Good day, everyone, and welcome to today's Bristol-Myers Squibb 2017 first quarter results conference call. Today's conference is being recorded. And at this time, I like to turn the conference over to Mr. John Elicker, Senior Vice President, Public Affairs and Investor Relations. Please go ahead, sir.

Thanks, Melody, and good morning, everybody. Thanks for joining us to review our first quarter results. With me this morning are Giovanni Caforio, our Chief Executive Officer; Charlie Bancroft, our Chief Financial Officer; Tom Lynch, our new Chief Scientific Officer; and Murdo Gordon, our Chief Commercial Officer. Giovanni, Tom, and Charlie will have prepared remarks, and then Murdo will certainly be available for Q&A as well. Before we get started and I turn it over to Giovanni, I'll take care of the Safe Harbor language. During the call, we'll make statements about the company's future plans and prospects that constitute forward-looking statements. Actual results may differ materially from those indicated by these forward-looking statements as a result of various important factors, including those discussed in the company's SEC filings. These forward-looking statements represent our estimates as of today and should not be relied upon as representing our estimates as of any future date. We specifically disclaim any obligation to update forward-looking statements even if our estimates change. We will also focus our comments on our non-GAAP financial measures, which are adjusted to exclude certain specified items. Reconciliations of these non-GAAP financial measures to the most comparable GAAP measures are available at our website. Giovanni?

Thanks, John. Good morning, everyone I think it's fair to say we just had an outstanding quarter with strong performance across the company. Our non-GAAP EPS was $0.84, up $0.10 over the first quarter of last year. We grew our revenue by 12%, with double-digit growth across our key growth drivers. Starting with I-O, sales for Opdivo were strong despite a more competitive landscape in the U.S. for lung cancer. In the U.S., while we have seen some share loss in lung cancer, our share has shown signs of stabilization in recent weeks. Outside lung, we've seen growth in renal, melanoma, head and neck, and bladder. Internationally, we've made great progress on access and reimbursement. Sales were strong across all indications. From a regulatory perspective, we continue to make progress, mainly the U.S. Opdivo approval in bladder, the Japanese approval in head and neck, and the CHMP positive recommendations for bladder and head and neck in Europe. We shared important new data from our immuno-oncology portfolio at AACR [American Association of Cancer Research], including overall survival data from Study 67, the TNB analysis from Study 26, and the early data from our IDO inhibitor. We also advanced our diversified portfolio with encouraging data from our FGF21 for NASH at EASL [European Association for the Study of the Liver], and we will work with the regulatory authorities to initiate a registrational program as soon as possible. We made significant progress from a business development perspective, investing in translational capabilities through deals like GRAIL and Foundation Medicine and broadening our immuno-oncology clinical development program through agreements to study combination approaches with different mechanisms across a range of tumor. As I look at the company going forward, I see tremendous opportunity, driven by our industry-leading portfolio and the very best people in the…

Thank you, Giovanni. Good morning, everyone. I'm truly delighted to be joining my first earnings call as Chief Scientific Officer of Bristol-Myers Squibb. As Giovanni mentioned, there is great potential in our pipeline. I knew this from my role on the Board of Directors, and it was the major reason I decided to join the management team. Our breath, and particularly our depth, in I-O position Bristol-Myers Squibb to make a major difference in how common cancers are treated and even allow us to begin to think about the possibility of cure for many types of cancer. In the short time that I've been in this role, just over six weeks, I've become even more enthusiastic about the opportunity with our pipeline and our people, both within oncology and across our cardiovascular, immuno-science and fibrosis assets. As I meet with my discovery and development teams here in Princeton, in Redwood City and across the globe, I'm struck that BMS is a company with remarkable R&D talent, particularly in antibody technology, medicinal chemistry, clinical development and in our regulatory operations. This, coupled with the strong integration between R&D and commercial, makes me very optimistic about our ability to execute and to continue to deliver transformational medicines to patients. As a leader, I have always believed that culture matters. At MGH and at Yale, I worked hard to build a patient-focused research environment. Ever since I joined BMS as a director, I knew the company had a patient-centered focus. But as I've been meeting more and more people in this role, I've been struck by just how patient-focused the mission is here. In every meeting, we hear the patient voice. Our research teams are driven to improve the lives of people with cancer, heart disease, fibrosis and rheumatologic disease. There are some…

