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

Good day, and welcome to the Bristol-Myers Squibb ESMO 2024 Investor Event. [Operator Instructions] Please note this event is being recorded. I would now like to turn the conference over to Mr. Tim Power, Vice President of Investor Relations. Please go ahead, sir.

Hello, everyone. Welcome to our investor presentation at ESMO 2024. Today, we're going to review some important data presented here at ESMO, including nivolumab plus relatlimab first-line lung Phase II study and also talk about the progress we've made in advancing our oncology pipeline. I'm joined today by Samit Hirawat, Chief Medical Officer and Head of Global Drug Development. Samit will have roughly 20- to 25-minute presentation followed by availability for Q&A. Moving to Slide 2. I'll refer you to our forward-looking statements. And with that, I'll hand over to Samit to begin the presentation on Slide 3.

Thank you, Tim. It's great to be here with you during ESMO to share exciting data across our solid tumor portfolio. With that, let's turn to Slide 4. At last September's R&D Day, almost exactly a year ago, we outlined our oncology strategy, focusing on extending and strengthening our leadership and durability in immuno-oncology as well as diversifying beyond IO. This weekend at ESMO, we are presenting some exciting data that further reinforce this focus. We will cover most, but not all of what you see on this slide in my presentation today. However, it is worth briefly mentioning some important things. We are the leaders in development of effective IO and dual IO therapy. For Opdivo plus Yervoy, we are presenting 10-year follow-up data in metastatic melanoma that shows transformational change in patients with advanced melanoma. Long-term data further supports use of nivolumab in a subcutaneous formulation. And we are advancing a Phase III trial for a new product, nivolumab and relatlimab in first-line non-small cell lung cancer. Relatlimab HD or high dose will be distinct from Opdualag as it is including a higher dose of relatlimab. Additionally, we are broadening our efforts beyond IO as we are advancing novel and exciting modalities across our emerging cancer research platform. Now let's turn to Slide 5 to further discuss our progress in cancer drug development over the last year. As mentioned, we are extending in IO with a new dual IO combination of nivo plus rela high dose in lung cancer, and we will talk more about this further later in my presentation. We have also filed our applications for subcutaneous nivolumab with respective health authorities, enabling us to extend the benefits of nivolumab starting early next year into the next decade. We look forward to the U.S. PDUFA date at the end of December this year, while the European submission is also under review. We have made significant progress in broadening our reach to new solid tumor targets with 8 new registrational trials underway or plans to initiate in the near future, and we are capitalizing on our differentiated development platforms. Lastly, we are broadening our clinical development into important solid tumors beyond where we have traditionally focused into areas, including small cell lung cancer, breast cancer and prostate cancer. When we put this all together, we have a bright and promising future in cancer research and development and the potential to bring more effective medicines to patients. Now let's turn to Slide 6, which provides an overview of programs across our cancer portfolio. And this slide highlights the breadth and depth of our oncology pipeline, which has evolved significantly even from a year ago. We continue to optimize our pipeline by deprioritizing certain programs and adding new ones that leverage our core strengths, focusing on best-in-class and first-in-class cancer medicines. I have highlighted the key programs we will cover today in purple boxes here. Turning to Slide 7. We'll start with our next dual IO opportunity. Earlier today, we presented late-breaking Phase II data from the relatlimab high-dose program in lung cancer. RELATIVITY-104 was a 2-part trial. Here, we are showing the trial design of Part 2 which was a randomized trial of nivolumab and relatlimab combined with chemotherapy plus nivolumab and chemotherapy. This new regimen includes a higher dose of relatlimab at 360 milligrams combined with nivolumab also at 360 milligrams administered every 3 weeks. The study endpoints evaluated overall response rate and PFS stratified across prespecified PD-1 subgroups and histology. What is most important is to understand here that this Phase II proof-of-concept study aimed to answer two key questions. First, can relatlimab high-dose add benefit to a PD-1 inhibitor and chemotherapy combination. And second, if so, in which subgroup of patients, we believe we have those answers today. Let's turn to Slide 8 and start with the safety summary of the study. As shown here, the safety profile was manageable and very comparable to the control arm, making nivolumab and relatlimab high dose in combination with chemotherapy, a viable treatment regimen for further development. Let's take a look at