Good day, ladies and gentlemen, and welcome to the Exelixis Fourth Quarter and Full Year 2018 Financial Results Conference Call. My name is Andrew and I'll be your operator for today. As a reminder, this call is being recorded for replay purposes. I would now like to turn the call over to your host for today, Ms. Susan Hubbard, Executive Vice President of Public Affairs and Investor Relations. Please proceed.

Thank you, Andrew, and thank you all for joining us for the Exelixis fourth quarter and full year 2018 financial results conference call. Joining me on today's call are Mike Morrissey, our President and CEO; Chris Senner, our Chief Financial Officer; P.J. Haley, our Senior Vice President of Commercial; Peter Lamb, our Chief Scientific Officer; and Gisela Schwab, our Chief Medical Officer, who together will review our corporate, financial, commercial, research, and development progress for the fourth quarter ended December 31, 2018. During the call today, we will refer to financial measures not calculated according to generally accepted accounting principles. Please refer to today's press release, which is posted on our website, for an explanation of the reasons for using such non-GAAP measures, as well as tables reconciling these measures to our GAAP results. During the course of this presentation, we will be making forward-looking statements regarding future events and the future performance of the company. This includes statements about possible developments regarding discovery, clinical, regulatory, commercial, financial, and strategic matters, all of which involve certain assumptions, risks, and uncertainties. Actual events or results could, of course, differ materially. We refer you to the documents we file from time to time with the SEC, which under the heading "Risk Factors" identify important factors that could cause actual results to differ materially from those expressed by the company verbally and in writing today. And, with that, I will turn the call over to Mike.

All right. Thank you, Susan, and thanks to everyone for joining us on the call today. Exelixis had a strong fourth quarter from a clinical, commercial, and financial perspective, which is a reflection of our momentum and overall success throughout the full-year 2018. As you'll hear from the team today, revenues, earnings, and cash continued to grow in Q4 as we executed on our strategy to make CABOMETYX the No. 1 TKI in renal cell carcinoma, or RCC. We're using the momentum from the last year to move aggressively in 2019 as we work to launch CABOMETYX in a new indication, initiate the next wave of potential label-enabling studies with cabozantinib, and rebuild our pipeline. As was the case in the third quarter, we again saw meaningful growth for the CABOMETYX RCC business in Q4, while demand notably decreased for the other TKIs in this indication. Our cumulative data from 2018 supports the projection that cabozantinib's best-in-class TKI profile continued to drive strong growth in the face of emerging competition from immune checkpoint inhibitor, or ICI, based therapies. The fourth quarter and full-year 2018 results highlight that the foundation of our business has never been stronger and that the prospect for future growth remains high in the face of widely anticipated competition. I'll begin today by providing a brief summary of our key Q4 milestones and then turn the call over to Chris, P.J., Peter, and Gisela for details of our Q4 financials, the commercial performance for CABOMETYX, an update on our emerging discovery efforts, and, finally, cabozantinib's development activities. Key highlights for Q4 2018 include, first, the significant growth in revenue, earnings, and cash based on the strong commercial performance from CABOMETYX in advanced RCC. Cabozantinib franchise revenue for the quarter and full-year 2018 was $176 million and $619…

Thanks Mike. I'm very pleased to share with you our strong financial results for the fourth quarter and full year of 2018. We are proud to say this is our eighth consecutive quarter and second consecutive year of operational profitability. The company reported total revenues of $228.6 million in the fourth quarter of 2018 and $853.8 million for the full year of 2018. Total revenues included cabozantinib net product revenues of $176 million for the fourth quarter and $619.3 million for the full year of 2018. Total revenues for the fourth quarter and full-year 2018 also included the recognition of $27.9 million and $164.4 million, respectively, of milestones from the company's commercial collaboration partners, Ipsen and Takeda. On a quarter-over-quarter sequential basis, our CABOMETYX net product revenues increased by approximately $13.4 million, or 8%. This was the result of an approximate $9.9 million increase in CABOMETYX patient demands, an increase in wholesaler inventory of approximately $3 million, and approximately $3.3 million which reflects the positive impact of the price increase we took at the end of the third quarter. Offsetting this price increase was an increase in our deductions from gross sales in the fourth quarter of 2018, which increased to 17.4%. This was a 1.5 percentage point increase when compared to the third quarter of 2018, which was 15.9%. The increase in our deductions from gross sales was primarily related to year-ending wholesaler inventory that will be dispensed to Medicare patients, many of whom will be entering the donut hole in the first quarter of 2019, as well as commercial patients that utilize our copay assistance program more frequently during the start of a calendar year. Overall, for the year, our CABOMETYX gross to net came in at approximately 16.3%, which is in line with our guidance that gross…