Thank you, Tom. Good morning, everyone. As Giovanni described, we delivered very strong financial results for the quarter due to robust performance across our portfolio. Let me provide some color on the performance of our key products. I'll start with Eliquis, which for the first time sold over $1 billion in a single quarter. In the U.S., Eliquis is now the leading agent in the NOAC class, which continues to grow as warfarin use declines. During the quarter, sales in the U.S. were driven by a sequential 8% increase in TRx and from the positive impact of lower Medicare coverage gaps compared to Q4 of last year. Internationally, Eliquis delivered strong growth of 51% compared to the same time last year, as the gains in established position as the standard of care in stroke reduction for AFib around the world. Sales for Opdivo in the U.S. increased 6% sequentially, driven by the demand growth that Giovanni described and some favorable inventory movement compared to Q4. We continue to expect our U.S. lung business to be under pressure this year assuming KEYNOTE-021G is approved next month. However, given the stability in second-line lung in Q1 and strong trends we've seen across other indications in our U.S. business, we now believe that there is potential for Opdivo to grow in the U.S. this year compared to 2016. During Q1, we saw sequential growth in international Opdivo sales when adjusting for the approximately $250 million revenue recognition catch-up for France and Germany we booked in Q4. We successfully negotiated reimbursement for Opdivo in several markets during the quarter, including a pan-tumor agreement in Canada as well as agreements for renal cell cancer and non-squamous-cell lung cancer in Italy. Based on positive sales trends and reimbursement successes, we continue to expect international sales for…

Okay. Thanks, Charlie. So we're ready to go to questions, Melody. We can go ahead and start.

Thank you. And we'll go to our first caller.

I think the operator is not going to announce the callers. So once you get prompted, if you can, go ahead and introduce yourself.

And, caller, your line is open. Please check your mute function. With no response, we'll move to the next caller. Next caller, please go ahead.

Hi there. It's Chris Schott at JPMorgan. Can you guys hear me?

We got you fine, Chris. Thanks.

Okay, perfect, just two questions here, maybe the first one for Giovanni and for Tom, just a question on lung and this ongoing debate around I-O/I-O versus I-O/chemo combinations. I think one of your competitors today announced that they're also going to be looking at chemo combos with PD-1 in addition to their I-O/I-O combos. I'm just interested in your latest thinking on the role you see chemo playing in this I-O landscape relative to CTLA-4 and some of the other next-generation agents you're developing. My second question was specifically on IDO and just more thoughts there on IDO as a combination agent and your approach to this target as you balance your internal asset versus the external partnership. So just any more color there would be helpful; and maybe as your answering that, just latest timelines on when we could potentially think about that internal asset moving forward into registrational studies. Thanks so much.

Thank you, Chris. This is Giovanni. Let me just start, and I'll ask Tom to comment on both of your questions. On lung strategy, as you heard from me and Tom, we are very comfortable with the fact that we have a very broad approach to first-line lung cancer, a very broad program. We are indeed investigating various treatment modalities, including I-O/I-O, I-O/chemo, and as you know, potentially the combination of the two approaches with I-O/I-O coupled with chemotherapy. And I am pleased that over just a few months, we have been able to evolve our lung cancer program, taking advantage of ongoing studies and that enrollment has progressed I would say very much in line with our expectations. With respect to IDO, the one thing I would say at a very high strategic level is that we see really promising modalities. I think the ability to complement external collaborations as we advance internal programs provide, again, in the same spirit, a very broad set of options for us in order to be competitive from a timing perspective at the same time as we develop potentially best-in-class. But let me ask Tom to give you his comments on both.

