pivotal trials is a very -- I mean that's how you prove activity, right? So we had, as you would imagine, a very robust debate at the Board level about what it would cost to do that? What was the risk to do that? What was the upside to do that? And it's the ultimate forward-looking estimation of -- you have limited amount of data. And then you project forward in terms of a base case, the downside case and certainly upside case of what it could look like if you're correct. But that's -- in essence, that's what we're paid to do, it is to make decisions that are far-reaching and very impactful across all the different components of the business based upon limited data. And there's -- we talked about this a lot. There an element of being really hard-core scientists and data people, coupled with -- and just opposed with the idea that you have to believe in what you've got and believe in the upside. And it's that balance if you're 2:1 or the other between data and bleeding, you can often go down a false road or false avenue that can end up being very nonproductive. So we made some tough decisions. We stopped a lot of programs that we thought didn't have the traction that we needed, and we went forward. And that was part of the drama. Certainly, from an investor point of view, everybody liked the idea, even back then that we've been focused, because it's a very easy bet to make then in terms of a binary bet in terms of is it going to work, with trials, whatever blah, blah, blah. Employees were -- it was a big cultural shift from the standpoint of having a very broad focus around different targets, different pathways, different molecules across the board to then saying, okay, guys, we're the cabo-only company, and we're going to place all our bets here. So end of the day, it's worked out, right, and certainly with -- even with the drama that we saw with the COMETs in prostate that not working, and the company essentially cratering in the beginning of 2015. We had -- literally were down to probably about 80 people overall on 1 floor of probably about 6 buildings that we had at that point in time. We had 3x more debt than cash. And we had basically one last shot in renal to make that work. But that's the -- it's the beauty of the business that we're in, because that data from midyear was so compelling and so differentiating in terms of both the PFS benefit, but also a survival benefit, strong response rate that had really redefined what a unique targeting TKI could do in that space. And since then, we've generated about $2.6 billion in cash through our U.S. commercialization efforts in terms of revenues and then our partnering milestones and upfronts and royalties. So it's put us on a whole different plane in terms of now being able to then have that sustainable cash flows and been profitable for the last 12 quarters. So it has certainly given us a lot of optionality in terms of how we then tactically proceed. But focusing on cabo was never our strategy, it was a pure tactic to get to a place where we are today in terms of profitable, generating free cash and lots of optionality going forward to build the pipeline.

Yes. I mean I want to talk more about the next stage of evolution, but I also want to, just for people in the room, just maybe run through what cabo has become, its approvals of monotherapy in frontline and second-line kidney cancer, second-line liver cancer. It's a leading single agent tyrosine kinase inhibitor in kidney cancer. And in the near term, you have quite a good amount of catalysts coming up. You have CheckMate 9ER, COSMIC-312, 313. Yesterday, we got the first proof of the -- your come back efforts in prostate cancer. But let's actually take a minute and reflect on the data that the prostate cancer, I'm sure people on the webcast looking forward to hearing your initial thoughts on that, Mike. What excited you about the data that was presented in the abstract in your press release?