Thank you, Chris. I'm very pleased to discuss in detail with you today of the strong performance of CABOMETYX in the fourth quarter. We achieved our goal of establishing CABOMETYX as the TKI of choice in RCC, as demonstrated by the strength of the Q4 business fundamentals, updated NCCN guidelines positioning CABO more favorably, and continued gains in market share in the TKI market. CABOMETYX demand grew at 6% in Q4 relative to Q3, an increase over the 5% demand growth rate in Q3 relative to Q2. The prescriber base of CABOMETYX continued to expand and grew by approximately 10% in the fourth quarter. Also, we saw broad utilization across academic and community settings, lines of therapy, and clinical risk groups. We are pleased that CABOMETYX continued to grow in Q4 despite the increased competition. In order to get a better sense for the evolving competitive dynamics, we analyzed the IQVIA National Prescription Audit data for the RCC oral tyrosine kinase inhibitor, or TKI, market, comprised of sunitinib, pazopanib, axitinib, and CABOMETYX. Based on this publicly available prescription data, CABOMETYX has continued to grow in market share, increasing to 34% in Q4. This represents significant progress and a share point increase of 2% relative to the last quarter and 11% from Q4 2017 a year ago. In fact, these data also show that CABOMETYX became the No. 1 newly prescribed TKI for RCC. This trend continues and we are using it promotionally to ensure physicians are aware of this fact, which is reinforced by the current NCCN guidelines in RCC. You may recall in September, the NCCN published updated guidelines for kidney cancer. CABO is now the only TKI that is a preferred regimen for poor and intermediate risk first-line patients, a population that makes up approximately 80% of the…

All right. Thanks, P.J. I'll provide an update on our partner programs and also on our ongoing efforts directed toward expanding our development pipeline. I'll start with an update on our partnership with Daiichi Sankyo. We were delighted to announce in January that MINNEBRO, the commercial name for esaxerenone, received regulatory approval in Japan for the treatment of patients with hypertension. MINNEBRO is a potent and selective mineralocorticoid receptor blocker identified during the research collaboration with Daiichi Sankyo from 2006. Exelixis will receive a $20 million milestone payment from Daiichi Sankyo upon first commercial sale of MINNEBRO and we are eligible for additional commercial milestones and low-double digit royalties. The approval was based on positive data from a registrational Phase 3 trial in Japanese patients with essential hypertension. There is a significant need for effective new agents in the treatment of essential hypertension in Japan, given that there are approximately 43 million adults who have high blood pressure. In addition, in the fourth quarter of 2017, Daiichi Sankyo initiated a second Phase 3 trial in Japanese patients with diabetic nephropathy, which builds off the Phase 2b trial previously conducted in this patient population. Switching now to our collaboration with Roche Genentech, the ongoing cobimetinib development program includes two Phase 3 clinical trials in melanoma patients. These are the IMspire150, or TRILOGY, trial, combining cobimetinib with vemurafanib and atezolizumab in previously untreated BRAF mutant-positive, locally advanced metastatic melanoma, and IMspire170, which was initiated in the fourth quarter of 2017 and is studying the combination of cobimetinib and atezolizumab in previously untreated BRAF wild-type metastatic melanoma. Both trials are now fully enrolled and Roche is guiding for potential regulatory filings later this year if the data are supportive. Roche is also sponsoring multiple additional earlier-stage cobimetinib trials, which are currently in progress…

Thank you, Peter. I'm pleased to provide an update on the progress of the cabozantinib development program in the quarter. I will start with a brief regulatory update on second-line HCC, as the fourth quarter was marked by important milestones. First, in November 2018, the European Commission approved cabozantinib as one therapy for the treatment of HCC in adults who have previously been treated with sorafenib. The approval followed a positive recommendation by CHMP for this indication in September 2018. And in late-breaking news, FDA approved our submission for second-line HCC on the date of January 14, 2019, in the United States. These important regulatory approvals expand the indications for cabozantinib beyond advanced RCC into a new indication, HCC, and we are very pleased with this progress that allows us to bring CABOMETYX to more cancer patients in need of new therapies. With these important milestones achieved, we are fully focused on the broader development and lifecycle management plan for cabozantinib, including single agent evaluations and combinations with immune checkpoint inhibitors through our collaborations with BMS and Genentech-Roche. As we have mentioned on prior calls, we have been planning to start multiple pivotal trials with cabozantinib in various tumor types in 2018 and in 2019 and have made progress in our work with such trials as planned. Two Phase 3 studies have already advanced to initiation in Quarter 4 of 2018 and I'm pleased to provide a little more color on these Phase 3 trials today. In October, we announced the initiation of a placebo-controlled Phase 3 trial of cabozantinib for the treatment of advanced radioiodine-refractory differentiated thyroid cancer patients who have received prior VEGFR-targeting therapy. This trial, COSMIC-311, will enroll a total of 300 patients globally. The co-primary endpoints of the trial are objective response rate and progression-free survival.…