Chris, thank you for your questions. And I just want to start by saying lung cancer is a really hard disease to treat. And you don't need to go any further than this morning's New England Journal, the paper by Charlie Swanton's group, looking at the genomic diversity of this disease, both within tumors and between tumors. It's a tough disease to treat. And we believe it's important for us to have multiple approaches to treating patients with non-small-cell lung cancer. We think the approach at Bristol-Myers Squibb that includes I-O/I-O and I-O/chemo and I-O/targeted drugs gives us a broad and diverse and balanced approach to how we're going to go after lung cancer. And I think that there may well be certain patients that are going to benefit from I-O monotherapy, I-O/I-O, and I-O/chemotherapy. I don't think there will be a one-size-fits-all approach to this disease. I want to comment on CTLA-4 and how important a target this is in lung cancer and other cancers as well. CTLA-4 I think has been validated as an important partner with Opdivo. Just look at our data in melanoma and the 067 study, which showed that we actually have a survival advantage, the only I-O/I-O combination that's been shown to have a survival advantage. And we think CTLA-4 matters in this disease and requires investigation. I think just to show you how deep our commitment to CTLA-4 is as an agent, we have two additional compounds with CTLA-4 that we're developing as targets. The first is the non-topopulated (25:55) CTLA-4, which has now entered into clinical study. There's a trial that we describe on ClinicalTrials.gov that you can take a look at. And our relationship with CytomX has developed the CTLA-4 compound as a Probody that we think might have a different…

Thanks, Chris, for the question. Melody, can we go back to the first group in the queue? I got an email from the team at Barclays saying that they were having trouble recognizing that they were up. So I don't know if it's Geoff Meacham or Paul Choi, but could you go to them please again?

Yes, definitely. Just one moment. And, Mr. Meacham, your line is now open.

Okay, great. So thanks a lot for the question. It seems the biggest opportunity longer term in I-O is really the cold tumors, breast, ovarian, colon, et cetera. To follow up on the earlier comment, CTLA-4 combos, do you guys feel, at this point, CTLA-4 combos plus a PD-L1 is going to be a component of that? Are there other targets across the I-O landscape and many different targets that look more viable in cold tumors versus hot? Thank you.

Geoff, thanks for your question. I think this echoes what I mentioned earlier, the concept of looking at resistance. Resistance is incredibly important, and it's really characterized a lot of oncology drug development. Just looking at where your drugs don't work tells you as much as you learn from looking where your drugs do work. What I explained earlier and outlined is the two areas I'm most worried about in terms of resistance – or think of as actually opportunities is what about the cold tumors that don't respond, and then what about the patients who have initial response to anti-PD-1 or anti-CTLA-4 therapy who then progress? Now, as you know, we have 10 targets in I-O that we're working on and 12 agents in those 10 targets. So we think we have multiple opportunities to look at combinations of I-O/I-O agents in these cold tumors. We presented data with our IDO asset, as you mentioned, and we also will be presenting initial data at ASCO this year with our GITR compound, which could be the type of agent that might play a role in the cold tumor space. But we will increasingly look at combinations of agents in that setting that might make a difference.

Thanks, Geoff. Melody, can we go to the next question? And if you have it in front of you, could you announce them on the line so that they know that they're ready to go? Thank you.

We'll go next to Gregg Gilbert with Deutsche Bank.

Thank you for that double prompt. Maybe, Tom, here's an opportunity for you to straighten us out of the timing of 227, whether it's final or interim, that would be helpful. And curious, Tom, if you have a view on whether PD-1s and PD-L1s can be different from one another other than by trial design strategy. And one for Giovanni real quick. In light of the activist investor interest and significant change in the board, curious if you can comment on what's changed in terms of how you're spending your time, the company strategy, options you're exploring to enhance value. Or do you see this as business as usual stuff? Thanks.

Gregg, thank you for your questions. I'll start off by commenting on the timing of 227. As we have published on ClinicalTrials.gov, we expect the data from 227 to report in the first half of 2018. And as we also have announced, there will be an interim look. And we'll look at that data with our data safety monitoring committee as the data matures. But we expect the data to be the first half of 2018. So the second question comes down to differences and similarities between PD-1 drugs. And I think my feeling is that I think there's highly likely to be much more similar between these drugs than there are going to be differences. Now we know, for example, that pembro [pembrolizumab] and nivo [nivolumab] recognize slightly different epitopes and there are slight differences in these compounds. But clinically when you look at the totality of the data, they really behave fairly similarly to each other. So I actually don't think we're going to find enormous differences between the various different PD-1 drugs. Now the difference between PD-1 and PD-L1 agents, certainly there's a possibility that there will be more information. We're looking out on that, and I remain open to looking at the data as it emerges across that area. Giovanni?

