Amgen Inc. (AMGN) Earnings Call Transcript & Summary

Hello. My name is Sarah, and I will be your conference facilitator today for Amgen's conference call following the 2023 European Society for Medical Oncology Congress. [Operator Instructions] I would now like to introduce Mr. David Reese, Executive Vice President of Research and Development. Mr. Reese, you may now begin.

Thanks, Sarah, and good morning and good afternoon, everyone, depending on where you're joining us from. We're very happy today to give an update on our oncology pipeline after the ESMO meeting, where we had some major data presentations. We have a lot of materials, so we're going to jump right in, in just a moment. I would note that all of the slides that we're presenting today will be posted on our website immediately following the call. On the next slide, is our standard safe harbor statement. And if we move to the next slide, you'll see the agenda today. I'll give an introduction to our strategy. Just as a reminder, very briefly, I do want to say a few words about BLINCYTO because we continue to generate very nice data with that molecule and reach more patients who can benefit from it. We're then very happy to be joined by Dr. Owonikoko, from the University of Pittsburgh Medical Center to talk about tarlatamab. And Dr. Kevin Kelly from the Sydney Kimmel Medical College, where he is the Chair of the Department of Medical Oncology to talk about xaluritamig. Both of these investigators have extensive experience with these particular bispecifics, and I'm sure you'll have questions for them. Then Jean-Charles, whom you all know, will take you through some of our precision oncology drugs with updates there as well. So just to remind you of our general approach to oncology, it is split between immuno-oncology and precision oncology approaches. Today, we're going to talk about updated data from our T cell engager platform, which I think is showing now real maturation and has reached some watershed moments with demonstration of efficacy in solid tumors, more than one solid tumors. We will also talk about updates on targeting cell intrinsic targets, in particular, AMG 193 and some additional LUMAKRAS data. The oncology portfolio in the 18 years that I've been at Amgen, I think, now is broader and actually more heterogeneous than it's ever been. And to me, the one real takeaway is that I think we have the opportunity to really change the natural history of some of these diseases where there is tremendous unmet medical need. And that's a theme that I'm going to come back to as we move through today's event. Now let me start with BLINCYTO. BLINCYTO has been around now for a decade. And as you can see, it is approved, of course, initially in the relapsed/refractory ALL setting. It is also FDA approved in individuals who have minimal residual disease or MRD positive. ALL after induction therapy, that's a situation where there's a high risk of relapse. We've demonstrated curative potential in that setting. And now with the E1910 data, which I'll remind you of in a minute, which is -- which FDA filing is underway. We will be moving into the first line. And one of the things that we have learned through now a decade of development of BLINCYTO is that efficacy improves and toxicity decreases as you move into earlier lines of therapy, almost certainly related to the lower tumor burden. This makes sense biologically, it makes sense immunologically. But that's a road map that I'd like you to keep in mind as we begin talking about tarlatamab and xaluritamig. I'd also remind you that the Golden Gate Phase III study is enrolling for BLINCYTO, which is a potentially chemotherapy sparing regimen in older adults. Older in this disease is considered 55 years or older where the prognosis is often much more poor. And this is a trial that I'm looking forward to very much. Again, I think we can change the natural history of the disease. In fact, if you look at the next slide, what you can see from the ECOG-ACRIN NCI 1910 trial, E1910 in both patients with MRD-negative disease as well as MRD-positive disease. You see an improvement of overall survival when BLINCYTO is incorporated into standard consolidation regimens for ALL. This is -- this, to me, really points to the direction where this field is moving and an ability to really enhance cure rates. Now here again, we show E1910 on the left, but this has been replicated in the pediatric setting as well as against historical controls in infants with first-line ALL. So again, you're seeing a theme here, increased cure rates when you actually administer the drug in a setting of low tumor burden. So I'd like to leave you with a thought about BLINCYTO. Here you see on the chevron illustrated its development over time. Now as I mentioned, the Golden Gate Phase III study is enrolling, which I think could redefine the standard of care in first-line induction therapy as well as consolidation and maintenance as part of the complex treatment regimens that these patients receive. And then I do want to put on your radar that we are developing a subcutaneous formulation of BLINCYTO. We've reported a few handfuls of patients previously at ASH. We've accumulated a fair amount more data since then, which we're hoping to show later this year or into next year. I think this is very, very promising and may -- there's a possibility of enhanced efficacy, but certainly, patient convenience and reach of the drug should be improved. So I think there's still ways to go with BLINCYTO as we really try to redefine the approach to ALL. Now we're going to switch to our solid tumor bispecifics, starting with tarlatamab. And tarlatamab targets DLL3, as you'll see in a minute, which is almost uniformly present on small cell lung cancer cells. Just so everyone understands the disease, we're talking about small cell lung cancer when it's staged is divided essentially into 2 stages, limited stage and extensive stage. Limited stage means the disease is confined to the chest in an area that can be encompassed by a radiation port and chemo-radiotherapy is the standard approach for those patients. The vast majority of patients, unfortunately, at the time of diagnosis have extensive stage disease and there are roughly 30,000 to 35,000 patients with small cell lung cancer in the U.S., in the major countries in Europe, relatively comparable epidemiology. Standards of care have not changed dramatically over the last 30 years. In fact, the same drugs that I learned as an oncology fellow in training 30 years ago are still employed in the clinic. Checkpoint inhibitors have been introduced. They have produced improvements in survival, but that has been very modest. In the third-line setting, which we'll initially talk about today, there is no accepted standard of care, and most agents have not demonstrated very much efficacy at all. So keep in mind the BLINCYTO paradigm of moving forward to earlier lines of disease as we talk about tarlatamab, but let's start with the foundational data for the program which is the DeLLphi-301 study. And with that, I'd like to invite Dr. Owonikoko to the stage to present the data that we've just shared at ESMO a couple of days ago, Dr. Owonikoko.

