Bristol-Myers Squibb Company (BMY) Earnings Call Transcript & Summary

Good afternoon. This is Tim Power from IR at Bristol-Myers. Welcome to our investor event at ASCO 2021. And as you have seen, we had some important data from our oncology pipeline presenters over the weekend, including for our next-generation I-O agent, relatlimab. So there's a lot to cover today in a short time. And I'm joined for today's discussion by Samit Hirawat, our Chief Medical Officer; and by Chris Boerner, our Chief Commercialization Officer. Samit and Chris will go through the slide presentation and then will be available for Q&A at the end. If we can go to our next slide, which contains our forward-looking statement. I'll hand over on Slide 3 to Samit to get the presentation started.

Thank you, Tim, and good afternoon, everyone. I'm happy to have this opportunity to discuss the data presented from our pipeline at this year's ASCO meeting, which is of course a very important medical congress as always. At this year's ASCO, BMS demonstrated that it continues to be a leader in oncology. We are pleased to be able to say that BMS is the only company that has successfully developed and executed pivotal trials for 3 separate I-O agents with Opdivo, Yervoy and now relatlimab. And with these agents, we've built a broad program that includes monotherapy and combination approaches. These therapies have provided improved outcomes across various tumor types and addressed the need for both early and advanced diseases. So what did we learn at this ASCO? First, we continue to deliver important new therapies with nivolumab, including those that are impactful in the neoadjuvant setting for non-small cell lung cancer and in upper GI malignancies. There is continued strengthening of evidence for the durability of benefit provided by nivolumab and ipilimumab. And importantly, we presented clinically meaningful data for our latest I-O doublet, relatlimab plus nivolumab, in a fixed-dose combination with relatlimab being a novel anti-LAG-3 antibody and the first of its kind to deliver successful Phase III results. Moving into the data right away. We wanted to highlight some important results from study CheckMate 816 for nivolumab plus chemotherapy in the neoadjuvant setting for non-small cell lung cancer. In April, this trial met its co-primary end point for pathologic complete response rate, as you've heard before. As you can see on the left, the PCR rate was a meaningful 24% in the nivolumab bar versus merely at 2% in the chemotherapy-alone arm. At ASCO, we also presented data on the surgical outcomes post treatment. First, adding nivo to chemo did not increase postsurgical complications. Moreover, the majority of patients who received nivo plus chemo had timely surgery with more complete resections and fewer pneumonectomies. Finally, the graphic on the right shows that with nivo plus chemo, an increased number of patients were able to undergo minimally invasive procedures compared to those treated with chemo alone. Overall, these data represent a potential option for patients with early-stage lung cancer when approved. I'll now move on to the data we shared in GI cancers. In particular, I want to highlight the data presented for CheckMate-648, our Phase III trial in first-line esophageal cancer, which met its primary end point of overall survival and, additionally, progression-free survival in patients with PD-L1 expressing tumors. I won't go through all the details because I know you've heard them all. But here are, as you can see, the trial design which studied nivo plus chemo as well as nivo plus ipi compared to chemotherapy alone. On the next slide, here are the Kaplan-Meier curves for the nivo chemo arm from the medical presentation. Let me share with you what's important from our perspective about what we see here. First, superior overall survival not just in PD-L1 expressors but in all randomized patients as well; and secondly, a very clinically meaningful result in both populations. On the next slide, here is a similar scene that you can see now showing superior overall survival for the nivo plus ipi arm with a hazard ratio of 0.64 for the PD-L1 expressors and 0.78 for all randomized patients, which potentially could present patients with a chemo-free option when approved. And finally, reflecting on each of the 4 curves shown, you'll notice that over 50% of patients were alive at 12 months. On the next slide, you can see a manageable safety profile was demonstrated for both active arms, nivo chemo and nivo ipi. Note that while nivo chemo had more treatment-related adverse events, fewer of them were deemed serious. So all in all, no unexpected safety signals. Now switching to dual I-O with Opdivo + Yervoy in melanoma, where we presented additional data this ASCO for CheckMate-067. At 6.5 years, these data represent the longest follow-up for a dual I-O trial. Nearly half of the patients who received this therapy remain alive. You can see the durability of this regimen continues to be demonstrated by the consistent shape of the overall survival curve with the tail of the curve clearly representing very good outcomes for many metastatic melanoma patients. As we celebrate the tenth year anniversary of Yervoy this year, we acknowledge this remarkable