Gregg, thank you. So from my perspective, let me say that I'm really very happy with the ability that we have to have broadened and strengthened the capabilities on the board. That's clearly very important for a company like ours. With respect to what our priorities are, they really have not changed. We are, as you have seen today, extremely focused on continuing to execute as well as we have for the short term. And we are extremely focused and I am extremely focused on delivering the value of an extraordinarily promising pipeline, which is probably the broadest and the deepest the company has had. So that's where we are spending our time. Those are our priorities, and clearly they are mine as well.

Thanks, Gregg. Melody, can we go to the next question, please?

Certainly, we'll go to Tony Butler with Guggenheim Securities.

Yes, good morning. Thank you very much, two brief questions. One, if I may around Eliquis, tremendous number for the quarter, but the COMPASS data will read out later this year at European Heart [ESC]. And I'm curious if it does possibly have an effect on the NOAC class as a whole. Or could it be viewed somewhat negatively for Eliquis at least in CAD/PAD? And that is does the class actually increase, or is this just dependent upon – or do we single out rivaroxaban? And then secondly, Charlie, an excellent gross margin quarter, but the question really is, how should we think about that going forward in the year? They bounce around little bit, at least they had last year. This is clearly a change from what we've seen in the last two years. Thanks for the time.

Thanks, Tony. It's Murdo here. I'll address the Eliquis question and pass it over to Charlie for your second part. First off, we're really pleased with the Eliquis performance in the quarter and continuing what was a great year last year. As Giovanni mentioned, total prescriptions, now we're in a leadership position in the NOAC class. And we continue to feel good about growth based on leading indicators that we look at in the market, which are NBRx trends as well as our cardiology trends. We're also seeing strong performance across the rest of the world. Now as I think about the COMPASS trial, it's actually an interesting question. And obviously, without data, it's hard to speculate. What I do see happening though is a continued strength of our profile and the divided daily dose of Eliquis in atrial fibrillation and VTE. What we need to wait to see is the quality of the COMPASS data at a different dose that Xarelto used in that trial versus the dose that they currently use in VTE and in atrial fibrillation. So it's hard to compare the COMPASS data to any of the data that exist already in the marketplace. And then the other question remains is what's the utility of a NOAC versus antiplatelet drugs in the patient population that was studied in the COMPASS trial? So I probably have some more questions related to that trial than I have answers, but we do not see it impacting our AF and VTE business going forward.

Hey, Tony, it's Charlie. Let me address your margin question. As you know, our gross margin is heavily influenced by product mix, with Eliquis, given the structure of our relationship with Pfizer, having a lower than 50% margin. So the growth of Eliquis with the loss of our higher-margin virology business has put pressure on our gross margin. So the uptick that we saw in the first quarter is almost solely due to foreign exchange when we think about it exiting out of the fourth quarter of last year, and that's more of a one-time impact that we see in the first quarter and not really repeating for the balance of the year.

Thanks, Tony. Melody, can we go to the next question, please?

Certainly. We'll go to Seamus Fernandez with Leerink.

Thanks very much. Can you hear me okay?

Yes, Seamus, we've got you.

All right, great, so a couple of quick questions. Can you guys talk a little bit about how in the wake of the settlement with Merck, how you plan to prosecute your rights around the patent estate that you've built up around PD-1? Obviously, it was successful with Merck, but they had a different time-to-market and there are many, many other PD-1s out there. So just wondering how aggressive you feel that you can be as a company, or if you think that that's not a value-added strategy. And the second question really is, I look at the trends on Eliquis and the still overall market opportunity with the substantial amount of warfarin out there. This feels like it could be well more than a $10 billion category. In fact, Eliquis may have an awful lot of runway ahead of it, particularly given your opportunity to extend the patent. So I just wanted to get a better sense of how much incremental opportunity you think there really is to grow the category in the U.S. and internationally. Thanks so much.