the efficacy results on Slide 9. The study answered our first question, confirming that relatlimab high-dose meaningfully boosts efficacy when added to a PD-1 inhibitor and chemotherapy and that this benefit was particularly seen in PD-L1 expressing tumors with non-squamous histology. Please note, we did not see a benefit in PD-L1 less than 1% or in squamous histology tumors. Moving to Slide 10. Let's further examine the data in light of our second question on the subgroup of patients. Within the subgroup of PD-L1 1% to 49% expressures and non-squamous histology, we see a compelling efficacy signal. The addition of relatlimab high dose doubled the response rates and improved median progression-free survival compared to IO and chemo alone with the upper end of the confidence interval not yet reached. Even more impressive is how this compares to the data for the current standard of care. Now let's turn to Slide 11 to contextualize these data and how it compares to historical benchmarks. While accounting for differences in study design and patient populations, the RELATIVITY-104 data in the PD-L1 1% to 49% non-squamous subgroup performed numerically better than the KEYNOTE-189 data with approximately 61% overall response rate and 9.8 months median PFS compared to 50% overall response and 9.4 months median PFS seen in KEYNOTE-189. Now, it's important to note the KEYNOTE-189 data shown over here is a 5-year follow-up from that study. PFS at the first readout of that trial was 8.8 months. An important comparison is to the published real-world data and from our own internal analysis, where we consistently see roughly 5.9 months PFS. Comparisons to these historical benchmarks reinforce our confidence to launch a Phase III study in this patient population. Now let's look at the Phase III trial design on Slide 12 posted last month. Our Phase III trial aims to enroll patients with non-squamous non-small cell lung cancer with PD-L1 expression between 1% to 49% in a head-to-head comparison with the standard of care, pembro plus chemo. The primary endpoint of this trial is overall survival, which is the most critical measure for this patient population. This subgroup represents roughly 20% to 30% of first-line non-small cell lung cancer patients. Now let's turn to Slide 13, which highlights why we are also considering a Phase III study in non-squamous non-small cell lung cancer patients with PD-L1 expression of greater than 50%. We saw encouraging clinical results in overall response rates and a median PFS of 13.8 months in this subgroup. These data open a potentially new development path to expand the addressable population of first-line non-small cell lung cancer patients in non-squamous histology. For a summary, of our nivo plus rela high dose program, please turn to Slide 14. Relatlimab high dose has shown improved benefits when added to PD-1 and chemotherapy. We are advancing to Phase III trial in the first-line setting in patients with non-squamous non-small cell lung cancer with PD-L1 expression between 1% to 49% with the potential for further expansion. Of course, relatlimab is not our only program in lung cancer. Let's turn to Slide 15 and look at our development portfolio within the lung cancer landscape. A significant unmet need remains in non-small cell lung cancer despite advancements in targeted therapies and immunotherapies as patients often experience disease progression and relapse. Beyond dual IO, we have multiple new modalities in development focused on improving patient outcomes across patient subgroups. We are now broadening our oncology development into small cell lung cancer. Small cell lung cancer represents 15% of lung cancer with a very high unmet medical need. It is an aggressive form of cancer with a poor prognosis and high rates of recurrence. Furthermore, first-line extensive stage small cell lung cancer represents about 70% of these cases. Standard of care has recently evolved with the addition of PD-1 inhibitors to chemotherapy. However, median overall survival from this combination treatment remains modest at approximately 1 year. Let's turn to Slide 16 and take a look at our anti-fucosyl-GM1, an asset we have not highlighted previously. Combined with nivolumab and chemotherapy, our anti-fucosyl-GM1 monoclonal antibody shows promise in improving patient outcomes. Fucosylated GM1 has emerged as a potential target as it is highly expressed in approximately 50% to 90% of small cell lung cancer with limited expression in the normal tissue. In earlier trials, this regimen showed promising durable antitumor activity in patients with refractory disease. In addition, we see the combination treatment having a manageable safety profile, which allows us to start treatment in the frontline setting, that is from the induction phase, whereas competitor programs are focused on maintenance after completion of induction. On Slide 17, I'll walk through important data highlights from ESMO on the Phase II trial. We evaluated the anti-fucosyl-GM1 combination with nivo plus chemo against nivo and chemo to mirror current standard of care of IO plus chemo for the treatment of first-line extensive-stage small