transformation in the outcomes of patients with melanoma where only a decade ago, the median survival is around 6 months. And there is an opportunity for a similar effect to be seen in first-line non-small cell lung cancer patients where we now have 4 years of follow-up with CheckMate-227 and 2 years of data for CheckMate-9LA, with the maturity of the curves for CheckMate-227, the longest duration of follow-up for dual I-O treatment in lung cancer. You can see the flattening of the curve and a distinct tail demonstrating that proven durability once again. We then look on the right side at CheckMate-9LA, where Phase III results had already validated that a small amount of chemotherapy can address the early part of the curve that we saw crossing over in CheckMate-227. These data showed durability as they mature. In fact, 38% of patients were alive at 2 years, which is relatively in line with what we observed with CheckMate-227. We are pleased with the evolution of the data in first-line lung cancer and we'll continue to follow, of course, the outcomes of these patients as well. Combined with the 4-year results from CheckMate-214 in first-line renal cell cancer, we now have long-term data across 3 tumor types where the promise of durability has been proven and delivered with dual I-O therapy. I want to switch gears a bit and talk to you about relatlimab, which we call rela. Rela is a humanized LAG-3-blocking antibody that restores the effector functions of exhausted T cells and reinvigorates their activity to direct an antitumor response. Earlier studies demonstrated that inhibiting LAG-3 and PD-1 checkpoints by using rela and nivo together enabled the T cell activation and enhanced the antitumor immune response versus nivo alone. That led to our Phase II/III study, which I will address briefly on the next slide. RELATIVITY-047 is the Phase II/III -- first Phase II/III study to evaluate the novel combination of rela plus nivo versus nivo monotherapy, a current first-line standard of care in advanced melanoma. Patients received either the fixed-dose combination or nivo monotherapy every week -- every 4 weeks. And by fixed-dose combination, I mean rela and nivo being co-formulated in the same vial. Patients were stratified by LAG-3 and PD-L1 expression as well as by BRAF mutations and metastatic disease stage. The primary end point of the study was progression-free survival by blinded independent central review committee. Secondary end points included overall survival and objective response rates, both of which the company remains blinded to at this time. Looking at the primary end point for RELATIVITY. You can see that rela plus nivo demonstrated significantly longer progression-free survival compared to nivo monotherapy with the median progression-free survival more than double the value seen for nivo alone. The benefit was seen early during treatment at the first -- at the time of the first scan and continued over time. The hazard ratio of 0.75 was clinically meaningful and highly statistically significant. On the next slide, we also see that the PFS results consistently favored a fixed-dose combination across all subgroups and stratification factors, including key factors like BRAF mutation status, LAG-3 expression, PD-L1 expression and disease stage. Moving to safety on Slide 16. The data from RELATIVITY shows that the fixed-dose combination of rela plus nivo was associated with a manageable safety profile and no unexpected safety signals. Overall, the type of adverse events observed were similar to those seen with nivo alone, although with occurrences at a higher rate, as expected with the dual therapy when comparing to a single agent. Because of this manageable safety profile, though, we believe rela plus nivo provides a potential to bring I-O treatment to more patients, including and especially those who may not be considered candidates for O + Y today. Now as we look to the future on the next slide, the potential for rela doesn't end with metastatic melanoma, and we are pursuing several additional studies. First, in a Phase III trial in adjuvant melanoma as we look to potentially take this treatment to patients with earlier disease. At this ASCO, you may have seen data presented this week and presented by Dr. Amaria from MD Anderson where nivo plus rela administered pre and post surgery achieved high rates of PCR and major pathologic response and encouraging data for an earlier-stage disease in melanoma for this combination. Now beyond melanoma, based on our understanding of the mechanism as well as signals from our Phase I basket trial, we also chose to pursue Phase II trials in 2 additional indications: hepatocellular carcinoma and non-small cell lung cancer. Now regarding non-small cell lung cancer, we expect to see safety data from our Phase II trial early next year, and that could enable us to move forward into Phase III. Now I just wanted to end here with a brief reminder of where we've been and the potential of what is yet to come. We look forward to sharing more data in the future as we continue to follow exciting and emerging science and bring more options to patients in need. And with that, I will now hand it over to Chris to give you his thoughts from the commercial side of things. Chris?