Thanks, Seamus. This is Giovanni. So let me start, and then I'll ask Murdo to give you his comments on Eliquis. So first on the settlement with Merck, we think that is an extremely important recognition the strength of our IP estate with respect to immuno-oncology and PD-1s. And I think that settlement is critically important, but also it reinforces our perspective that we will continue to vigorously defend our IP across the board, and that clearly is a priority for the company. It has not changed. If anything, it actually has become stronger post the settlement with Merck. Let me just also say that we also are very focused on Eliquis, and we agree with the significant long-term potential there is in this market. And I'll ask Murdo to comment on that. Obviously, before I do that, let me just get back to the IP. It's clearly an important priority for us. It's also one that we share with our partner, Ono, and we are working very closely with Ono on any IP-related issue. Murdo, on Eliquis?

Thanks, Giovanni. Thanks, Seamus. We're also, as Giovanni said, we're very enthusiastic about the trajectory of Eliquis so far, and we're working closely with our partners at Pfizer to capitalize on the untapped opportunity that still exists in the market. As you mentioned, over half of the TRx business is still a warfarin-based treatment patient population. And if you look at the leading indicators, new-to-brand shares, warfarin is about 27% of the OAC [Oral Anti-Coagulant], so still a good delta to capitalize on in terms of leading indicators. We continue to invest appropriately in Eliquis. The other thing I would mention is, outside of the U.S., in some markets around the world, the NOAC class is even more underpenetrated in the market. So when you combine the U.S. opportunity and the European opportunities, we continue to feel that there's a large growth trajectory and a high ceiling for that brand.

Thanks, Seamus. Melody, could we go to the next question, please?

Certainly. We'll go to Tim Anderson with Bernstein. Timothy Minton Anderson - Sanford C. Bernstein & Co. LLC: Thank you. A couple of questions. If I look at the totality of things that have happened over the last call it six months, including programs you've advanced like IDO and chemo combo and backup CTLA-4s and other things like that, a bear's point of view could say that this foretells problems with Yervoy and Opdivo, specifically in front-line lung. So my question is can you say you're as confident as ever in the possibility that CheckMate-227 is going to deliver the goods in this regard in front-line lung, or is it fair to say that Bristol's internal confidence in this is less than what it was perhaps a year ago? And then the second question is still on CheckMate-227. There have been some rumblings about additional changes potentially to the construct of that trial above and beyond what you talked about earlier. I'm wondering if you can just clarify whether any of this is correct since the latest updates that we've had. And this includes things like expanding various arms to new subgroups and timelines for readouts and sizing up the trial and that sort of thing.

Tim, thank you. This is Giovanni. Let me just start. So obviously, a lot has happened in the last 6 to 12 months in first-line lung cancer. As I said earlier and Tom reiterated, this is a very difficult disease to treat. It has evolved very rapidly. You just think of the data that was released in the last 12 months, and it's clear that our competitive position has changed and continues to evolve. My perspective is that today we have the broadest program for first-line development for our immuno-oncology portfolio in lung, and we are extremely focused on executing that program. Execution is on track, and I'll let Tom answer you with respect to study design.

And, Tim, thanks for your question. I think a couple things to say. I think the first is that again, just emphasizing how complex and difficult a disease lung cancer is to treat, it's highly unlikely that there's going to be one approach that's going to fit every patient who presents in first-line. We believe strongly that Yervoy/Opdivo is certainly one of the options that needs to be pursued and developed, and we're looking at that in first-line lung cancer. And we're also looking at it in the setting of what appropriate biomarkers might be important in evaluating the potential for this combination and for really all of our drugs in first-line lung cancer. So I think that's important to emphasize. The second question you asked was about changes to CheckMate-227, and there have been no changes to CheckMate-227. The design is the same design as it was the last time it was discussed.

Thanks for the questions, Tim. Can we go to the next question please, Melody?