All right. Thanks, Gisela. Exelixis had a strong fourth quarter and continues to see solid growth in all aspects of our business as we move into 2019. Our notable regulatory, commercial, and financial performance in 2018 provides a strong platform for rebuilding our product portfolio with future cabozantinib label-enabling trials and adding new product opportunities through internal and external R&D efforts. I would like to emphasize two key issues that are important to keep in focus. First, Exelixis will continue to build on the strength of CABOMETYX and its status as the TKI of choice for RCC. Our commercial efforts target the broad RCC opportunity and we continue to target every eligible RCC patient every single day so that RCC patients can potentially benefit from CABOMETYX at some point in their journey from first- to third-line therapy. We are not surprised by the success of ICI combination strategies in RCC and expect that the vast majority of these patients will need subsequent therapies as they become refractory to front-line ICI-based combinations. Based on market research, KOL feedback, and, most importantly, the real-world trends that we've seen recently, CABOMETYX appears to be the TKI of choice for current and future ICI refractory patients. As we've said previously, we see the eventuality of an evolving first-line RCC dynamic, with multiple ICI combinations as standard of care, as an opportunity for CABOMETYX to fortify its position as the go-to standard of care for second-line ICI refractory patients, which could provide a compelling growth opportunity for years to come. Second, we seek to expand cabozantinib's indications and advance next-generation molecules to build on the potential of this approach. Cabozantinib has demonstrated broad clinical activity in multiple tumor types as a single agent and shows encouraging activity, tolerability, and safety when combined with ICIs. Our…

[Operator Instructions] And your first question comes from the line of Andy Hsieh with William Blair. Please proceed.

First question for P.J. In the field of ATC, as we see more options for patients, just wondering if you have experienced an evolution in patient management dynamics recently. What I mean is do interventional radiologists feel more comfortable sending their patients to their oncologist or hepatologist colleagues?

Certainly, as I mentioned, excited to be approved in HCC. And it's early days for us in the launch and I think as we've kind of indicated previously, this is a market that will have to be built by us and by others, as you kind of indicated, with many recently approved therapies. What we are seeing and hearing from the KOLs is that the patient flow dynamic, if you will, with regards to the interventional radiologists, is sort of starting to flow more quickly into the oncology setting, which isn't really surprising given that there are many more options now beyond sorafenib, which was the only option for a long time. I think the Decision Resources Group data that I referred to in our script quantifies this patient dynamic over time and, in the first-line, it's expected to increase by 37% drug-treated patients by 2025 and in the second-line, it's an even higher number, a 69% increase. And that's being driven by the patient flow dynamics and just the fact that the drug treatment rates are expected to rise. So, very pleased with the indication and happy that it gives us the opportunity to have these multi-tumor discussions with the community oncologists, where we can talk about the great data in HCC as well as RCC.

Thanks, P.J. So, second question has to do in the RCC setting. So, obviously, strong data from Keynote-426, which, in my view, at least, reinforces the potential for the IO/TKI combination. And one of the unique, I guess, chemical aspects of cabo is that it targets XL, which is unique among all the approved TKIs. I was just wondering if, Peter, you could remind us what are some immunological or, I guess, tumor escape mechanism implications have been published out there.

Thanks for the question. I'd be happy to answer that one. And in a broad sense, it's a kind of modern version of a kind of broader question that we've been asking for a while. If we go back to the original development of cabozantinib as a single agent in renal cell carcinoma, I think people were reasonably comfortable with the idea that VEGFR inhibition would result in clinical benefit. And the question always was, for the additional targets that cabozantinib inhibits - you mentioned XL is one, MET obviously is another one - what are they going to add? So, we would usually answer that question around the role of MET and XL in the tumor biology of RCC, it's role in the vascular, and then, more recently, as I'll get to in a minute, the role in tumor immune microenvironment. And I think the combination of that really came with the CABOSUN data, comparing cabozantinib head-to-head with sunitinib, which is a very good VEGFR inhibitor but does not inhibit some of those other targets, that showed that cabozantinib was superior and led to our approval in front-line RCC and now its position as the leading TKI in renal cell. So, now, as you indicated, the question is, in the context of combination with an immune checkpoint inhibitor, what might we expect beyond the VEGFR inhibition. And obviously the 426 data with axitinib and pembro kind of lays a baseline or a benchmark, if you would like. Axitinib is a very selective VEGFR inhibitor what might the role of MET and XL be. And I think there's a pretty compelling scientific rationale and also data with cabozantinib that suggests that those targets might be impactful. So, with XL, since you mentioned it, it's known to be expressed, and its…

Thanks for that really detailed answer. So, last one for me for Chris. I think last quarter, you talked about the inventory on-hand, two and a half weeks. Is there any sort of update in terms of the quantity for this quarter?