Thanks, Samit. That is a great segue for my first slide as I'd like to remind you of the legacy that we have had in I-O for over 10 years now. Combined sales for Opdivo and Yervoy have grown to $8.7 billion last year, reflecting the success of these 2 products which, either alone or in combination, are considered a standard of care across 11 tumors. And as we've said previously, we expect Opdivo to return to growth in 2021 and that our I-O business will play an important role in company growth beyond 2021. We see that growth supported by several recent approvals and positive data readouts: first, in first-line lung, where we have 2 different regimens to offer patients; second, in upper GI, we have had 2 recent approval as well as a positive Phase III readout, which will round out a very comprehensive book of business in upper GI cancer; third, we have important opportunities in other tumors, notably first-line lung -- renal -- first-line renal, where we have launched the 9ER regimen and in muscle invasive bladder cancer with CheckMate -274 where we have a PDUFA date of September 3. Finally, we're excited about our next I-O combination with relatlimab, which you just heard about from Samit, and the ability to expand our leadership with first-line melanoma with a fixed-dose combination, and I'll talk to you more about that in a moment. But let's first take a look at our growing upper GI franchise where we have options for patients, addressing multiple areas of unmet needs, including: first, the first approved I-O adjuvant treatment for esophageal GEJ cancer with CheckMate -577; second, the first approved I-O treatment in first-line metastatic gastric cancer with CheckMate-649; and with the recent success of CheckMate-648, a potential launch in first-line metastatic esophageal. Overall, we have the most comprehensive offerings across upper GI cancers, including those involving the esophagus, GEJ and stomach. Shifting now to metastatic melanoma. Bristol-Myers has demonstrated its leadership with Opdivo + Yervoy and that regimen over the last number of years. We now have an additional potential opportunity with the fixed-dose combination of rela plus nivo. As Samit mentioned, rela combined with nivo delivered a clinically meaningful PFS benefit versus PD-1 monotherapy, which is a great accomplishment knowing the high bar of PD-1 mono efficacy in melanoma. On Slide 11, we're also able to see the PFS for this option as well as demonstrated with Opdivo + Yervoy. This new regimen provides an opportunity for additional first-line melanoma patients to receive a combination treatment. On this slide, you see the current U.S. market landscape in first-line melanoma, which you can broadly think about in 3 distinct segments. Today, the combination of Opdivo and Yervoy represents roughly 1/3 of the market, reflecting the significant durability and long-term survival benefit this regimen affords patients. The next 1/3 of the market represents PD-1 monotherapy use. For a regimen like rela plus nivo, with a differentiated safety profile and a demonstrated clinically meaningful PFS benefit versus PD-1 monotherapy, this segment would be a clear near-term focus. There remains further opportunity for I-O penetration in the BRAF mutant type segment, the last 1/3 of the population. Many times, those patients with a BRAF mutation are automatically prescribed with TKI. Over time, we expect that prescribing behavior could shift, allowing these types of patients to benefit from the durability of I-O. Overall, in melanoma, physicians are looking for additional options to achieve strong efficacy and long-term durability for more patients. Over time, and with additional data, we'll be able to determine where we can further expand the commercial opportunity. While we were only able to highlight a few elements of our ASCO data and oncology program today, remember, we have a broad franchise that is expanding over time. In our solid tumor business, our inline brands continue to perform well, including Opdivo, where we expect growth later this year. We look forward to the potential launch of the fixed-dose combination and extending our existing leadership in melanoma which also continues to develop -- while also continuing to develop new agents with novel targets across various platforms. On the hepatology side, we continue to advance our new launches, including our 2 CAR-T products Breyanzi and Abecma, and further establish new medicines in myeloid disease with Reblozyl and Onureg. Finally, we look forward to strengthening our leadership in multiple myeloma with additional developments in our CELMoD agent portfolio and with other BCMA assets like our T-cell engager and ADC. To sum it up, as you've seen at this conference, we continue to meaningfully advance our leadership in oncology with a broad set of programs across tumors and stage of disease. Importantly, the data presented here not only advance the science of I-O, but also support near-term commercial opportunity. And with that, I'll turn the discussion back over to Tim.

