Good day, and welcome to the Bristol-Myers Squibb 2019 Second Quarter Results Conference Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Mr. John Elicker, Senior Vice President, Public Affairs and Investor Relations. Please go ahead, sir.

Thanks, Orlando, and good morning everybody. We're here to discuss our second quarter earnings as well as the news that was press released last night. With me this morning, Giovanni and Charlie will prepared remarks. Chris Boerner, our Chief Commercial Officer will be here for Q&A, as well as Fouad Namouni, our Head of Oncology Development is here for Q&A as well.

Thank you, John, and good morning, everyone. I'm pleased to speak with you today about our strong performance in the second quarter and the progress we've made on our planned acquisition of Celgene. But, first, let me start by discussing the results we announced last night and frame what it means for Opdivo and how I think about our outlook going forward. With respect to 227 results, as you know, we announced important results last night. We had a successful outcome, but also a part of the trial that didn’t meet its endpoint. Starting with the results of Part 1a, this is the third major tumor in which Opdivo+Yervoy shows an overall survival benefit in the first-line setting. And we believe these results represent a potentially differentiated opportunity in first-line lung cancer. If approved, the combination would provide an additional and chemo-sparing treatment option for patients. And I have full confidence in my commercial team's ability to execute in this competitive marketplace. As we noted in the press release, we also saw in an exploratory analysis, an overall survival benefiting in PD-L1 negative patients. We’ll be sharing all the data at an upcoming medical meeting. So, I won't go into the specifics today. Now, turning to the results of Part 2. They were not what we had hoped for. There are, however, important aspects of the study results to keep in mind. First, we're looking at one-year landmark analysis. Opdivo plus chemo performed consistency with the experimental arms of other successful trials. The chemo control arm somewhat overperformed compared to what we regularly see. The performance of Opdivo was consistent with our expectations, but the trial was not positive.

Thanks, Giovanni, and good morning, everyone. Building on Giovanni's comments about Checkmate-227, I'll start by providing some additional color regarding dynamics in our I-O business during the quarter and how we think about it moving forward. In U.S., we continue to see strong share across indications including both metastatic and adjuvant melanoma, first-line renal cell and second-line lung. As expected, we also continue to see the size of the eligible pool of second-line lung patients decline, which has impacted demand sequentially.

Thanks, Charlie. Orlando, I think, we're ready to go ahead to the Q&A.

Absolutely. Thank you. Our first question from Tim Anderson with Wolfe Research.

Hi. Thank you. First question is, results of 227, how do they influence, how you now look at the odds of success on Checkmate-9LA? Does that trial have CTLA component? I’m guessing you'd say the odds of success with that one have to be higher. Also, can I confirm that that trial stratifies by PD-L1 and not by TMB, which seem to make sense in light of the Part 1 findings. And the second question is on Part 1 itself. Just directionally, hoping you can see whether the magnitude of clinical benefit was the same in PD-L1 positive versus negative patients. It seems that across few different tumor types now, CTLA-4 combination may offer its greatest value in the PD-L1 negative segment where we know that PD-1s at least by themselves don’t really seem to work. Is that potentially where the best and easiest positioning of the combo would be?

Thank you, Tim. This is Giovanni. I'll ask Fouad to answer your two questions. I just want to start and reiterate how excited we are to have an opportunity to play in first-line lung cancer, which was what we now know with the results of Part 1a of 227 and particularly with Opdivo+Yervoy, which as you know we’ve been successful in establishing in two other tumor types, melanoma and renal. So, I think that's important news for us. And I'll ask Fouad to give you his perspective on 9LA and PD-1 expression.

Thank you, Giovanni, and thank you, Tim, for the question. We are confident in 9LA. Let me remind us what 9LA is. 9LA is the combination of two active Checkpoint inhibitors, Yervoy and Opdivo in first-line lung cancer added to two cycles of chemotherapy where the goal is to really manage the early progressions but also create a load of new antigens that will continue to stimulate the immune system. So, we continue to be confident in 9LA in the way it is designed. For your other questions around Part 1, and in fact, this is third time we see Yervoy and Opdivo perform in terms of overall survival versus standard of care in first-line -- in a cancer and this time in first-line non-small cell lung cancer. I think, we have seen clinically a significant and meaningful overall survival of the combination in first-line lung cancer PD-1 positive. We have seen improvement in overall survival in the exploratory analysis of PD-L1 negative. Overall, I would say, Tim, we are seeing the same pattern for Yervoy and Opdivo that we have seen in other tumors, like melanoma and renal cell carcinoma.