I mean, so I'd answer that in a couple of ways here. So, I guess the focus for us is really on the demand growth. Right? As P.J. mentioned, we saw a 6% demand growth in Q4. That was preceded by a 5% demand growth in Q3. And as P.J. pointed out, we are starting to see patients roll off ICI and into CABOMETYX. So, we continue think of this as a demand story. Now, that being said, we saw some fluctuation in the wholesale inventory levels and so we did see weeks on hand rise from around two and a half weeks to around 2.9 weeks. But, again, we do still think this is very much a demand story and not an inventory story. You can refer to Slide 10 in the slide deck because that lays it out for you pretty well there.

And your next question comes from the line of Michael Schmidt with Guggenheim Securities. Please proceed.

Great to see sort of the next IND being active here, XL092. Could you maybe comment a little bit more how that molecule might be similar or different compared to CABOMETYX and how you think about potential development paths, especially if it's similar? Do you see that sort of as a lifecycle management type opportunity or are there maybe other indications that might be interesting based on the mechanism?

Yes, this is Peter. I'll take that one. So, obviously, over the last several years at this point, we've developed an enormous amount of information and understanding of cabozantinib, clinically but as well as pre-clinically. And I think there's been advances in our understanding of the relevance of the various types of cabozantinib to the treatment of cancer, as I just outlined. So, we think there's something special about the kind of VEGFR combination, the kind of core of the cabozantinib profile, if you'd like. So, XL092 retains that, of course. It's a VEGFR inhibitor. It's a novel TKI. It has a potential for a differentiated clinical profile. The Phase 1 trial that Gisela described earlier is obviously aimed at getting us the data to see whether the pre-clinical profile - what we see and that, obviously, we like - actually translates into people. And based on that data, we'll be guided about how and where to further develop the compound.

And then a question on guidance. On your R&D guidance for 2019, I'm just curious if you could share what do you have included there in terms of additional potential cabo Phase 3 trial starts throughout the year.

Michael, it's Chris. So, thanks for the question. I'd say the biggest driver of our expense increase 2019 versus 2018 is really just the ongoing enrollment in 021 and 311 and 312. And there's a very small amount related to new studies that we've budgeted for but we haven't talked about yet. But the majority is around the studies that have already started in either 2017 or 2018.

And then how should we think about additional data disclosure from COSMIC-021, in terms of some of the extension cohorts here. Some of those are fully enrolled for some time now, I believe.

This is Gisela. Thank you for the question. COSMIC-021 is actively enrolling across 20 expansion cohorts. And, certainly, we are very excited about the different indications that we are generating data on. Just to say that we are not going to present or publish premature data. We are looking to accumulate sufficient data that is mature and that will be presented at the appropriate time at conferences and publications. So, we will certainly update you on that. Just referring back to one such presentation, and that was the DOS escalation experience presented at the ESMO conference, and I referred to the data earlier on in the call. We're very pleased with the tolerability profile of the combination. There were no dose-limiting toxicities and we saw good signs of clinical activity in RCC.

And your next question comes from the line of Kennan MacKay with RBC Capital Markets. Please proceed.

This is Justin on for Kennen. Congratulations on the quarter and thanks for taking our questions. A couple of quick ones from us on the increased inventory build that you sort of saw to close out 2018 across the board. Wondering if you're getting any feedback on potentially what drove that? And then going forward into 2019, how you expect that to potentially impact growth in cabo sales moving forward. Thank you.

This is Chris. So, looking at it, we do think that the big portion of that is the future demand that wholesalers would see. But we also saw, generally, over the biopharma industry, several companies talking about their increase in fourth quarter wholesaler inventory. And some of that could be related to the price transparency measures that have been put in place, including SP17 in California where, in certain circumstances, you have to report, in advance, your price increase to the government. So, nothing specific that we can point to but those are some of the things that we think may have driven it.

And your next question comes from the line of Boris Peaker with Cowen. Please proceed.

I'd like to add my congratulations on the excellent quarter. And I just want to probe a little bit further on the liver cancer opportunity. Can you just comment - you mentioned there's an overall growth in HCC. What fraction of this growth is driven by TKI versus checkpoint inhibitors right now in the marketplace?

This is P.J. I'll take that question. As I kind of referred to, it's sort of early days for us in the approval. So, I wouldn't really want to speculate on, certainly, a dynamic market, with new approvals both on the IO front as well as TKI moving forward. What I would say is that I think we're certainly well-positioned. We have a concentrated customer universe, in terms of the value of HCC, 95% of which we already covered with our RCC salesforce. And we know that prescribers, through market research and early anecdotal feedback, if they've used CABOMETYX in RCC, which at this point is a significant number, they're more likely to use it in HCC. So, I think those are all encouraging signs but recognize that it's early in the launch and wouldn't really want to speculate beyond that.