Yes. It's a good segue. It's the mental, emotional, physical scars from the COMETs still live with us all in terms of that effort. We had such, I would say, remarkable data back in the 2013, 2014 time frame around the ability of cabo to impact the appearance, the generation of -- and driving the disappearance of bone lesions by bone scan, that it was a really attractive new way to think about the molecule, think about the disease. Since prostate cancer certainly in its advanced stages is a bone-predominant disease. That was -- if you go back, that was a very controversial topic, certainly not a clinically validated endpoint, something that we went into fully knowing the risk involved in kind of pushing the envelope around this new way to look at response assessment. But that didn't work, right? And if you go back to the story of COMET-1, which is about 1,000 patients global pivotal trial looking at patients post NHTs, post chemo against prednisone. For a variety of reasons, that trial didn't work. Hazard ratio is 0.9 for overall survival. We saw a benefit in PFS. We saw a clear benefit in patients with visceral disease so that was the first sense of activity. We saw very interesting activity in terms of the delay of skeletal-related events. So from a regulatory point of view, it was a clear miss, and we -- I was proud of the way that we owned that upfront and kind of pivoted quickly, acknowledging the negative data. But we have kept kind of in that prostate cancer area in terms of thinking and talking to investigators and analyzing the data really for the last 4 or 5 years very intensely. It's something that we all really believed in, and we all think cabo has activity. The challenge for us is to learn from the COMETs and then design a better approach to be able to prove that. And I think we've done that to a large degree with the COSMIC-021 study. Cohort 6 is the main cohort that we talked about. You talked about just now and that we have data on at ASCO GU this week. We had a press release last night with the abstracts coming out. We took a step back and simply asked the question, one of the learnings from COMET was we tried to hit a grand slam and we struck out, right? So the question was, can we look at the disease, and this is a very complex disease from the way it develops, from the way it evolves in men, sometimes over a decade or more. And can we basically dissect out various components in that disease? And then ask relatively simple questions around the ability, in this case, cabo and the combination of cabo with atezolizumab, the PD-L1 inhibitor from Genentech-Roche, around its ability to have an impact on different parts of that disease in a separate but sequential fashion, right? So -- and we've carved that out, at least right now in 3 different segments. The first part, what we're doing now in cohort 6 is looking at men with metastatic castration-resistant prostate cancer, so CRPC, who have failed on at least 1 novel hormonal therapy, XTANDI, abiraterone, some of the newer ones from either Bayer or J&J. To ask the question, can this double it? Cabo/atezo actually be effective in that -- either first or second-line population, depending upon when they got the first NHT to be able to delay the use of chemotherapy. Very simple hypothesis, very simple, I would say, response assessment tools as well. We're specifically -- one of the learnings from the COMETs is certainly bone component of this disease is very important. Per, again, clinically and regulatorily validated endpoints, you can't use bone for response assessment. It's very hard to measure, especially in a positive shrinking sense. So we have decided to focus exclusively on measurable disease, soft tissue, visceral mets, where we saw some hints of activity in COMET-1 to be able to, again, make the problems simpler, if you will, kind of slow down, ask some very fundamental questions using clinically validated tools like RECIST 1.1 to ask the question about response assessment. And that mandates that you have patients with measurable disease, either in the visceral or soft tissue disease. So we did that. We started that back in -- at the end of 2018. We've expanded that cohort twice. We've added additional cohorts. The first data set was -- came out, again, with the abstract form in our press release yesterday. First 44 evaluable patients, we have a 30-plus percent response rate, either in the overall population or in this high-risk population. The disease control rate's about 80%. We have a duration of response in the 8-plus month range, good overall duration of treatment, about 6 months. The tolerability was very, very encouraging, low level of grade 3, 4 AEs. Our study treatment discount rate was about 7%. So we're seeing very good activity in a population of patients who got previously either 1 NHT or 2 NHTs. It was about a 50-50 split in the overall population. About 27% of patients actually had chemotherapy for castration-sensitive prostate cancer early on in their disease. So some were even pre-feeded with chemo, and we've seen good activity across all 3 of those different subpopulations. So it's very encouraging from the standpoint of having an endpoint in terms of response rate by RECIST 1.1 that there's no debate about, that's important and having the level of activity as well as tolerability, which looks very encouraging, especially in the context of the old data that we have from COMET-1.

Absolutely. And also, I mean the treatment paradigm has also been involved in since COMET-1 as well using the NHTs ahead of chemo.

Yes, exactly.

So this is almost a very good layup for you to come back. But how do you think about PD-1 and immunotherapy? And how that was moving alongside while you guys were essentially reevaluating how to approach prostate cancer? What made you think about combined with PD-1 in here, when initial data showed single agent -- a very low single-agent activity with PD-1 in prostate?

Yes. Well, that's true with both cabo and various PD-1 -- PD-L1 molecules. The response rate by RECIST is relatively low. But we've got a lot going on in that space. Certainly, we've got very encouraging data with cabo and nivo in RCC. We had some data with cabo and nivo and cabo and nivo at being HCC at ASCO-GI a couple of weeks ago. That's showing good response rates, duration of response, all the same kind of phenotype in terms of single agent, nonrandomized data, which is showing pretty encouraging overall activity. So we're seeing a clinical phenotype that looks encouraging, right? And that prompted us then to do this large basket study, where we've got 24 different cohorts now across 12-plus different tumors looking at the cabo/atezo combination. So the -- since the early advent and success of the PD-1, PD-L1 molecules, there's been a very intense focus across the industry, looking for combination partners that can both kind of drive the tail up and move the OS, PFS to the right, if you will. That's mostly been unsuccessful with novel IO/IO combinations. If you look at the real successes that have come with really chemo with the TKIs, maybe some other maybe more standard therapeutic approaches, things like cabo. Cabo has very, at least seeing this as a single agent, very rapid onset of action, can certainly see tumor shrinkage quickly, release antigens, goes after the TAM kinases as well as VEGF, and kind of inducing more immune permissive environment. So the combination of cabo, which is kind of immune inducing along with the PD-1 molecule, which is obviously very well set up to then release the break and let that checkpoint go is certainly a big part of that process. So -- but data looks encouraging across different tumor types, and we're excited to now be able to have one more indication like prostate cancer to be able to add into that mix going forward, for sure.

I mean just based on response rates, I would think that the setting, the prostate cancer setting is one where really I think surprised -- I was very pleasantly surprised to see that sort of a synergistic effect. Clearly, the TAM kinase targeting has been highlighted by a lot of KOLs to us as to being a very unique component of cabo, and that's what makes it work really well with immunotherapy. Going back a bit, how much of that was actually part of the initial design of cabo, thinking about hitting the TAM kinases and the wait was...