Great. Thanks very much, Chris. Drew, if we can go to our first question once you've compiled the roster, that would be great. It might take a minute to get this compiled. If we can go to the first one when you're ready, that would be great. Thanks.

Our first question is going to be from Geoff Meacham from Bank of America.

I just have a couple for you on melanoma. On the development side, are there lessons to be learned from mutations in LAG-3 as a resistance mechanism? I'm just trying to think of indications beyond what you've highlighted. And then from a commercial perspective, how do you think about the positioning of who is more optimally suited for nivo ipi versus nivo rela?

So maybe, Geoff, thank you. I can start and then certainly, Chris can chime in on the commercial front. On the mutations of LAG-3, one of the strengths that we carry at BMS is because of the conduct of the studies and collection of tissue and materials over the years from our various trials, we have the ability to now mine the data, go back and take a look. Also, as you saw in the presentation this time, we did have the preliminary data from our LAG-3 expression perspective. But now we can go back and dig deeper into the correlation, as you said, of where patients may not have responded because they carry the LAG-3 mutation. But those are data that are still to be produced. We still need to dig deeper into it, things that will follow in the future. That will also, as you very correctly said, will help us in our pipeline as well as we bring more treatments into the melanoma landscape. But today, we don't have the data to really comment on.

And on the commercial side, what I would say, Geoff, is first, we continue to believe, as I mentioned in the presentation, that the initial opportunity for rela plus nivo is going to be against the roughly 30% of the business that is currently I-O monotherapy today. Given the data that we presented at ASCO, we think that's the most obvious near-term opportunity. In terms of how physicians are going to think about ipi nivo versus rela nivo, I think there will be some physicians who may wish to avoid some of the toxicities that you've seen with ipi. So they may choose to use the rela nivo combination for those subset of patients. However, what I would say is that given the considerable long-term OS benefit that we've seen with ipi + nivo, and as we said in the presentation, we are going to need to see the rela nivo data continue to mature to better understand exactly which portion of those patients are going to be preferred for physicians to go with rela nivo versus ipi nivo. And then, of course, longer term, as we mentioned, there may be additional opportunities in that BRAF mutant population.

Our next question is from Greg Gilbert from Truist Securities.

Can you hear me okay?

Yes.

Great. 2-parter. First, can you speak to your confidence in filing with PFS data? And then secondly, I realize it's somewhat early days for LAG-3 as a mechanism, but do you see the potential for differentiation between LAG-3s that does not seem to exist in the sort of PD-1-versus-PD-1 world? I'm talking about drug versus drug here not driven by a clinical trial strategy.

Sure. Thank you. On the first question on filing with PFS, so PFS is an accepted primary endpoint from a regulatory perspective, and we certainly look forward to sharing the data with the regulatory authorities and using that for our filing. We certainly are continuing to follow patients for overall survival. As I said earlier, we at the current time remain blinded, and we'll supplement that data in the future as that becomes available. On the potential for differentiation versus other LAG-3s, I think it's early days. We don't know what others have that might be different. But what we do have at the current time different is the fixed-dose combination with Opdivo. And we are going to continue to pursue that. As you know, we have the Phase II studies ongoing in non-small cell lung cancer, hepatocellular carcinoma and then soon to start the adjuvant setting in melanoma, and we'll continue to pursue that fixed-dose combination. But in terms of differentiation, I think we have to see where the data evolves and how we then are able to compare our data to the others in the same indications to see if there are any differentiations to be really followed from that perspective.