And just a follow-up question, maybe kind of similar in the RCC space. I mean, we obviously saw the Keynote-426 data, which is an IO/TKI combo, and we saw results for Opdivo plus Yervoy. How do you see that dynamic playing out for front-line setting? Do you see maybe Opdivo and Yervoy giving way to more TKI use in the front-line setting or just your general take on this data dynamic?

It's Mike. I'll be happy to provide some color there. as I mentioned a few minutes ago, I think we've always viewed the different either IO or IO/TKI combinations as being potentially very enabling and successful and moving into the first-line setting kind of en masse. So, I think at steady state - obviously, this is all data dependent and we certainly need to see more of the 426 data this coming weekend at ASCO-GU. But it wouldn't surprise us if there's numerous different combinations that take up the bulk of the first-line setting. And as we talked about in our prepared remarks, that's fine by us. It makes sense based upon the data but it also provides, I think, a pretty compelling opportunity for CABOMETYX as the, potentially, go-to standard of care for second-line. And we're seeing some of that data emerge right now, as P.J. mentioned in his prepared remarks, and some of the data from Brand Impact Rx. But it's a dynamic playing field. Lots of moving pieces. We're certainly in that game with 9ER and excited about having what are arguably the two best-known RCC agents, in terms of nivo and cabo, being combined. The only two that separately show overall survival by themselves. And, certainly, then the opportunity to look at the triplet, cabo/nivo/ipi, as well. So, exciting times for us in this indication and we're certainly very committed to continuing to work with our partners and do additional work ourselves to continue to move the goal post as high as we can going forward.

Your next question comes from the line of Ted Tenthoff with Piper Jaffray. Your line is now open.

So, two questions for me. Firstly, on the COSMIC-311, I think you said that you were going to provide an overall response update on the first 100 patients. Would that be potentially registrational?

So, you heard me say that objective response rate and durability of response and PSF are co-primary endpoints. And that the study foresees assessing the objective response rate in the first 100 enrolled patients with sufficient follow-up. And any regulatory steps would be, obviously, data dependent. But as a co-primary endpoint, we feel positive that could potentially support such action.

And then with respect to some of the other updates from Cosmic-021, when could we expect data coming in from some of those other cohorts? Thanks for taking the questions and congrats on a great quarter.

As I said earlier, we definitely want to present mature and stable data from the expansion cohorts and we will do so as soon as we have those data with sufficient follow-up. And so we will certainly keep everyone in the loop and communicate when such data are available and when they will be presented.

Your next question comes from the line of Peter Lawson with SunTrust. Your line is now open.

Thanks for taking the questions. Mike, just what are the puts and takes we should be thinking about 9ER and how it could stack up versus 426? Anything we should be thinking about? And any color around the timing of 9ER?

So, look, I don't want to speculate too much on an ongoing pivotal trial. It's just bad form. Probably not something we should be doing. That being said, in my answer to Boris, I think the idea of combining two agents which improve overall survival together in this indication, which is obviously sensitive to both, is very attractive and we're doing that with 9ER with cabo and nivo and we're looking forward to doing that with the triplet as well, with cabo/nivo/ipi. So, again, we have lots of interest in this space, a lot of momentum in this space. We feel like we understand not only the clinical pharmacology and the clinical activity of these agents, as you heard from Gisela today, but as Peter mentioned as well, we really have a fundamental understanding of how cabo is working and interacting in the tumor immune microenvironment. So, we've got a lot of things going for us and we have to just get it done now and get the data and then move forward with that data in hand.

And what kind of feedback are you getting from physicians around the use of combinations versus sequencing with individual TKIs and IOs? What's the preference?

P.J., you want to take that?

Yes. I think we talked about it a little bit in my prepared remarks. What we've seen, in terms of the landscape, is that sequencing still plays a large role. We do think that, to an extent, single agent TKI will maintain some position in the first-line. And as Mike kind of alluded to, certainly, the combinations, we've always thought would be big players in the first-line. But, importantly, looking at all the market research we've done, regardless of whether it's nivo/ipi or PD-1 TKI, really the indication is that cabo will be the therapy of choice in the second-line there. And that's certainly been borne out so far by the data we've seen. In Q4, as I mentioned, approximately 90% of the patients progressing on nivo/ipi went on to receive cabo. So, we're pleased with that and that's really a key part of our messaging. It aligns with our positioning in the NCCN guidelines and really the data in our label and the depth and breadth of both the METEOR and CABOSUN studies. So, we think the patient flow dynamics of those patients coming off the combination nivo/ipi, those patients will continue to increase in second-line and I think we're well-positioned to capture those patients.