Next, we’ll hear from Seamus Fernandez with Guggenheim.

Thanks very much for the questions. Just a little surprised by the new process with the beep. So, just a couple of quick questions. First off, I was just hoping the team could comment on whatever is possible to comment on relative to the Senate bill at this point and kind of the relative exposure that Bristol has to U.S. spending alone and then prospectively in combination with Celgene, and just hoping to get a little bit of your thoughts on the Senate bill itself. And then, separately, just hoping to get a little bit of color on 227. More than anything, the question really is, when we see the data presented, both from the Part 2 and the Part 1a study, are we going to not only fully understand, but also will come away with Bristol proving that the benefits of adding Opdivo on top of chemo therapy having clear benefit also in non-squamous patient perhaps on PFS. And then in terms of Opdivo+Yervoy, are we likely to walk away with a clear view that not only is Opdivo a monotherapy showing improvements, but that Yervoy on top of that is additive?

Thank you, Seamus. This is Giovanni. Let me start with your question on what's happening from a pricing and policy perspective. Obviously, as you’ve seen the environment is very fluid in Washington. I think, these are early days. And it’s difficult to point to specific outcomes of the dialogue that is happening in Congress. As you know, we’ve discussed for quite some time the perspective that I have that we're at the beginning of the period of change from a policy perspective. And from my point of view, I actually support discussion about change because I think it's important to address the affordability issues that patients have. Now, remember, those affordability issues are clearly primarily driven by benefit design and the very high out-of-pocket expense patients have because of the way insurance plans are designed, and some of the misaligned incentives in the system. And one of the things that I would like to say is that we are concerned with many of the proposals that we are seeing, including some of the proposals that are included in the Senate bill, because while they are very punitive in many ways for the industry and particularly for companies that are focused on innovation, they don’t really benefit many patients, they only benefit about 2% of patients in Medicare as currently grafted, and they don’t address some of the misaligned incentives in the system. So, I think, we’re all for change and we have proactively proposed policy solutions that address issues that are making it difficult for patients today. And I think what we've seen so far, doesn't address some of the big issues that we see in the marketplace. Now, as I said before, I think, it's important to remember, from the very beginning, I felt that it was important for us during the period of policy transitions and changes, to have a broader and more diversified portfolio that goes across multiple types of reimbursement with more growth opportunities into different diseases and more launch opportunities to accelerate really the lifecycle of the portfolio. And I think that's what the Celgene acquisition does for us. I think, it becomes even more important, given the uncertainty of these days and times to have a broader portfolio, to have a more diversified portfolio. And I think it's critical for us. And that's what we do through Celgene. Now, with respect to your question, going to the next level of detail, when we look at the impact of some of the measures that are being proposed, we can see, for example, that yes, there could be a potential impact for Revlimid. But, I would also remind you that some of those measures only would come into effect in 2022, when we would be really, I would say, at the end of the lifecycle of that asset. And also, there are other parts of our portfolio where there would be offsets going in the opposite direction. So, I think it's early to give you a definitive answer on what the impact would be. But, I think that given the broad portfolio we have, I think you would see ups and downs. And many of them, obviously, would be impacted by where every one of the products would be in their lifecycle. And again, Opdivo -- sorry, Revlimid would be towards the end of that cycle. Chris, do you have anything to add?

Yes. I mean, Seamus, let me just pick up where Giovanni left off. I think, you have a number of different proposals embedded within the Senate bill, there are a number of different additional steps that are going to be required before anything is enacted. And obviously, as you know, specifics matter here. What I would say is that, as Giovanni mentioned, having a more diversified portfolio is better. When you look at it on a product by product basis, obviously, the allocation of the business by payment mechanism is going to be important in determining the impact. What I will say is, from a BMS exposure, what we've said previously, I'll just remind you, is that looking across our business, we have about 24% of our business in Part B, about 26% in Part D, and relatively little Medicaid exposure. With respect to Celgene, remember, we’re two separate companies, so as for specifics, you would need to ask them. But multiple myeloma, for example, is primarily a disease of the elderly. So, we would skew a bit more towards the Medicare population.