Yes, we'll go to Jami Rubin with Goldman Sachs. Jami Rubin - Goldman Sachs & Co.: Thank you, a couple questions, and maybe, Murdo, you can answer this first. CheckMate-067 at AACR generated more questions about the combo than answers. Do you see the results leading to an increase in market share for the combo in melanoma or a decrease? And following up on data that came out at AACR, there was an interesting study that showed high tumor burden as a marker of success for PD-L1. I just was wondering if you were able to adjust CheckMate-227 to account for this new learning, if at all. And then lastly for you, Tom, and I appreciate your introductory commentary on your priorities. That's very helpful. Just wondering though, you have been on the board for two, three years, and I appreciate you've only been in this position for a couple months, but just curious to know what sort of changes you have thought about putting in place. What are you planning to do differently compared to the former head of R&D? For example, are there plans to change the way that you go about study designs? I think initially Bristol's focus had been on creating high-value trials showing overall survival. Those trials certainly make a lot of sense, but not when your competitors are trying to beat you to the market. So I'm just wondering if you are changing the way you approach some of these clinical trial designs. Do you have the right people in place? Do you have the right regulatory people in place just compared to where the priorities were before? Thanks very much.

Thanks, Jami. Murdo, why don't you start?

Sure, thanks for the question, Jami. On CheckMate-067, first off, I thought Dr. Larkin did a great job presenting these important data from CheckMate-067. And as Tom mentioned, the first overall survival data set we have with an I-O/I-O combination. We have been out, obviously, getting feedback from customers in the U.S. and around the world. And I would say that the data have largely been viewed positively as confirmation of the effectiveness of Opdivo plus Yervoy in this patient population. I would agree with you; there are more questions that still people have because everybody is trying to segment patients in first-line metastatic melanoma. And we continue to work with physicians and investigators on the trial to answer those questions as best we can. I will say, for those physicians who have used the regimen and continue to believe that they're getting the best results possible in terms of response rates or overall survival in metastatic melanoma, this data set confirmed that belief. I would say for those physicians that were skeptical or overly concerned about the toxicities, and the toxicities are significant, we recognize that, but we're able to help most experienced treaters through the toxicities and we're able to help them also discuss them with their patients. For some of those physicians that remain, I would say, still concerned about the balance of efficacy and safety, particularly in the high expressers. Now, there are other descriptive analyses that have been completed. And I will remind you, obviously, it's hard to compare the active treatment of the combination to PD-1 monotherapy because the study wasn't powered to do that. So people are inferring from the data set certain things that we've yet to statistically prove. But we have seen also some good descriptive results in BRAF mutant patients where the combination of Opdivo plus Yervoy looks particularly good. So there's a lot still to unfold here. But nonetheless, my summary to all of that is net positive. Thank you.

Jami, Tom Lynch here. So a couple things. First, you have two questions I'll address. One is the mutational burden question. The second are the changes within the R&D organization. So regarding mutational burden. I think this is a great example of data that we're very excited about. As you know, when we went back and looked at 026 and we found that when we stratified patients by mutational burden, we saw that the response rate was superior for nivolumab versus chemotherapy. But it's early data and it's going to require confirmation in larger data sets and it's going to require more evaluation to have a sense of what its ultimate role is. And what I really want to stress is how difficult it is to develop the appropriate biomarkers. And just think about a drug like trastuzumab. And when we started developing trastuzumab, we thought that expression of immunohistochemistry of HER-2 was going to be the ultimate marker. It turned out to be HER-2 amplification that was the marker. Think about the development of erlotinib and gefitinib. We thought it might be expression by IHC. We then thought it was going to be amplification. It turns out it was EGFR mutation that was the most important. And then also think about siltuximab and its development in colorectal cancer. It wasn't EGFR expression, it turned out to be RAS mutation. So these are all very complex areas that require a very committed and disciplined approach to biomarker development. So the second question you asked comes down to the changes within R&D. I want to just start off by saying, and I emphasize what I said in my opening remarks, is that the quality and depth of the people and the talent at Bristol-Myers Squibb in R&D I think is exceptional.…

Thanks Jami. Melody, can we go to the next question, please?

Yes. We'll go to Alex Arfaei with BMO Capital Markets.

Thank you for taking my questions and congratulations on a great quarter. Sorry if I missed this earlier. But was there any inventory or deferred revenue impact on your I-O franchise, particularly in Yervoy? I'm just following up on an earlier question. How should we think about the patent life for Eliquis in major markets? Thank you.