And your next question comes from the line of Silvan Turkcan with Oppenheimer. Please proceed.

Congrats on the quarter and thank you for taking my question. Could you just tell me a little bit about maybe some details, if you can, about your BD strategy? I mean, you have $850 million in cash and are you looking for an asset that kind of combines well with TKI? Are you looking for another TKI? I mean, you just put one into the clinic. Could you give us a little color there, please?

It's Mike. I'm happy to say a few words there. So, look, we have a very broad approach here, looking for appropriately valued assets that span the range of opportunities from late pre-clinical to early clinical to late clinical to even commercial. So, we have a pretty big appetite and I think a pretty good vision on where we could go with any of those kinds of assets. We did a number of early stage deals last year, while looking, certainly, much more broadly and much more deeply at a wide range of assets. So, we're looking for the right compounds. We're agnostic to the modality. Certainly, we have interest in biologics as well and our deal with Invenra, I think, underscores that. But looking for the right kinds of products, the right kinds of drugs that have good data that would allow us to either develop them or pursue them in some strategy that would allow us to move them into the commercial setting as quickly and as efficiently as possible. So, I would think broad, I would think deep, and as we, again, get traction on the next wave of deals, we'll be talking much more about that in great detail.

And maybe could you also tell us a little bit about your internal estimates, in terms of size and speed to peak of the HCC opportunity with respect to RCC? And since there's a lot of overlap in KOLs and prescribers, is there a way to maybe get some margin expansion as the sales ramp up in HCC?

Well, as we talked about previously, we were able to - and as P.J. just alluded to a few minutes ago - the fact that there's a good overlap between the HCC market and the RCC market has allowed us to use our existing commercial footprint, with a few enhancements, to be able to address both simultaneously. So, the idea is that we don't need to invest a lot more to be able to really address this large underserved market opportunity with second-line HCC. That being said, I think we've been very consistent along the lines that this is a market that we have to help build, or we build, from the standpoint that with all the new therapies coming into the marketplace now on the TKI side, on the IO side, that we need to help kind of educate not only oncologists but also hepatologists around the value that some of these drugs can bring and then help move them move their patients, again, toward the medical oncology sleeve. That could be very productive for them and for us. So, this is going to be a slower ramp. We've been very consistent in talking about that. We haven't given guidance on HCC since there's many, many moving pieces that are certainly hard to model and hard to predict, both by us and by others. But that's fine. I think we're certainly very committed in terms of where we stand today commercially and we were ready to go within, literally, minutes of getting the approval letter for second-line HCC a few weeks ago. But, also, we are continuing to reinvest in this indication as well. The COSMIC-312 study, again, looking at cabo/atezo in the first-line HCC setting, is just clearly a sign of our interest, the value of the market, and the large unmet medical need that these patients actually have. So, a lot of work to do but we're certainly excited to be in the mix here and looking forward to pushing it forward.

Your next question comes from the line of Stephen Willey with Stifel. Your line is now open.

Thanks for taking the question and congratulations on the quarter. Maybe just a question for P.J. to start. Do you happen to see any kind of noticeable difference, just in terms of duration of therapy, for those patients who are progressing off of the ipi/nivo regimen versus those patients who have progressed off of a prior TKI?

With regards to duration, I think relatively early in the patient flow dynamic to really make any comment on that specific patient population. We've always said that the duration we're seeing is similar to what we've seen in the trials. So, something we certainly track very carefully and we're encouraged to see many - 90% - of the second-line nivo/ipi progressors going onto CABOMETYX and, certainly, some data and metrics that we look forward to tracking in the future.

And then just with respect to that 90% of patients who are progressing, I guess, on ipi/nivo. So, the updated CheckMate 214 abstract that was released yesterday I thought had an interesting data point in it, in the sense that I think it suggested that only 55% of patients in that trial actually went on to receive systemic second-line therapy. I know there's still another 10% or 15% of patients that are still on study drug. But just curious if you think this dynamic is kind of specific to the trial environment that is CheckMate 214 or do you think that some of these novel front-line regimens might be somehow kind of changing the proportion of patients we subsequently see on second-line therapy? Thanks.

Steve, it's Mike. I'll take that one. It's probably a little early to comment there. We saw that data, too, and it certainly looks interesting. And we asked the same question that you asked there. Is that a consequence of having it be at clinical trial sites as opposed to the community, etc.? So, we're certainly pleased to see NRx trends improve, relative to what's happening in Q1 and at the end of Q4, relative to what we think is patients coming on cabo after progression with ipi/nivo. So, that's all internally consistent with the timing of that launch and the PFS data that we saw with 214. So, great question. Early days. Certainly, we'll be able to track more of that as time goes on and we get more data, more patients, just more data and numbers. Right? So, stay tuned.