The next question is from Chris Schott from JPMorgan.

Great. Just 2 for me. First, can you just talk a little bit about the adjuvant data in RCC we saw from Merck this weekend and what that means in terms of the size of the frontline metastatic market as we think about Bristol's portfolio positioning there? And then my second one was on LAG-3 and just some of the newer indications. Can you just elaborate a little bit more in terms of what you're seeing in the basket studies and the scientific rationale for LAG-3 in HCC and non-small cell versus the signal you're seeing in melanoma? I'm just tying to get a sense of your confidence in those additional indications versus the signal that we've now kind of confirmed in melanoma.

Sure. Maybe I -- yes, go ahead. Go ahead, Chris. I'm sorry.

Yes. So as it relates to the adjuvant RCC data, I mean, obviously, it's a little bit early to comment specifically on how that market is going to evolve. I think what we're going to have to wait and see is exactly the percentage of patients who get responses when -- if and when the regimen is approved and then what exactly those responses look like including their durability and how fast they occur. One of the things that we have learned from our experience in melanoma in the adjuvant setting is that, in general, physicians do view these as 2 distinct segments. And so we are seeing real-world data where patients who have been exposed to an I-O therapy in the adjuvant setting and, unfortunately, if that patient does progress into a metastatic setting, that those patients are then again considered eligible for I-O therapy. And we've seen that in other non-I-O indications as well, for example, in the breast cancer setting. So I think it's a bit early to give a precise answer to the question. But as where we sit today, we don't see any obvious near-term impact on our first-line metastatic business as a result of these data in the adjuvant setting.

Yes. And in terms of the LAG-3 indications and why we are pursuing the non-small cell lung cancer hepatocellular carcinoma indication is some preliminary data that we saw in the Study 020 for LAG-3 when we were looking at combinations are responses that we saw in non-small cell lung cancer as well as indications of activity in hepatocellular carcinoma. But I think we need to certainly delineate the activity due to nivo or what could be driven through nivo rela. And that's why if you look at the trial design for non-small cell lung cancer, we're looking at nivo rela chemo versus nivo chemo. That serves 2 purposes. One, defines the dose at a higher dose that are being tested in this particular study; and second, give safety; and third, gives us a good insight into the contribution of components that we don't then have to do in a Phase III trial that we will pursue further. So those are the reasons why the trial design is the way it is and the basis of pursuing based on the prior indication and prior data that we saw in the earlier trial, which was a nonrandomized study. And that's why our randomized Phase II study we thought was important to garner that confidence before we start a Phase III.

The next question is from Matt Phipps with William Blair.

It's nice data at ASCO overall. I was wondering how are you all thinking about the potential of the Opdivo-Yervoy combination in esophageal based on the 6-grade trial, maybe not as impressive as a vivo chemo, but there did seem to be some physician interest and then chemo reaction. And then as far as LAG-3, this is something that I guess from an investor standpoint, the initial data was not really impressive enough for many to look at it too much. But you obviously moved forward and had a bit of a gated development to get to this point. Is there any learnings from the LAG-3 experience that can help deal with any of the earlier I-O combo pipeline?

Samit, do you want to start and then I can...