cell lung cancer. In the study, we observed encouraging overall survival benefit. Our Phase III study starting early next year will evaluate this combination against the current standard of care atezo plus chemo focusing on the most meaningful endpoint again of overall survival. Next, Slide 18 summarizes the safety data from the trial. Overall, the regimen shows a manageable safety profile with pruritus or itching as the most common side effect, which usually resolves within 1 to 2 cycles. This drug is just one part of a much larger registrational pipeline in oncology for BMS and let's take another look at our current cancer portfolio on Slide 19. Here, as you can see, we are also advancing new oncology programs across our other platforms, including protein degradation, ADCs, radiopharmaceuticals and targeted therapies. Let's turn to our protein degradation platform and androgen receptor ligand dependent degrader, or AR LDD on Slide 20. At last year's R&D Day, we shared our excitement about the targeted protein degradation platform and its potential. Our first asset is the androgen receptor ligand-directed degrader or AR LDD program, being studied in metastatic castration-resistant prostate cancer or metastatic CRPC. Metastatic CRPC has a low overall survival rate of only 35% at 5 years. The androgen receptor is a key driver of prostate cancer. And AR LDD effectively inhibits the receptor's activity, thereby locking the downstream signaling that promotes tumor growth and survival. In this dual mechanism of action. Number one, the degradation of AR; and number two, AR antagonism results in a deep AR pathway inhibition that proved effective in overcoming resistance mechanisms seen with traditional androgen receptor inhibitors. At a medical conference earlier this year, we shared initial Phase I data and the clinical benefits observed. Here, we see an example of a 69-year-old male patient who was heavily pretreated and showed a compelling PSA response and a confirmed radiographic partial response after treatment with the AR degrader. On Monday at ESMO, we plan to share an important update from this Phase I study. So let's turn to Slide 21. AR LDD has the potential to be an important new option for patients. We are encouraged by these data, which show a PSA response across multiple dose levels and in both AR wild-type and AR mutant castration-resistant prostate cancer. Now historically, metastatic CRPC patients reported median radiographic progression-free survival of approximately 6 to 9 months in chemo-naive patients. However, treatment with AR LDD shows encouraging durability in patients who previously received ARP inhibitors and are chemo-naive. These findings underscore our plans to initiate a Phase III study in 2025. Now let's move to ADC bispecific that we licensed from SystImmune on Slide 22. This ADC is engineered to recognize and bind to EGFR and HER3 on the surface of cancer cells. EGFR and HER3 play critical roles in cell growth, survival and differentiation. By targeting EGFR and HER3, the ADC can selectively deliver the cytotoxic drug to cancer cells, pairing normal cells and reducing systemic toxicity. We are closely collaborating with SystImmune, a Chinese biotech company, on the global study to generate data in a broader population across multiple tumor types. Let's take a look at some recent data generated thus far in Phase I studies conducted in China on Slide 23. In these studies, promising single-agent activity has been seen across tumor types. Data presented here at ESMO covered gastrointestinal and genitourinary indications highlighting expansion potential, especially in metastatic bladder cancer where there was a 75% response rate. Now granted, it is a small number of patients. The data also showed a manageable safety profile with myelosuppression as the most common treatment-related adverse event and very low rates of interstitial lung disease. Outside of this trial, we have seen exciting data emerging from another Phase I trial in breast cancer, and we plan to provide an update at our future medical Congress before the end of the year. This data builds on the initial data sets presented last year. Now with Phase III studies already ongoing in China and with data from these Phase I studies and our global Phase I study, we are planning initiation of registrational trials across solid tumors in 2025. Now speaking of breast cancer, let's switch gears to radiopharmaceuticals on Slide 24 for an overview of this exciting platform. We believe radiopharmaceuticals have the potential to be a transformational modality to treat cancer and RayzeBio, a company we acquired late last year, enables future success as it brings number one, a late-stage asset; number two, an IND engine; and number three, an in-house manufacturing capability, which is critical given the real-time supply chain demands for radiopharma. RayzeBio's approach involves the use of actinium-based radioisotopes conjugated to molecules that specifically target cancer cells. This precision targeting uses an alpha emitter allowing for higher potency and efficacy and reduced side effects compared to traditional radiation