This is Charlie. As I mentioned in my comments, in the fourth quarter of last year, we booked the deferral that we had on our books from France and Germany. That equaled about $250 million. So that happened. When you look fourth quarter versus first quarter, we did have that one recognition last year. We did see some slight inventory builds for both Opdivo and Yervoy in the U.S. Opdivo was about $30 million, and for Yervoy about $20 million.

And with respect to your question on Eliquis patents, so our composition of matter patent expires February 2023. We have a pending request for restoration of patent terms, which goes until November of 2026.

Thanks, Alex. Melody, can we go to the next question, please?

We go to David Risinger with Morgan Stanley.

Dave? David R. Risinger - Morgan Stanley & Co. LLC: Sorry, I was on mute. My apologies. So I have a few questions for Tom, please. And congratulations on your new role, Tom. With respect to the first one, that is Opdivo's chemo lead-in strategy to enhance Opdivo's performance in lung cancer is something new. Could you talk about that a little bit more? And then second, with respect to the forthcoming MYSTIC trial readout this summer, I guess what would you focus Bristol-Myers investors on when we're looking at those MYSTIC study results and thinking about cross-trial implications for CheckMate-227? And then finally, I think you had mentioned that in the first half of 2018 you'll have the CheckMate-227 results. But CheckMate-227 is a very complex study with three segments. Do you expect all three segments to be reporting in the first half of 2018? And also, what is the interim look in the fourth quarter of this year? Is that just in the positive cohort? Thank you.

Thanks, David. Let me just start by giving you our perspective with respect to your question on MYSTIC. And as we've commented before, these are different trials with different molecules and different study design. We've actually I believe really worked hard to optimize dosing and schedule. So it's very difficult to really speculate on any implications on the results of one study versus the other. Now I'll ask Tom to answer your question with respect to chemotherapy and then provide some context on timing for CheckMate-227.

Thanks, Dave. So regarding the rationale for looking at Opdivo in chemotherapy, I like to think that there might be two ways that this could be helping in this. As you know, chemotherapy has been the backbone of how we've treated lung cancer for the past 30 years. And in studies, and again it's not just with I-O, but it's also looking at targeted agents. There may be a rationale for the fact that the rapid progressors might be better suited by getting chemotherapy on board early, that there may be some rationale for that. The second thought is perhaps the addition of chemotherapy to I-O agents could help to create a more favorable environment for immune recognition and activation of the immune system. So I think there are good rationales for looking at chemotherapy with Opdivo, and I think one of the things that we're very interested in is how the combination of Opdivo, Yervoy, and chemotherapy performs. And we'll be looking at that quad combination to see where that might fit into the treatment of patients with non-small-cell lung cancer. Regarding the timelines, obviously, these patients were accrued to the different portions of the trial at different times. I don't anticipate that we would have all of the data in the first half of 2018. As you know, this is an event-driven world that we live in, in these trials and that we're going to need to see how the data matures. But we do hope to have the first data in CheckMate-227 at that point. Thanks.

Thanks, Dave. Melody, can we go to the next question, please?

We'll go to Andrew Baum with Citi.

Thank you, a couple of questions, please. Just continuing on the theme of chemo at your CheckMate-568 trial, if I were to ask you whether your intended rationale by combining chemo with dual checkpoint blockade is simply addition or some form of immunopotentiation, I'd be interested in your current thesis. Second, you mentioned Charlie Swanton's papers recently published, where he clearly showed that chemo increases subclonality, and thus may actually impair subsequent long-term outcomes arguing against increasing chemo as first-line. Do you think there is merit to that hypothesis? And then finally, I noted you had started a fairly substantial Phase 1 trial with your next-generation CTLA-4. I know you consolidated serum (56:11), presumably enhanced T-REG depletion. I know that that mechanism has been downplayed by Bristol as being contributing to Yervoy's efficacy. I'm just interested in whether there's a rethink of that and obviously the implications given the other CTLA-4 molecule in Phase 3 trials. Thank you.