Thank you so much, Dr. Reese. Good afternoon, everyone, from Madrid. My pleasure to be part of this. And I want to quickly go through the data for tarlatamab, an immunotherapy agent targeting DLL3 in small cell lung cancer. As you've heard, this is a construct that binds to CD3 on the surface of immune cells and DLL3 on the surface of cancer cells. Vast majority of small cell lung cancer cell will have DLL3 expression on surface, more than 90% with different degrees. And with this binding, that this leads to activation of the T cell and T-cell mediated cell lysis. Next slide, please. The DeLLphi-301 study was designed to test 2 different doses of tarlatamab in patients with extensive stage small cell lung cancer who had received at least 2 lines of therapy. They have good performance status and they have disease that we can measure and follow to estimate the efficacy of interventional agent. A lot of patients have brain metastasis and this was allowed in the study, it's been treated and stable. 32 doses tested were at 10 milligrams and 100 milligrams of tarlatamab. This was given on day 1 with a small dose, followed by the definitive dose on day 8 and then subsequently on day 15 and thereafter every 2 weeks. The primary endpoint for this study was overall response rate. And other important endpoint that we look for is duration of response as well as other measures of activities such as overall survival. Next slide, please. Majority of the patients on this trial, which was more than 200 patients across the different parts of the study were male, 70% of them. Median age is comparable to what we see in patients. And if you see here more and sort of the patient that received at least 3 lines of therapy. This is a population where we currently don't have any standard of care. Important to also note that majority, about 70% of these patients received anti-PD-1 or PD-L1 as part of their frontline regimen before coming on trial. DLL3 expression was assessed and was present in more than 90% of the patients. Next slide, please. So looking at the 2 doses of tarlatamab tested on this study, there was definite evidence that this agent has activity. Our objective response rate was 40%, while the lower dose of 10 milligrams, and 28% at the higher dose of 100 milligrams. We did see patients with complete responses with both doses, especially the higher dose of the drug that was tested. So with this evidence, we -- I think it's safe to conclude that we have preliminary evidence that tarlatamab at the 10 milligram dose, especially would be efficacious in heavily pretreated small cell lung cancer based on this objective response rate of 40%. Next slide, please. And this is our one of our plot showing individual patients on trial. What is most noticeable here is the vast majority of patients, even if they did not meet the objective response criteria showed some reduction in tumor burden, whether you look at the 10 milligrams or the 100 milligram dose, and this was regardless of the level of presence or absence of DLL3 expression. The red diamond signs there showed those without undetected DLL3, where we still saw evidence of too much shrinkage, including patients treated at 10 milligrams to add objective response by resist criteria. Next slide, please. This is what we call the [ DOR ]. The [indiscernible] tells us for those patients who had response, how long did the response last for. Number 1 seem to notice a patient that had responses, it occurred very, very early. The time to response was about 1.4 months in terms of median. And the duration of response when measured in terms of median has actually not been reached. Majority of the responders continue to benefit more than 6 months from when they achieved a response and 56% of the responses were still ongoing at the time the data was summarized for this ESMO meeting. Next slide, please. One thing about small cell is we've had agents in the past that will induce responses but those responses don't last, and oftentimes, they do not combat into other efficacy significant progression-free survival and overall survival. What is particularly encouraging with tarlatamab is defined by the responses that we are seeing also converted into signal in terms of PFS at 6 months with more than 40% of patients achieving progression free survival at that median time -- 6-month landmark time point. And most impressively is the overall survival. In the third line setting where we do not have anything to really treat patients effectively. Majority of the patients do not last very long. It's usually measured in a matter of months. If anything, you treat the patient with work or doesn't work, a majority of our patient, unfortunately, do not do well at this line of therapy. While this is still very, very early, I think this early data is quite encouraging that 57% of patients treated at the 10 milligram dose and more than 50% of the patient treated with the 100 milligram dose were still alive by the time of the data summary for the meeting. Next slide, please. That's always concerned with this immune-mediated strategy, whether or not patients can actually tolerate them. Vast majority of patients treated on both doses experience some type of adverse events. As would be expected, majority of these were immune-related sort of, what we call CRS, which is more or less on target. In fact that we see when we use this type of construct. Grade 3 or higher toxicity were much less frequent. And dose reduction or dose interruption was infrequent. If you look at the 10 milligram dose, only 40% of patient had their doses reduced or interrupted, and a slightly higher proportion at 29% at the 100 milligram dose. Speaking specifically to the CRS, about half the patients on the 10 milligram dose experienced some type of CRS, mostly Grade 1 and 2, no Grade 3. And with the 100 milligram dose, above 55% had grade 1 and 2 while 6% experienced grade 3. These were mostly easy to manage. Majority of this patient, as I highlighted before, did not have to permanently discontinue the treatment. And there's ongoing efforts to see whether or not this regimen will be amenable to outpatient use like what we did when we first started where we are to admit patients into the inpatient setting to administer the drug. Next slide, please. With respect to the cytokine release syndrome or CRS, we are particularly interested in whether or not there is neurotoxicity the ICANS. This is a subset of CRS that could be life-threatening. Two things that I want to quickly highlight here is the observation that CRS was likely consigned to the first 2 doses that the patients got on trial. So when we look at the 10 milligram and the 100 milligram dose, majority of the CRS was observed with the first 2 doses by the patients were given. Once you pass the first 2 doses, the rate of CRS overall went down. More importantly, ICANS, which is a much more concerning type of CRS was quite infrequent, less than 6% of patients across all the doses and time points experience any type of ICANS event. And when you look at those that had it, it was predominantly observed with the tarlatamab at a much higher dose of 100 milligrams. This was also reversible with the use of tocilizumab. Some of the patients has requirement in terms of additional intervention for CRS of using oxygen and vasopressor support in [indiscernible] real instances with 1 or 2 patients. Next slide, please. So to conclude, based on the aggregate data that we have now about tarlatamab, especially the 10-milligram dose and the durable anticancer activities. The safety is very encouraging. A lot of what we saw with CRS that were easily manageable with the overall response rate of 40% in this third line setting or higher. This is really, really meaningful for our patient population, and this is a durable response that is lasting more than 6 months in more than half of the patients. The most common treatment associated adverse events was the CRS, occurring mostly with the first two doses in cycle 1 and generally manageable with supportive care. Discontinuation was very low at 3% with tarlatamab, while we continue to use in-patient monitoring for the first cycle, there is ongoing effort to study the safety profile without the need for inpatient monitoring at all or maybe a shortage duration of inpatient monitoring as we continue to develop these agents. So at this current point, tarlatamab at the 10 milligram dose has been selected for further development. And I think result overall, we'll support the use of these agents in previously treated small cell lung cancer. Now I'll be happy to take questions later on. Thank you.