And Seamus, this is Fouad, for your second question around what we will see, what we will understand when we show the data from both Part 1 and more data from Part 2, will be the following. Let me start with Part 1. When we will show the data from Part 1, clearly, we will understand the performance of nivo and added benefit of Yervoy on top of nivo, very clearly. And we will understand also the overall pattern that we are seeing with immunotherapy combinations, like depth of response, rate of complete responder, the durability of response and the long-term survival. And I think these elements would be very clear when the data will be presented. For Part 2, in addition to what was reported in the press release today, what we will understand is one, the performance nivolumab plus chemotherapy versus other combination of PD-1 and PD-L1 agent in chemotherapy. Basically, I think, we will see that the performance of nivolumab in chemotherapy is very consistent with what we have seen with other PD-1s and PD-L1 agent combined with chemotherapy. I think, we would see that the chemotherapy comparator arm in Checkmate, which we said on Part 2, has outperformed what we would expect in the standard of care. Let me remind us that, what we know from the activity in standard of care in terms of median overall survival, the chemotherapy is between 13 to 14 months, and this is supported by real-world evidence. I think, the chemotherapy arm outperformed in 227, maybe underperformed in other studies. And I think, this is broadly what people will take when we go in depth into the data from Part 2.

And next, we will hear from Chris Schott with JP Morgan.

Thanks very much. Just two questions on 227. I guess, first on the Part 1 study. I know, you can't go into specific numbers yet. But, do you feel you need to see an overall survival hazard ratio close to that seen with chemo Keynote-189 for this offering to be competitive, or should -- or are you more focused on those factors, like CR rates and depth of responses you think about what you need to be to find kind of role and a niche for that combo? My second question was on Part 2 and that stronger kind of control arm survival rate. Can you give us any color in terms of crossover rates in second-line -- to second-line I-O in Part 2, and was that meaningfully different than what we’ve seen with prior competitor studies? Thanks very much.

This is Chris. Maybe let me start and then I'll turn it over to Fouad. As we said previously, as you think about how physicians have conversations with patients, very rarely do they have a discussion around hazard ratios. And so, we believe, what is important and what we hear from customers is, how are patients performing over time, so that would imply landmarks are very important. And one of the key drivers for treatment choice in first-line lung cancer continues to be our product showing durable efficacy. And so, when you step back and you think about first-line lung cancer, despite all of the progress that we've made in that area over the last few years, we need to keep in mind that the majority of patients progress within one year. And in doing so, many of the spaces actually burn through two options, I-O and chemo. So, there is a need for more options that potentially spare patients from chemotherapy and have proven OS and durability benefit, and that's what we hear consistently with respect to how physicians are going to make choice in first-line lung cancer.

And that with regards to the Part 2 crossover, to I-O therapy in the chemotherapy reference arm, we have seen a level of crossover, about a third of patients, which is consistent with many other trials. And we do not believe that the crossover is the reason explaining the output performance of our chemotherapy arm.

Next, we will hear from Navin Jacob with UBS.

I just wanted to understand the rationale behind the non-squamous as the primary endpoint for Part 2. The squamous data looked quite good in comparison to 407. And if you could, as it correlates to that, could to clarify if 9LA has a similar design as Part 1a where the primary endpoint is also specific to non-squamous…

Thank you, Navin, for your questions. So, first, the non-squamous and the primary endpoint, this is the largest population, this study. The squamous population is pretty really a small percentage of the first-line. Data available to us at the time showed, there is benefit, and most of the benefit we see in the non-squamous population. Therefore, the study was really designed to ask the non-squamous question. For 9LA, the study is designed to go to all commerce in terms of histology and biomarkers. And looking at the first-line population of patients is different where you combine both checkpoint inhibitor and diminishing the number of cycles of chemotherapy.

And next, we will hear from Terence Flynn with Goldman Sachs.