therapy. Additionally, a major advantage of RayzeBio is the manufacturing capabilities. The Indianapolis facility is intended to meet clinical and commercial manufacturing demands. We currently source isotope from multiple suppliers but plan to transition to mostly internal manufacturing by Q1 2025. Now let's turn to Slide 25 and review the lead asset, RYZ101. RYZ101 has been in development in GEP-NET or gastroenteropancreatic neuroendocrine tumors, small cell lung cancer and recently metastatic breast cancer indication, targeting somatostatin receptor 2, which is overexpressed on the surface of certain cancer cells, including breast cancer, as shown on this slide. In the lead trial, action 1, specific to GEP-NET early data looks promising in patients who have received Lutathera previously and have disease progression. In a small subset of patients, we see encouraging confirmed response rates with 29% of patients showing a partial or complete response. As regards to safety, RYZ101 is generally well tolerated with no treatment-related discontinuations. The Phase III trial versus standard of care remains ongoing with data expected in 2026. Now let's talk about an exciting new program we initiated with RYZ101 in metastatic breast cancer on Slide 26. In an independent study, RYZ101 was leveraged as an emergency investigational new drug for a participant with estrogen receptor-positive breast cancer. Clinically, because of pleural effusion and metastatic disease, this patient had severe dyspnea, and her disease was heavily pretreated with 11 prior treatments. Looking at the image on the left, this patient showed extensive cancer activity in both PET and CT images. On the right-hand image, this is a remarkable change compared to baseline with a near complete response after 2 cycles of radiopharmaceutical therapy. Now, we do need to interpret these data with caution because this patient did experience disease progression after 2 cycles of treatment. However, the drug did what it is supposed to do and in the recurrent ascitic fluid no SSTR2-positive cells were isolated. Data supporting the expression of SSTR2 in HR-positive breast cancer, and these clinical experiences support the initiation of our new Phase I program, currently underway for patients with ER-positive, HER2-negative metastatic breast cancer. On Slide 27, you can see we have multiple assets in the Rayze clinical program. In addition to our ongoing and new studies, we are expanding the Rayze pipeline and have several assets in preclinical development. Later this year, we will advance RYZ801 targeting GPC3, overexpressing several malignancies, including hepatocellular carcinoma with IND clearance expected by year-end. We are also targeting CA9 over-expressed in hypoxic tumors with a potential lead indication in renal clear cell carcinoma currently in pre-IND stage. Now let's turn to Slide 28, and I will briefly review our PRMT5 inhibitor. While this program is not yet in registrational development, it is an important drug from our targeted therapy pipeline. Our PRMT5 inhibitor is the next-generation oral small molecule, which selectively targets cancer cells with MTAP deletion, a common genetic alteration found in approximately 10% of all cancers. Early Phase I data has demonstrated promising results across a basket of tumor types, including lung cancer, melanoma, mesothelioma and beyond. We have seen a promising efficacy signal and good tolerance with no dose-limiting heme related toxicities like those that have been observed with the first-generation nonselective PRMT5 inhibitors, including neutropenia, thrombocytopenia and anemia. We are currently in dose escalation to identify a recommended Phase II dose to take forward either as a single agent or in combination. We plan to present updated Phase I data at a medical conference later this year, which will be focused on safety, tolerability and updated efficacy of the single agent. Demonstrating meaningful durability of response and disease control rate is going to be important. Additionally, we plan to rapidly advance the clinical development plan sometime next year. Let's turn to Slide 29 to summarize our key clinical development program in oncology. Here, you can really see the breadth of progress we have made in this past year. We have established a strong foundation in IO and continue to expand and diversify with novel agents across multiple platforms. What excites me is that we now have 8 new programs in our plans to enter registrational trials with a study flow of data expected through the end of the decade. And let me now close on Slide 30. In summary, what I would say is we are making significant advancements in the fight against cancer and dedication to pursue better outcomes for patients. We have strengthened our oncology pipeline and continue to execute our strategy of extending leadership in IO and diversifying beyond IO, and we look forward to providing future updates. This is truly a positive and exciting time not only in solid tumor oncology, but also at BMS with a potential for meaningful new treatment options for patients. With that, I'll turn the discussion back over to Tim. Thank you.