Your next question comes from the line of Paul Choi with Goldman Sachs. Please proceed.

I was wondering, either Mike or P.J., if you could maybe comment on how you think about the volume-price dynamic over, let's say, 2019 and 2020 in RCC, given the changing landscape. Correct me if I'm wrong, but I think, by my math, volume contributed approximately 20% of your growth for cabo in 2018. So, as you think about this short- to intermediate-term, how do you think about maybe volume versus the price growth? And can the price increases that you've taken annually, is that sort of a reasonably proxy going forward here?

Thanks for the question. Fair question. You're vectoring toward a question around guidance or at least data that would drive that math and I'm not sure we're comfortable sharing that with you. We've certainly grown and continued to grow in quarters like Q4 and Q3 of 2018 in the face of pretty significant competitive pressure. In fact, we were the only TKI that was growing in that timeframe, both 6% in Q4, 5% in Q3. We had a big quarter in Q1, obviously, when we launched based upon the first-line approval in 2018. So, I guess the question you're asking is certainly a key issue. Again, what we're looking at is the dynamic around what's happening in first-line and the impact of what happens in first-line and the consequence it has or the impact it has on the second-line. So, time will tell. I think our theory, as we've outlined several times today and provided some supporting data, is that we will continue, hopefully, to grow in the second-line setting and that volume pick-up will drive growth on top of everything else that we're doing in that regard. So, many moving pieces. We acknowledge that. It's a complex dynamic but it's one that we've got, again, great data, great team, and a lot of momentum going into 2019 to be able to cover that.

And maybe one for Chris on the guidance on the R&D component. I guess with the CheckMate 9ER coming toward the finish line here in the not too distant future, can you maybe explain to us is it primarily the COSMIC-311 and -312 components of the R&D expense that are driving the year-over-year increase? And, also, apart from that, will there be any completion of enrollment timeframes that you're willing to share for either trial?

So, I'll take the financial piece of that. So, I would say that 9ER is a significant contributor to the expenses in 2019. And it just doesn't stop when data - the trial doesn't stop when data is read out or revealed. So, that will continue beyond any point of readout. And then 311 and 312 will be a big contributor from an incremental perspective, just because they were started later during 2018 and so the incremental piece will be more significant, I guess, on 311 and 312 when you compare '18 versus '19.

In terms of the accrual, at this time, we're early in these trials and we don't provide guidance on the accrual dynamics at this point. I think I can refer you back to what we said earlier regarding the 9ER study, where Bristol-Myers in their quarterly call indicated that results could be expected in the second half of 2019.

Your next question comes from the line of Jeff Hung with Morgan Stanley. Please proceed.

Thanks for taking the questions. I just wanted to clarify. The $3.3 million inventory build, that was on top of the leftover inventory build from 3Q, correct?

I mean, that's the way to look at it, yes. So, as I said in my prepared remarks, we did see an approximately 700 unit increase in inventory during Q3 and Q4 in total. And as I said to Andy earlier, we did see, in Q4, weeks on hand go up from two and a half to 2.9 weeks. But the real story here is the demand story. As P.J. has said, I said earlier, Mike said, this is 6% demand growth in Q4, 5% demand growth in Q3. And, overall, in 2018, when you compare it to 2017, I mean, demand grew almost 66%. So, it is a demand story. We provide the inventory information so there's clarify on where the growth comes from and transparency but I'd really focus it on the demand story.

And then you guys talked about BD and potential opportunities you guys have been looking at. But I guess in terms of capital allocation, how do you think about balancing profitability and acquiring additional assets? Thanks.

Jeff, it's Mike. Yes, we've spoken of that extensively. There's both strategic and tactical reasons to be profitable and to stay profitable and that's certainly an important part of our equation over the last few years and certainly going forward. But we certainly have a lot of room to maneuver, in regards to the financial depth that we've built over the last few years, to be able to add assets that we think have value and can generate value going forward. So, it's certainly a balancing act but it's one that I think we've got the experience, the depth, both clinically from an R&D point of view, commercially, as well as financially, to be able to manage pretty well.

Your next comes from the line of Asthika Goonewardene with Bloomberg. Please proceed.

Thanks for taking my questions and for squeezing me in at the end. First question. On cabo, you have the first-line RCC approved for more than a year now. I was wondering, with all the real-world data at hand, does the real-world data for PFS, is that consistent with what's on the label, the 8.6 on the label? And then I've got a couple more.

Yes, this is P.J. As I mentioned a bit earlier, I think the duration is consistent with what we've seen in the trial. So, it's certainly encouraging and gratifying that the patients are getting the benefit that we saw in the data.