Sure. I can certainly start from the second question, LAG-3 first, I guess. So there are 3 elements that one has to really look into and not just from the melanoma perspective, which of course is a big deal, as Chris pointed out earlier. If you think about today, many of the patients are not getting treated with a doublet I-O. 1/3 of the patients are not even getting the I-O background therapy for their melanoma. So generation of a Phase III trial data, to us, does seem quite a game changer, I would say, because look at the competitor arm, which is single-agent PD-1, which is a very active therapy. So showing more than doubling of the progression-free survival in the Phase III trial is, to me, I think pretty impressive from a patient and physician perspective. The second part of LAG-3 is if you look at the totality of the data that have been presented at ASCO not just by us but shared by others as well in other tumor types, it starts to tell you a story that LAG-3 does have a place, but we just need to ensure the right combination and the right patient population or the right indication needs to be pursued. We've seen the data in melanoma. We -- I talked a little bit earlier about what we saw as hints of activity in non-small cell lung cancer, hepatocellular carcinoma, and that's why we've initiated those trials. We saw some data at ASCO in colorectal cancer in the MSS type. So I think in totality, there is activity but LAG-3 will require a little bit of a finesse in terms of choosing the indications, choosing the combinations and how we proceed in the right tumor types and the right indications. On the esophageal side, O + Y, I think you already mentioned and from a development perspective, I would say, we are very proud that we've now got 3 different trials showing the activity of Opdivo and Opdivo + Yervoy in the upper GI malignancies from -- starting from the adjuvant setting to the first-line metastatic setting. As you very correctly said, not all patients may want chemotherapy. Not all prescribers may want chemotherapy. So there is space for that O + Y combination, once approved, to be able to deliver for patients an additional effective therapy that has shown an improvement in overall survival. I don't know if there's a commercial perspective to that. Chris, you would like to add.

We're obviously, as Samit mentioned, very pleased to have yet another positive trial here, in this case, for patients with squamous cell carcinoma. Obviously, we're pleased to have 2 regimens that we could potentially offer to these patients. As Samit mentioned, we have heard from a number of physicians we've shown these data to. And you've heard it a little bit even in the ASCO presentations that there will be patients who would like a chemo-free option and obviously Opdivo-Yervoy provides that. And it certainly provides optionality for physicians to target the regimen that they wish to choose to the patients they see in front of them. And along with the approved indications that we have, this gives us the broadest opportunity in first-line metastatic upper GI cancer. So obviously excited about it. And once it's approved, we very much look forward to being able to launch it from [ commercial ].

The next question is from Dane Leone from Raymond James.

Congratulations across the entire set of updates that your team's had at ASCO, great outcomes across the board. I want to ask what some of the feedback has been at ASCO today. We actually had a session that did focus on some of the new BCMA therapies for multiple myeloma. To date, how many treatment centers do you have up and running that have access to Abecma? And what's been the feedback in terms of getting patients on Abecma as a commercial product versus enrolling patients in some of these other BCMA-directed clinical trials that are ongoing and ramping?

Well, let me say at the outset that the 2 CAR-T therapy launches are both going very well from a commercial and medical standpoint. We have over 60 accounts that have been activated with Breyanzi and we're certainly increasing the number of accounts that are activated to treat with Abecma. So I think that in general, on the Abecma side, the feedback from accounts has been very positive about the profile. The demand for the product has been very significant, and in fact, the demand for Abecma is pushing the boundaries of our ability to supply at the moment. We're obviously doing everything we can to continue to ramp up supply in order to meet the demand that we're seeing from these patients. As you know, it's the first BCMA cell therapy here. We've demonstrated a significant improvement in the available options that are available to these patients, and the feedback from physicians has been very positive and there's a significant desire to get patients onto therapy. And so we're very much in the road right now of continuing to ramp up supply to be able to meet this demand. And we continue to work very closely with the customers. But what I would say at the outset is that the feedback has been very good from a commercial and medical standpoint. And our focus is on continuing to push forward both this launch as well as leverage the broader platform that we have in cell therapy with the launch of Breyanzi as well.

The next question is from Steve Scala from Cowen and Company.

The addition of relatlimab appear to have most pronounced benefit in PD-L1 low-melanoma patients. Does Bristol expect to label along these lines or used to be stratified by biomarkers? That's the first question. Second question is Roche is developing a PD-1 LAG-3 bispecific, and they cite some theoretical advantages of that modality. What are your thoughts on that modality? And are you pursuing a similar target?