Thanks, Samit. And Chuck, if you could remind participants how to ask a question, that would be great and then just go to our first question, please.

[Operator Instructions] And our first question for today will come from Mr. Evan Seigerman with BMO Capital Markets.

Two for me. Well, I appreciate the comparison to the PD-1 chemo in the real world in the non-small cell lung cancer setting, can you just walk me through what gives you the confidence to move into this large Phase II program based on a small subset of patients? And then just a follow-up there, taking a step back -- how do you see this combination of Opdualag plus chemo fitting in with a rapidly developing non-small cell lung cancer space. And I'm specifically referring to some of the bispecific data we saw at World Lung and then today as well at ESMO.

Thank you for your question, Evan. For the first question, what gives us the confidence. Remember, this is our Phase II signal-seeking trial, which set out to ask 2 very important questions. First, to look if rela added on top of the IO plus chemotherapy backbone and very systematically then explore that if it did, what is the next subgroup of patients where the drug works versus where it doesn't work. And there are 3 ways to look at this area. Number one, within the trial is that consistency of the data. And you may not have noticed it, but on Slide 10 of the presentation this morning by Dr. Gerard, we saw the data in the Part 1 of the study. That data then carries through in the randomized portion of the study. So 2 parts. We saw the efficacy in PD-1-positive non-squamous cell lung cancer in Part 1 that carried through in the randomized portion of that same study. Number two, in the randomized portion, we clearly saw the efficacy of rela on top of the IO plus chemo backbone that was superior to what was present in the control arm. Number three, when we look at this data in the real world now, how the range of data for the control arm that is the standard of care today has evolved from the days of KEYNOTE-189 to what today we have seen in the real world is somewhere around 6 months or so. These data bodes fairly well. And we believe that now there is a hypothesis that we need to test in the Phase III study, and we are confident these data put us in that position to be able to test the hypothesis of comparing rela plus nivo plus chemo versus pembro plus chemo. One important thing that the Phase II study allowed us to demonstrate already the contribution of components, meaning rela addition to the IO plus chemo backbone. So therefore, we can now avoid having a third arm in the Phase III trial. So that's for the first part of your question. The second about the landscape evolution in non-small cell lung cancer, especially the bispecifics that you talked about, I think, yes, we acknowledge there is a huge movement and that's great for patients that we are moving the non-small cell lung cancer treatment from target therapies to naked antibodies, to ADCs, bispecifics and beyond. And every one of them will add to the ultimate outcome for patients. It's important to note -- it will be important to understand these data from a geographical application. And especially if you think about the bispecific data that you refer to PD-1 plus VEGF versus PD-1 alone, how that data will evolve and the global studies are conducted because remember, VEGF showed a differential activity when you look at Caucasian population versus the Asian population. And secondly, in population of non-small cell lung cancer below 50% expression of PD-1 that standard of care is not a single agent IO. So I think it will be important to understand all of that. Then of course, we have to keep on thinking about what the durability is. And we saw very importantly what the control arm performance was, of course, of pembro as well as what the durability will be, we will continue to watch out for. So I think it's great, the portfolio is evolving. But as you also saw that what I showed, our portfolio is also evolving. So we are very competitive in the target therapy space, in radioligand space, evolution of our own IO franchise and beyond the IO franchise in lung cancer itself.

The next question will come from Chris Schott with JPMorgan.

Just 2 for me. Just can you talk a little bit about nivo and LAG-3 in the greater than 50% population, is that something we should expect to move directly into Phase III? Or is there more Phase II work to be done there? And should we think about chemo as a piece of that program? Just any additional color there? And my second question, I know it wasn't part of the discussion, but we're approaching the Opdivo subcu PDUFA. So just talk a little bit about just -- how do you think about how quickly we should anticipate that new format ramping? Is that something that could go quickly? Or will that take some time to convert over the franchise?

Thank you, Chris. As always, great questions. So on the first question of nivo plus LAG-3 in the more than 50%, we are actively working on that design of the study. And while today, I'm not ready to provide you the specifics of the population as well as the control arm, but soon we will be able to share that information with you. We are excited about the data that has been generated and looking forward to discussing that very, very soon. In terms of the Opdivo subcu, as we have discussed before, truly excited by the PDUFA date coming up in December towards the end. We look forward to bringing that into the IO franchise beginning of very early next year and extend that benefit to patients all the way through into the next decade. As Adam has spoken about before, we look forward to conversion of patients, about 30% to 40% of the U.S. business to the subcu franchise. And certainly, we have about 4 years to accomplish that, looking forward to that conversion as well.

Our next question will come from Mohit Bansal with Wells Fargo.

Two for me, if I can. Samit, could you please help us understand why -- like we understand the subgroups it worked in, but in the subgroups such as squamous and then PD-1 expression less than 1% where the combination of rela did not work. Could you -- what could be the rationale behind that? And the second part is, is it reasonable to think about combining rela with pembro plus chemo combination given that nivo plus chemo combination also did not fare well in the prospective trial. So -- just trying to get your thoughts on that -- those 2 questions.

So thank you very much for that. So for the first question about why the drug did not work in less than 1% PD-1 expressures are in squamous histology. Look, I think in the previous trials or in the past trials, actually there was a symposia today where Dr. Louis presented some of the data of IO plus chemo in less than 1% expressures of PD-1 as well as in the squamous histologies and the continuous evolution of the data has shown that PD-1 plus chemotherapy generally has had lower activity in squamous population as well as in the less than 1% expressures. Here, we actually, when we look at the multi-weighted analysis, we see similarly. If you combine the two, less than 1% expression and the squamous histology, we do not see that activity. And therefore, we are focusing on the non-squamous and greater than 1% expression, that's where the best efficacy has been observed. I think biology is quite interesting. And I think the Darwinian expression as we say, the tumors evolve and we'll have to continue to find new ways of treatment of these cancers as we understand and learn what we are going to gather from there. From the perspective of the control arm and the combination arm, certainly, there are opportunities of combining -- or there was an opportunity to combining rela, pembro, chemo, but we have our own PD-1 and chemo. And we did show within the trial that the investigational arm certainly performed better than the control arm in the subgroup of interest that we have shown over here. We do believe that these data are very comparable, as we have seen within the trial, and then we take that beyond from both the range of data sets that have been available from KEYNOTE-189 and then beyond in the real-world setting for the control arm. If you think about what nivo plus chemo has shown. And then if you take that data and look at it as it relates to the pembro plus chemo data, I think it fares pretty well in their compatibility. So we are confident in terms of conducting our study with nivo plus rela plus chemo and comparing that versus pembro plus chemo, and that hypothesis is now being tested in the Phase III.