Yes. I'm not sure the TAM kinases were known back when we designed cabo back in the 2003 time frame. So the main focus for cabo, going back to a trip that Peter Lamb and Kirk McMillan and I took back years ago was really focused around tying up targets that block VEGF as well as targets that are -- to kind of drive resistance to VEGF therapy, which was where the whole MET/AXL axis came in relative to VEGF. And it was just being discovered back in 2003 or so by some really elegant work that showed that if you inhibited VEGF with a small molecule or biologic, you would see pretty dramatic upregulation in MET and AXL, which then made the tumor cells much more invasive in terms of their phenotype that really drove metastases, which is what you also saw clinically as well in terms of, say, VEGF inhibition by back then SUTENT or Avastin, right? So there were early clues around how to make a better molecule, and I think we did that in a very rational fashion by tying up both VEGF and the targets that drive resistance into a single molecule. Certainly had some targets come along for the ride there. When -- as we were profiling the compound more and more both at the molecular level, different kinase panels, but also different cell types, it was pretty clear that we were hitting all the important players from a cell biology point of view in terms of the tumor, the tumor vasculature, all the different cells involved in that and the immune system on both the adaptive side and the innate side of the immune system. So we've got a lot of things going for us, which seem to be validating well in preclinical models and now to see that now transform into the clinic is certainly very gratifying, right? So our job is to simply, first, broadly profile cabo as much as we can, which we're doing. Again, you -- we talked about this a few minutes ago, we have -- with the 3 trials, we'll start with Genentech this year in prostate, lung and renal. That will give us 12 large global randomized pivotal trials that we're doing either ourselves with our partners, Ipsen and Takeda, with our clinical partners, BMS and Roche as well as with our partners at the NCI. So it's a pretty broad cut of different large label-enabling trials. We expect 6 of those to read out this year that could drive 4 new indications in 2021 if we have good data and we file and that kind of stuff. But that's only really half the effort. The other half is, can we make a better cabo in terms of understanding the clinical pharmacology, the activity profile, having a decade-plus of experience. Can we make believe next-gen molecule. I think we have that with XL092, and we'll talk more about that later. We have another molecule XL265, where we've taken the TAM kinases as a focus and made that a relatively selective molecule for those as well. So we're taking the learnings from cabo and now dissecting pieces out of that to be able to understand if we can have more selective activity around that. So that's exciting, too, for sure.

Yes. Before we -- I'm going to definitely dig in more on the pipeline and what you got coming up. But I also want you to talk a little bit more about the broader COSMIC-021 study. I mean this is 20-plus arms. This is a really big study, a very interesting effort to figure out why to -- how to fit cabo in many different tumor types. First off, I mean how much does that cost? But two, maybe tell the audience a bit more about that study and what you're aiming to achieve out of it?

So we've done these kinds of baskets before. That was one of the early trials we did with BMS which had a similar focus. That was more focused on looking at PFS as an endpoint. This is pure response rates. But it's a simple way to be able to broadly profile, in this case, a large number of tumors and indications that we have some level of activity with cabo and asked a simple question, can you improve the profile by having this cabo/atezo doublet in terms of all the relevant endpoints in terms of response rates, duration of response, can you take the best of both worlds and then find other indications to pursue quickly, as we've done with, say, renal and liver and now prostate with 021. It's a very cool design because we have the built-in -- with the protocol, the built-in flexibility and optionality to rapidly add patients to follow up on early signs of activity. We can add cohorts relatively easily, too. So we have the flexibility to mix and match and merge and purge different approaches here to be able to really take early signs of activity and then rapidly profile those and pursue those and refine our questions based upon what we're doing. So it's worked out really well. We've expanded several cohorts so far, certainly, and most notably, both prostate and post-IO lung. We've added cohorts for those as well in terms of single-agent activity cohorts for cabo and atezo. We've looked at different cohorts for renal and others. So again, I think we're really maximizing the value for our understanding of this doublet's potential but also for patients as well. So we're really excited about that, and it's been something that we can drive quickly ourselves, which is nice, too.

And also in structuring a trial like this and figuring out what the right patient population to go after, you need to have a good understanding of what is the patient's unmet medical need, not just now, but in many years from now. So I mean looking at the lung cohorts that you're looking -- that you are studying, you're looking to go after checkpoint inhibitor failed patients as one. Even the prostate cancer, it's patients who've had 1 or 2 NHTs, who now -- your next option is chemo and that's what makes the cabo/atezo combo look so good because it has similar efficacy as chemo, but much better safety.

Yes. Exactly.

How do you look forward like this and figure out what the patients needed?

Well, it's a good question. And I think what we've done here, if you look at across prostate, lung, renal, bladder, others is you have individual cohorts, looking at individual clinical phenotypes, clinical indications. So bladder, I think we had 4 indications. Lung, we have 3. Renal, we have 3, et cetera. Prostate, now we've got 4 or 5. So it's being able to understand what the need is now project forward and say, this is the most likely scenario where the need will be later. I mean the example of IO refractory lung, I mean that's a need right now, right? All these people getting pembro or pembro, chemo front line. And some do really well. Unfortunately, many of them progress relatively quickly, and you have -- those patients have an immediate need for better options in terms of this refractory, IO refractory phenotype, right? So I think we've got the right mindset in terms of understanding what the current need is, what the need is in the foreseeable future, but we have the flexibility then to add cohorts rapidly as we either get data or as we see data emerging on the upside.