Maybe as you think about the outperformance of the chemo arm in Part 2 of Checkmate-227, can you help us think about your adjuvant program and maybe again any puts and takes there as you think about design or performance of those control arms and just confidence level in those trials, given that chemo outperformance that you saw in 227 Part 2? Thank you.

So, in the adjuvant space, I think -- so first, we're very pleased to see our data in the first-line setting with Part 1a. And we believe the performance of immunotherapy in the adjuvant setting really will be good and makes us really confident in the adjuvant setting. In terms of standard of care, there were so many evolution and changes in the management patients in terms of supportive care and how we do it in the metastatic setting of lung cancer. There was not a lot of progress in the adjuvant setting in terms of standard of care and comparators. And we believe that our adjuvant and early program is pretty strong, and actually we have, as you probably have seen, started steady in this stage 3 setting of lung cancer of nivolumab and nivolumab plus Yervoy versus durvalumab and looking at overall survival in this population of patients.

And next, we will hear from Steve Scala with Cowen.

The Company seems to be suggesting that the tail of the Opdivo+Yervoy curve will impress in Part 1a when its presented, it has been mentioned two or three times on this call already. In Keynote-189, Keytruda plus chemo saw 69% of patients alive at 12 months. So, I’m interpreting Bristol's positive tone as Opdivo+Yervoy will show more than 69% of patients alive when Part 1a is presented. And I’m just wondering, would you suggest, I consider other interpretations of the Company's positive tone? So, that’s the first question.

Go ahead, Steve.

Yes. The second question, is the Company dealing with crossover differently in 9LA than Checkmate-227? Thank you.

Yes. Steve, let me just start before Fouad answers your two questions, by saying, I think, we’ve really pointed to the types of benefits that I-O and plus I-O. And we’ve seen consistently in tumors where Opdivo+Yervoy has provided a benefit. The physicians really have valued deep responses, complete responses, the durability of responses, and as the follow-up of those studies has continued is when we really started to see the value of the trend. So, I think that's an important consideration. But, Fouad?

I think, Giovanni summarized very well our observations on the lung cancer. I would just add, this is a pattern that we have seen in other -- two other major cancers, in melanoma and in renal cell. For the question on 9LA crossover, the crossover is not systematic in 9LA, it was not systematic in 227. So, patients, when they progress in the study on the chemotherapy arm, they will receive standard of care therapy by their physicians.

And this is Chris. Let me just jump in here and make a couple of comments. So, first, as I mentioned, there's still considerable unmet need in first-line lung cancer. And we think there's an important role that Opdivo+Yervoy has to play. And I think that as you begin to think through what that opportunity could look like, remember the experience that we've had without Opdivo and Yervoy in melanoma and renal cell which are the two diseases in which we've seen significant benefit there. And what we see with the regimen is significant responses, including impressive CR rates, you see durable responses, you do see a compelling overall survival benefit that has that characteristic plateau in the Kaplan-Meier curve, which appears to be somewhat differentiated from other mechanisms. And you see a side effect profile that has advantages relative to chemotherapy. So, we think there's an important role to play for dual I-O therapy without Opdivo+Yervoy in that setting. The other thing I would say is that when you look at Opdivo+Yervoy and the clinical trials also, we see in the market when we hear back from customers is, it's very consistent with their real world experience with the regimen. Why is that important? Well, first, that’s frequency, not the case with other modalities. And second, we have a strong base of Opdivo+Yervoy users across tumors. In fact, and when you look in lung cancer, of the highest prescribers in lung cancer, just over half of those physicians have used Opdivo+Yervoy in another tumor, notably melanoma and renal. And if you look at all of the targets in lung cancer, just over 30% have used Opdivo+Yervoy in other tumors. And those physicians account for about 45% of the total opportunity in lung cancer. And those are the very physicians who have made Opdivo+Yervoy a standard of care in those other tumors.

And next, we’ll hear from Matt Phipps from William Blair.

Thanks for taking my question. I guess, on the Part 1a or Part 1, you’ve now had Opdivo+Yervoy show improved survival, at least numerically in both the biomarker positive and biomarker negative population. So, how do you think about the totality of data when you're talking to these physicians and regulators as far as patient populations? And then, secondly, can you talk just generally about the potential for approval neoadjuvant lung cancer, based on pathological complete response rate?