Your next question will come from Matt Phipps with William Blair.

Just curious what was the rationale for evaluating this higher dose of relatlimab in lung cancer versus what you -- it's approved in melanoma and also going with the Q3 week regimen, which does not point that Opdivo is, I believe, currently approved for now.

Yes. Thank you for those questions, Matt. So on the high dose, that's very much based on what we have done in the preclinical space, looking at occupancy and certainly degradation as well as looking at the overall immune cell impact when we combine relatlimab and nivolumab. So we have to look at that data and try to achieve concentrations that are best. And in non-small cell lung cancer, we feel that -- we found that higher doses would be required to achieve an effect as opposed to what we had observed in melanoma. And that's why we did the Phase I study also to look at the safety of the combination of relatlimab. And there, as you saw from the presentation this morning, we did test out even a further higher dose of 720, but settled at the 360-milligram dose over here. We went with a Q3 week regimen over here because the chemotherapy is given every Q3 weeks, so it matches very well with the dual chemotherapy or platinum doublet that is normally used in non-small cell lung cancer, and therefore, nivo, rela will be given along with the chemotherapy within the same regimen.

The next question will come from Trung Huynh with UBS.

Just 2 from me. So first, and perhaps a somewhat unfair question, but you pointed out the real-world setting for Keytruda plus chemo has evolved to be around 6 months versus the KEYNOTE-189 where we see a median PFS of 9 months. So what is actually evolving here? Is it just simply people discontinuing earlier? Is there a discrepancy in the baseline population versus the real world setting? Or is there any other reason that you've concluded? And then just second, in the Phase II, you mandated GCSF administration prophylactically to address the neutropenia. You're also going to do that in the Phase III. Just wondering how much of a burden is this for patients and prescribers and how common is it to administer GCSF in the first line with the standard of care?

Thank you for those questions. Look, as I said, over time, treatment of patients with non-small cell lung cancer continues to evolve. And I think as we think about the data, there's a range of data for IO plus chemotherapy. And what we are seeing now in the real world, and again, this data -- this similar data for OS was also presented earlier today, where IO plus chemo OS data in the real world does not look same at this time compared to what was shown in KEYNOTE-189. So this range of data, as we look at them and again large populations from the data that have been collected in the real world, really support our own data within the trial and our hypothesis that the progression-free survival at this time is somewhere around 6 to 6.5 months. And that's what we're taking as a hypothesis that we are going to be testing in the clinical trials as well. From the growth factor perspective, truly based on the emergence of the data in the Phase I portion of the part 1 of the clinical trial, where we notice neutropenia and the best way to manage that was by allowing the use and then mandating the use in patients who are greater than 60 years of age and that's what is based on the guidelines and that is very commonly done in this particular patient population. So this is not a burden, at least we don't believe it's a burden and the physicians tell us and investigators tell us the same. And so we are carrying that forward in the Phase III trial as well and looking forward to readout of those trials and data.

Next question will come from Stephen Scala with TD Cowen.

What is Bristol's best case scenario for filing rela plus nivo in first-line non-small cell lung cancer and do you think you ultimately need to show overall survival benefit to see substantial uptake? So that's the first question. Second question is Bristol is confident in EGFR HER3 bispecific ADC based on China-only data. Just generally speaking, how would you characterize the reliability of data generated in China and your level of confidence in it being replicated globally?

Thank you, Steve. As is always the case, great questions. For the first one, look, it's an event-driven trial. We are just initiating the centers and beginning the enrollment in the clinical trial. We have not yet said what the filing date would look like. But certainly, over time, we will be able to commit and confirm those dates for you. And overall survival is the primary endpoint. So we are certainly looking forward to showing superiority on overall survival for appropriateness in this particular patient population. For your second question on EGFR HER3 patient population data, we are totally acknowledged that the primary data right now that we have are certainly from China only clinical trials. And certainly, we are doing the global study right now as a Phase I to identify the right dose. If the dose is comparable to what we have seen in China as well. And that emergence of the data is going to continue to dictate how we are proceeding further. Now having said that, a substantial amount of data across multiple tumor types showing single-agent activity in those multiple tumor types. So we do have to take that along. And certainly, as we have learned, many of these ADCs are coming through now that have shown efficacy in the Chinese patient population. And now that efficacy is also translating in the global settings when done appropriately. So we will continue to evaluate that. But certainly, we are looking forward to taking this now into the global trials and the registration trials with more than 850 patients already treated with this drug in China. So we are very confident with the emergence of these data.