So, Andrew, thank you for your questions. I'll break them down into the three parts. The first is the rationale behind CheckMate-568, whether it's due to addition basically of the two mechanisms versus whether it might be more immunostimulatory. Again, I think that there might well be arguments you can make in both respects, and I keep an open mind to both of those. And I indicated that in response to Dave's question earlier. So I actually don't know which of those two is going to turn out to be most important. I think again with regard to Charlie's work on the complexity, the genomic complexity and heterogeneity within tumors, I do think we need additional time to see how this plays out in terms of understanding how this impacts the therapeutic classes that we use. And again, I await the collaborations that we'll be doing to look at those endpoints. And finally on CTLA-4 and our next-gen areas, I think that T-REG depletion still is a valid endpoint in developing I-O agents. Whether it's the principal activity of CTLA-4 or not I think is something that is an open question. But among our agents that we're developing in I-O, we have multiple drugs that we think have as part of their mechanism depletion of T-REGs. So I do think that there is potential for T-REG depletion to be important in the activity of I-O agents in our next-gen compounds as we bring them forward.

Thanks, Andrew. Melody, I think we have time for one more question.

We'll go to John Boris with SunTrust.

Thanks for taking the questions and congrats on your new role, Tom. First question for Murdo, just if you look at Opdivo's use currently, what percent is coming from lung, melanoma, renal, and if we can get a split in lung, in particular in first-line versus other line use in the U.S.? The second question has to do with the use of the diagnostic for PD-L1 testing. It seems as though many centers have gone to an exclusive use of testing, and a very high percentage of the centers are now testing. Just any thoughts on how that may shape the landscape, especially as Merck is attempting to pivot into first-line lung use? And then third and final question on the FGF21, just your thoughts, Tom, around single-agent use in fibrosis and NASH versus combination use. So one of the major competitors that has farnesoid X antagonists has certainly moved very aggressively on developing relationships and combination relationships with both the Tobira and Addis assets to maximize efficacy in this area. Just your thoughts and your openness on the importance of combination relative to single-agent therapy would be helpful.

John, I'll address your question on testing first. Basically, what we're seeing in first-line is about 80% of patients are receiving a PD-L1 test. It is becoming, I think as you mentioned, a standard practice, and we continue to feel that that will happen in markets outside the U.S. as well as testing proliferates. So all new early diagnosed first-line lung cancer patients will have a PD-L1 result. In terms of Opdivo use overall, we've been very pleased. I think the team in the U.S. as well as the teams around the world have done a great job both defending competitively as well as establishing the launch trajectory of Opdivo. And right now in the U.S., about 55% to 65% of our total business is in lung, all lines and non-small-cell lung cancer all lines. And when you go worldwide it's higher than that given that we have a significantly lead over Keytruda.

So, John, thank you for your questions about FGF21. I'd be delighted to have a chance to offer some thoughts on this. We're very excited about FGF21 because it's the first of our agents to come forward from looking at fibrosis and NASH, and we think this drug has a number of favorable features about it. It impacts the metabolic drivers, fat accumulation. It has anti-inflammatory properties, and we also believe it's going to impact fibrosis within the liver itself. And when you think about the kinds of patients who get NASH, these are patients who are dealing with obesity, patients who are dealing often with diabetes, metabolic syndrome. It's a difficult group of patients to treat. And so how we develop these drugs, whether they're single agents or combinations, I think a key look at side effect profiles, at lipid profiles, at making sure that we're able to achieve endpoints that are meaningful, meaning reduction in fibrosis ultimately, I think will be the ultimate goal. Whether it's a single agent or combination, I obviously remain open to both opportunities. I think there are reasons to think our drug might work as a single agent. However, I think it would be important for us to also consider what combination therapy in the future might look like in this setting. It's an important unmet medical need that we are very happy to be able to offer some options for.

Thank you, Tom. Again, we are in a very strong position financially, commercially, clinically, and I believe we are well positioned to continue to grow the company and play an even more important role in the lives of our patients. Thanks, everyone.

Thanks, everybody. This concludes our call. As always, if you have any follow-ups, give us a call or shoot us an email. Thank you.

Ladies and gentlemen, again, that does conclude today's conference. Thank you all for attending.