Matt, thank you for the question for Part 1a and Part 1. Overall, as we said, we have seen clear benefit, clinically meaningful and statistically significant of Yervoy+Opdivo in the primary end of the study, which is in PD-1 positive. We have also seen good survival benefits in the PD-L1 negative. I think, the totality of the data will be seen. We’re not going to comment on our interaction with health authorities. They will be happening in the next days and weeks. But, I think, we have seen benefit across the board in terms of biomarkers, in terms of the totality of the data. In terms of neoadjuvant, I think, major pathological responses in lung cancer has not been actually used historically as an approval and endpoint by the USFDA by our health authorities. On the other hand, it’s going to depend on how meaningful is the data. And when we see our data from study 186, we will be able to interact with authorities and see what will be the outcome of that.

And the only thing I would add to that is that, while obviously we’ll have to see how the discussions with regulators proceed, what I would say is, just consistent with what I've mentioned previously, we think A, there is opportunity in first-line lung cancer from an unmet need standpoint and we think that Opdivo+Yervoy has the opportunity to provide an important treatment option in PD-L1 patients across the full spectrum of PD-L1 patients, positive patients.

And next, we will hear from Jason Gerberry with Bank of America.

Hey. Thanks for taking my questions. I just wanted to come back to comments about Opdivo in 2020 and just make sure I understand the moving parts, correctly. So, it sounds like, it’s flat or down year, driven by the drag on sales for lung cancer in 2020, presumably with like maybe flattish sales in renal with growth coming from melanoma and some of the other tumors. So, I just wanted to make sure, I sort of had a rough sense of how the moving parts will evolve before you get the lung expansion opportunity in 2021. And then, my second question, can you just elaborate a little bit more on FTC trends as it pertains to defining markets? And really what I'm getting at here is, whether or not your sales process could include companies that have meaningful share in the injectable space of moderate to severe psoriasis? Thanks.

Thanks, Jason. Let me ask Chris to give you some perspective on different dynamics impacting the Opdivo business today and into next year. And then, Charlie can comment on the FTC and where we stand, and what's the process there.

Yes. So, let me just start with, as we think about 2020, obviously performance in ‘19 becomes very relevant. So, where we are today, we continue to see strength in our core business. We continue to lead in virtually every tumor in which we are promoting. And as you think about our core tumors, the large tumors, first-line metastatic melanoma, second-line renal cell, second-line HCC, we continue to see I-O shares at or greater than 50%. And then, we've talked a lot about the two big growth drivers that we have for this year, which are notably first-line renal cell and adjuvant melanoma. And again, there, very happy with the continued performance of the teams. In first-line renal cell in the U.S., we are holding share at around 35%. Obviously, we have seen some impact of I-O plus TKI, mainly in the favorable patients less so in intermediate and poor, which is where we are indicated. And then, outside of the U.S., we’ve seen good uptake in key markets, notably Germany and Japan, where we have access and we expect additional access approvals later in the year. Similar story on the adjuvant melanoma side, U.S. shares still holding around 70%, in spite of competitive entries there, and outside of U.S. is still very early in terms of access. As we think about 2020, while we're very pleased with the results that we presented yesterday for Part 1a, given the competitive dynamics, the timing of data readouts, we do think there will be some pressure on Opdivo in 2020. But the growth picture becomes much clearer as you get into 2021. And exactly what that profile looks like is going to be informed obviously by the opportunities we see with 227. You will continue to see a stabilizing of the dynamics in second-line lung cancer, which is important. Just to remind you, we expect second-line lung in the U.S. to stabilize in terms of I-O eligible patients at the end of this year, little bit later as you get into ex-U.S. market. And then, clearly, we will be looking for some key study readouts that will inform that near-term growth picture, notably 9LA in lung cancer. We’ve got first line GBM, 9ER in first-line renal cell and then first-line studies in head and neck and esophageal. As you get later out, clearly, the adjuvant programs become important.