The next question will come from Chris Shibutani with Goldman Sachs.

Perhaps if I could ask a question about the MK cooperative PRMT5 inhibitor, you will have some updated data in the second half of this year at this meeting, ESMO, competitor had some data and there are several other players there. Maybe just around that particular approach. Talk about your confidence in monotherapy. Perhaps, is there something that you can help us understand that characterizes the potential differentiation either with your molecule and approach and which particular tumor types might we be expecting the updated data to come when you present it later this year.

Thank you, Chris, for the question. For PRMT5, we are not presenting our data here, but soon to be hopefully and within this year, another conference. We have seen the competitor data. One thing that we can say is what is going to be important is to observe what the duration of response is because stability of that response would be important and the duration of response is going to be important as well as the disease control rate beyond just the overall response rate. And so continue to watch out for that. And I think our data set is going to be much larger than what you have seen or you will be seeing from the competitors. So we're really looking forward to that. Will this drug have a very high response rate as a single agent, I think that is still to be confirmed. But certainly, both approaches will be very important, and I think are interesting, not only as a single agent but also starting to think about combinations, and we are already thinking about those. And soon, we will be initiating combination trials with this agent as well.

The next question will come from Kripa Devarakonda with Truist Securities.

This is Billal, on for Kripa. I had a couple of questions. Do you think the relativity control arm is more representative today's real-world population than the KN-189. And also what did you see in the 720-milligram arm in Part 1 that got you to go down to the lower dose and finalize on that? And then, I guess, for future combinations, what do you think is more prevalent to address here. I mean, how big of a concern is T cell exhaustion with these IO therapy? Would it make more sense to combine with PD-L1/4-1BB like agonistic type thing? Or do you think targeted cytotoxic agent is more of a priority?

Thank you very much. So -- as I've said earlier, I think [indiscernible] that we do believe that this control arm that we used and the data that we've seen within the trial is truly representative of what you see in the real world. By the way, if you look at other IO plus chemo combinations in this space, repeatedly we are seeing a PFS similar to what we have seen in the control arm for nivo plus chemo within RELATIVITY-104. In terms of the dose, we tested both 360 and 720 dose in Part 1 of the study. The 720 dose ultimately ended up being more toxic, and we saw that safety that you saw in the presentations today of the triplet of nivo, rela, chemo was very comparable to nivo plus chemo, and that's why we have taken the 360-milligram forward. And there, we've seen the efficacy already in the subgroup. So now we have the Phase III on hold. For future combinations, I think that one will need to be tested out that beyond the chemotherapy, can we go to chemo-free regimens or not. Our jury is still out. We continue to see the evolution of that data. Some would say that while ADCs are a different way to approach it, but the payload is still chemotherapy. So that would be one approach, whether 4-1BB on top of PD-1s and then attaching them to another PD-1 or another IO therapy will work or not. I think we just have to do the trials and understand if the efficacy stands or not. Thus far, we are far away from that.

Next question will come from Charlie Yang with Bank of America.

My first one is, I'm just wondering like whether the -- the real-world how data set for the NSCLC performed worse than the trial, is that partly due to the fact that they are how in the real-world setting the outperformance score to patients who are on PD-1, and that's how part of that real-world study and that's driving this lower survival data. And then my second one is regarding kind of your anti-fucosyl-GM1 Phase III plan. I was wondering if there's a plan to include how the biomarker as part of the [indiscernible] analysis or if that's even possible, just given kind of the data that we saw was somewhat modest. And obviously, if you can talk about the emerging how trend with [indiscernible] and the B7-H3 ADC in that frontline space as well.