Sure. So on the first one, Steve -- Samit here. On the first one around the label specified by subgroups, we obviously will not comment. The way I look at the data, the way we've looked at the data and the way the physicians and presenters have looked at the data, across the board, there is a benefit for the patients. If you look at the hazard ratios and how the overall forest plot looks like, yes, the -- you -- from a naked eye view, it may seem as if there is a difference. But is that difference really meaningful? We don't believe so. So we'll continue to pursue our continued dialogue with the regulators in terms of seeing where it goes. But certainly, we see the result equally applicable to all patients at this time. On the second question around PD-1 LAG-3 bispecific from Roche, we'll have to see what the data would tell when they come through. We have the fixed-dose combination that we are pursuing. Now of course, our overall oncology pipeline is filled with other medicines that we are continuing to pursue. Beyond the LAG-3, as you know, we have the TIGIT, we have the TGF beta, we have the IL-12, IL-12 as well as the IL-2 in the Phase III development. So we don't have a LAG-3 PD-1 bispecific at this time in our pipeline, but I think our approach that we've taken, we are quite satisfied and we'll continue to pursue that in the other indications that I spoke about earlier.

We have another question from Greg Gilbert from Truist Securities.

Just 2 follow-ups. On fixed doses in general, can you talk about where you're drawing the line as a company between combination therapies that should be combined in a fixed dose versus those that should be kept apart? Is that simply technical considerations? Or are there other factors that you're considering because there could be presumably an infinite number of potential fixed-dose combinations as you think out over the years? And my second question is a bit more outside the box, I'd say. Are there cases where a competing PD-1 or PD-L1 has a favorable label versus Opdivo, where it would be unethical for you to conduct a U.S. trial to sort of mimic that part of the label, but where you are considering doing such a study outside the U.S. is the way to sort of normalize labeling over time?

Yes. So maybe I can start and certainly, Chris, if you want to add, we can go there, too. From the fixed-dose combination perspective, I think you mentioned one of them, the technical approach, whether the drugs can be formulated to be in the fixed-dose combination or not. And one of the things that also is a driver is the volume of the drug that will need to be delivered, can it be actually given if it's going to be subcutaneous? But of course, IV-wise, can the 2 drugs be formulated? And I think one of the key factors is going to be, if there is going to be a lot of dose changes or toxicity associated with one drug, can you really have a fixed-dose combination when you have to toggle one versus the other drug? So I think both from a clinical safety as well as the technical point of view, one has to keep those things in mind. It could be applicable to the larger part of our portfolio but we have to assess it on a case-by-case basis as we go forward. Right now, we've started with rela + PD-1. On the second question on the label-driven geographic development part, I would say that we are a global company. As we think about the development of medicines, we think about it from a global perspective and more and more so that we include Japan as well as in China as well as many of the other geographies in the overall development of a molecule. We have not devised strategies to shortcut the indications that we are seeking just because we can't conduct the study in the U.S., so we go outside, conduct the study and bring the data to the U.S. We've not taken that approach. We don't think that applies to Opdivo at this time because our labels are pretty solid in either -- in any indication that we have pursued thus far. And we'll continue to pursue the global development that we have ventured out to do at this time. I don't know, Chris, if you want to add anything.

The only thing I would add on the fixed-dose combination, and we've certainly heard this from physicians who have seen the rela data, is that it does offer a convenience element, particularly for those accounts where infusion time is a concern or the availability of infusion chairs. And so in that regard, it potential -- it has the potential to provide a real customer benefit above and beyond, surely that's one of them.

We currently don't have any raised hands. We were just wondering if Andrew Baum from Citi had any questions at all. And if anybody else has any questions, please do raise your hands.

All right. If Andrew would like to ask a question, can you just patch him in and we can see if he has other questions for us.

If he has a question, he'd be able to unmute himself.

I was just in listen-only mode. So forgive me.

Andrew, we can hear you now.

No, I was just -- I was literally just in listen-only mode, so I didn't have any pressing questions. I think you've run through a gamut of them.

Very good. All right. Well, I think that's it for today. Thanks for being efficient, everybody. You know where to find us if you've got follow-ups, and thanks for your time.

Thank you.