Yes. Just, Jason, in regard to your question on the FTC. We believe that the divestiture of OTEZLA will satisfy the FTC’s concerns and allow us to close the transaction on a timely basis. We won't have full clarity on who is an acceptable bidder until we present a draft sales agreement to the SEC. But had to had some preliminary perspective from the FTC. So, we feel directionally we have a good understanding. As I mentioned in my comments, based upon what we see today, we believe we will be able to run a robust process that will generate significant bidding interest.

Next, we’ll hear from Umer Raffat with Evercore.

First, I want to touch upon a trial I feel like we haven't had much discussion on, which is your LAG-3 Phase 3 melanoma, which is due perhaps in the next 12 months or so as per ClinicalTrials at least. And my question is, we know it’s fully enrolled. And where is your expectation and how are you thinking about the trial? I also noticed it has a PFS primary endpoint, not an OS. So, I was curious to get your perspective on that. And then, secondly, it was helpful commentary on the market shares in various Opdivo indications. I was curious if you could give a bridge on a dollar basis on progression of sales from 1Q to 2Q, for Opdivo U.S. Thank you very much.

Thank you. Why don’t we start with Fouad on the LAG-3 program.

Thank you, Umer for the question and the melanoma development -- LAG-3 is a Phase 2/3 study. Looking at the addition of LAG-3 to Opdivo -- comparing to Opdivo. I think as you mentioned, we will have in the next month the first readout from the Phase 2 part and we will see if we hit the threshold to move to the Phase 3 part, as we said earlier.

And then, let me just comment on Q2 dynamics. So, Q2 dynamics, Opdivo was down slightly about 1% in the U.S. And that’s really a function of a few things. First, we continue to see pressure in second-line lung cancer, due to the decline in the overall opportunity. That's what I referenced previously around the percent of I-O eligible patients. What I will say though is within the pool of I-O eligible patients, we continue to hold a market share for Opdivo of around 40%. We've also seen some competitive impact in tumors outside of any discussion around lung cancer, notably in first line renal cell, and I referenced those previously. Again, they’re in that market. I’m very proud that the team’s holding share of roughly 30% to 35%. And the impact really of I-O TKI has been confined to favorable patients. And where they have gotten additional uptake outside of those favorable patients, has mainly come at the expense of monotherapy TKI. And then, beyond that, I think we’ve spoken to what the near-term opportunity looks like for Opdivo.

And next, we'll here from David Risinger with Morgan Stanley.

Thanks very much. I just wanted to pivot to ask about the TYK2 development program. Could you just provide an update on key trial progress, and when you expect enrollment to complete, and when you expect to be able to share key results? Thank you.

Sure, David. Good morning. This is Giovanni. So, as you will remember, we have two studies in our Phase 3 program for TYK2, and both studies are progressing rapidly through the enrollment period. And the design of those studies requires a one year treatment period. And so, we do expect to see data, one year after the completion of enrollment, which I expect to be sometime towards the end of next year, in that timeframe.

The only thing I would just add to that is we continue to be very excited about the profile of TYK2 and how it can play an important role in psoriasis. As you may recall, psoriasis is a debilitating disease, has very serious comorbidities, there's a considerable psychosocial cost associated with the disease. And based on what we've seen with TYK2, at least in the early days, we're seeing very good activity approaching biological efficacy with an easier mode of administration. So, we think it has an important role to potentially play in psoriasis. And from a commercial perspective, we're still very excited about the opportunity here.

Orlando, do we have any more questions?

And there are no further questions. I'll turn the call back over to Mr. Elicker for additional or closing remarks.

Thank you. Thanks, everyone. In closing, this is an important time for us at Bristol-Myers Squibb. We've had another strong quarter demonstrating our ability to execute on many priorities. We delivered good financial results, driven by strong commercial execution. And going forward, we will continue to advance our pipeline and progress the integration planning with Celgene, including the divestiture of OTEZLA. The data that we announced in first-line lung cancer has really the potential to help us bring a new and an important option to more cancer patients who continue to have important unmet needs. Thanks everyone for participating in the call.

Thanks, everybody. Tim and I, as always, are available for follow-ups. I appreciate you joining the call this morning.

And this concludes today's call. We thank you for your participation. You may now disconnect.