Thank you very much. So our dose -- from a first question perspective, from a real-world perspective, we believe our dose looks pretty good from a PFS perspective, the outcome that we've seen is pretty good, very much in line with what the real world data is. The real-world data also is within the range of data sets that we've seen in the set that we've collected as well as, as I said earlier, all the other IO plus chemo trials that have been done. And certainly, data continues to evolve and has evolved from the days of 189 reporting versus where we are today. We are very confident. I don't think the real world data is worse than the trial. This is within the range and the evolution of the treatment of these patients in the first-line setting on non-small cell lung cancer. In terms of the anti-fucosyl-GM1, certainly we will be collecting a lot of the biomarkers and testing them. As far as the fucosyl-GM1 itself, it is highly overexpressed in about 50% to 90% of patients with non-small cell lung cancer. So to select the patient population will be very difficult. But certainly, we will do the biomarker analysis for future evolution and to understand the activity of the drug if it is limited to a certain patient population or not. One thing that is going to be very important in small cell lung cancer, overall survival is going to be the most important thing. There are no trials that have shown remarkable benefit when it comes to progression-free survival. Our Phase II study certainly put that as the primary endpoint. But the real meaningful outcome is the overall survival, especially as we think about the extensive stage disease and how it progresses. So we have taken that forward. Landscape over there is evolving as well. We are watching the space very carefully. B7-H3 is a good one that you said. And I think it's a good activity that we've seen that far and looking forward to continued evolution of that one. Again, this disease is so fatal and so bad in terms of these unmet medical need that we have that more drugs would be better. And so every target that has activity in this is always going to be welcome. So we're watching the space very carefully on that one. But looking forward to initiation of our Phase III trial early next year.

The next question will come from Sean McCutcheon with Raymond James.

Just a couple for me. Can you speak to the dose flexibility or how you're thinking about the go-forward dose with AR LDD moving into a Phase III program, just given the [indiscernible] the 900 mg BID dose? And how critical do you think getting to that higher dose or some intermediate dose above 600 mg BID will be to get sufficient degradation? And then second question, I would be interested in getting your thoughts on what seems to be a general lack of PFS benefit for the anti-fucosyl-GM1, but a signal on overall survival, what your interpretation on the underlying mechanism is there and maybe why we're seeing this dynamic? And if we should expect to see that be replicated in our Phase III?

Yes. Thank you for both the questions, Sean. On the first one on the dose for AR LDD, very well said I think there is a QTc signal that we have to be aware of. But with that QTc signal, primarily Grade 1 and Grade 2, nonsymptomatic, well managed with those interruptions are those decrease. We did not have any dose discontinuation due to QTc prolongation. With that said, though, it's an important one, and we'll keep an eye on it. As we look towards late phase development of this particular molecule, we are changing the formulation and the formulation will then bring our very manageable [indiscernible] AUC smoothening of the curves. So hopefully, we will be able to better manage the QTc signal over there as well as we look to initiation of these trials next year. For the anti-fucosyl-GM1, as I said earlier, across the board in small cell lung cancer, it has been very difficult to show a progression-free survival benefit. So I think what happens is the biology of the tumor changes with every treatment, and therefore, they add to the overall survival benefit that we have seen even in the earlier trials as we think about the evolution of this landscape. So we have a very strong hypothesis emerging from a randomized trial in the first-line setting that we have now got to do and prove in a Phase III study, then that's what we are starting now. And I think it's an encouraging signal in terms of this disease, which has a very high unmet medical need.

Your next question will come from James Shin with Deutsche Bank.

Seems LAG-3 was assessed at baseline in RELA-104. Did the BMY team do any exploratory look at LAG-3 expression levels and outcomes? And then secondly, Samit, there is some OS data from Part I of RELA-104, seems pretty in line with 189. Any concerns PD-L1 may not be the best biomarker?

Yes, sure. Thank you very much. For the -- James for the first question on LAG-3 expression, we did look at LAG-3 expression. And interestingly enough, the driver really is the PD-1 expression, not necessarily differentiating based on the LAG-3 expression. So we have to continue to monitor it. But within this trial, LAG-3 did not turn out to be a marker of great differentiation in terms of the activity of the drugs that have been used. So we have to understand that even better going forward. So we'll continue to collect that data, and let's see where it takes us. In terms of the data, I think Part 1 data is very consistent that we have seen over to Part 2. In part 1, we did see an overall survival also. But again, these are continuing collection of the data. We had a longer follow-up in Part 1. So we have an overall survival of about 22 months and approaching higher, but we again have to follow these patients longer. For part 2, we only have a 10.7-month follow-up thus far. So we don't have enough information on the overall survival. And as that data matures and becomes available, we'll certainly be sharing that. And overall survival remains the primary goal endpoint that one has to follow in the Phase III study. So we are conducting that study with that intention in mind. What I would say overall with the portfolio that we have in non-small cell lung cancer, we have multiple opportunities to impact the outcomes for these patients. And that is a truly exciting part of conducting drug development at BMS and looking forward to bringing many, many of these updates in 2025 to you.

Thanks, Samit. We're out of the time for today. Really appreciate your interest and participation, especially over the weekend. If you'd like a follow-up, just reach out to anyone in the Investor Relations team. Thank you.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.