Excellent. And then so far we've seen some first-line RCC combo data with Keytruda, Bavencio, nivo/ipi, etc. And with it, some certain comparative arms. I was wondering if you could tell me, given what you know about the trial recruitment sites for these trials, as well as 9ER, you could maybe tell us which one of them might be a better representation of 9ER's arm.

It's Mike. Fair question. I think it's one that it is probably not wise for us to comment on. It's always challenging and arguably somewhat dangerous to make cross-trial comparisons, either retrospectively or prospectively. So, we don't really do that and would not want to give any insight or guidance there for you. Sorry.

Last one. On XL092, we've been trying to dig up the patent for this to see if we can get a little bit more color as to what exactly this is targeting beyond MET and VEGFR and we think we might have found it. It's the one ending in 342 which hits MET, VEGF2, CKET, XL, etc. And also has some associated use patents in RCC and HCC. I have to ask is this the right one? And then could you maybe contrast 092 versus cabo in terms of potency on the targets, please?

It's Mike. So, I can answer that question with an emphatic no. That's not it. The patent application hasn't published yet so I would stop looking, at least in the short-term. In terms of the pre-clinical profile, again, we'll defer that to a later date once we have some additional clinical momentum that would allow us to kind of put the full story in context. As you could imagine, this is a pretty competitive space right now and there's no need for us to enable the competition with any early looks at the data. But, again, we're excited about having discovery back in action. I mean, that's in our DNA, going back almost 20 years. So, it's very gratifying to have Peter and crew back in the lab cranking out new molecules and moving them forward into Gisela's world. And, certainly, again, as both Peter and I talked about today, we have just tons of information and knowledge from cabo that we're able to put into a next-generation, potentially better molecule. And we're going to get some clinical data to see if our hypotheses are correct. So, stay tuned.

Thanks for taking the question, guys, and look forward to more assets coming out of the clinic this year.

Your next question comes from the line of Tim Call with The Capital Management. Please proceed.

Congratulations on a great quarter again. You have zero debt. If you find the right acquisition, what would be the optimal debt level for the company?

Tim, it's Mike. Thanks for the question. Glad you could join us today. Yes, it's a question that we probably don't want to speculate on, in terms of how we might tactically put our capital to work, either cash that we have or debt that we might access. We certainly have done debt in the past. We had a convert that we used to help finance the company going back to 2012 so we have some experience there. Chris and his expertise, his time back at Gilead, did a lot of debt work. So, we have the depth of knowledge and experience, from a capital perspective, to be able to, I think, really ask the right questions, do the right analysis, and then deploy the right level of capital in the right way to help us maximize value for the company. But it would be probably unwise to speculate on that prospectively.

Because of your excessive free cash flow, your cash balances rose by 86% this year to around $2.73 per share. Should some of that be returned to shareholders to offset future stock option dilution by buying back shares?

Yes, that's a good question. Chris, you want to start with that and I'll provide some commentary.

Sure. I mean, at this point in our life as a company, to us, the most important driver of shareholder value is really around the deployment of capital in finding new assets and less so on a shareholder return around a buyback or dividend. We think it's more important to really drive shareholder growth through the acquisition of assets and deploying that in BD and acquisitions and licensing. And that is the most important piece of what we're doing today. Now, I'll never say never. I won't say that we won't do a share buyback or a dividend in the future but that is less of a priority than actually driving company growth through acquisitions or through BD and licensing.

Yes, I would just reaffirm that. Look, we're focused on long-term value creation and sustainable growth over the long-term. And I think, as a company and where we've come from and where we're going, that seems to be the best way to build value. But thanks for the question.

Many of your partner contracts have embedded very attractive escalation clauses. When you reach certain cumulative product sales and reach those thresholds or exceed those thresholds, your percentage of future product revenue received as royalties or collaboration revenue goes up nicely. Did you reach any of those thresholds last year and do you expect to reach any in 2019?

So, yes, we did reach, I guess, two thresholds last year, to talk about that. So, the first threshold was around the cumulative $150 million of revenue. And the first $100 million of that $150 million was at a 2% royalty rate and the next $50 million was at 12%. After we got through that cumulative piece, we hit our now minimum royalty of 22%. So, on an annual basis, our minimum royalty will be 22% in 2019 and beyond from the Ipsen perspective. And then the other one is when Ipsen hit a cumulative $100 million over four quarters, we did receive a milestone or earned that milestone during the second quarter, which we recorded during the second quarter. On a future basis, I'm not going to speculate on the future potential for them to hit the next set of commercial milestones but just know that the royalty rate is, at a minimum now, from Ipsen, 22%.

Thank you. And at this time, there are no further questions. So, with that, I will turn the call.,.

Great. Thanks, Andrew, and thank you all for joining us today. We certainly welcome your follow-up calls with any questions we were unable to get to during today's call.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program and you may all disconnect. Everyone, have a wonderful day.