Thank you, Dr. Owonikoko, that was great, and we'll come back to you, of course, for questions. I just want to end this section by returning to the development program for tarlatamab, DeLLphi-301, which you just saw, we actually have begun after discussions with the FDA submission and as these data continue to evolve, we will update you on that potential registration pathway and what time lines may look like. But that's something, obviously, of great interest to the field. In addition, the DeLLphi-304 study, which is a second-line trial head-to-head against standard of care chemotherapy, with overall survival as a primary endpoint is now briskly enrolling. And we have studies planned in the first line on induction maintenance setting as well as in the setting of limited-stage disease, again, returning to the BLINCYTO paradigm and moving into earlier lines of therapy with lower tumor burden. We anticipate that within the next 6 to 8 months, we will launch the studies in earlier lines of therapy, first-line induction maintenance and limited-stage disease. So this is going to be a comprehensive broad-based development program based on the clinical profile that we've seen to date. Now we have a second BiTE with activity in solid tumors as well, xaluritamig, and we'll move into that in a minute. Unlike small cell lung cancer, the treatment landscape for advanced prostate cancer is much more complex. Patients often when the disease progresses beyond local therapy, they receive a variety of hormonal therapies. Then ultimately, potentially chemotherapy and now radioligand directed therapy as well. There still remains substantial unmet medical need across that spectrum. The disease cannot as of yet be cured. And the question is how to best position xaluritamig in this complex treatment landscape. I do believe we have the foundational data to begin thinking about that. And I would now like to ask Dr. Kelly to present the data that he shared just a few days ago in Madrid at the ESMO meeting. Kevin?

Yes. Thank you, David. Good morning, good afternoon, everybody. It's my pleasure to present some of this data that we just recently presented at ESMO. And this was in patients with advanced prostate cancer targeting STEAP1 which is a cell surface antigen, which is highly expressed in prostate cancer and is associated with poor survival. Important to note, it is highly expressed in prostate cancer, but minimally expressed in other normal tissues. Often I'm asked, is the differences between the PSMA expression in STEAP1 again, is that STEAP1 is more homogeneously expressed throughout tissue in the higher levels and it doesn't have the off-target effects that PSMA may have also. So this makes it a very good target for immunotherapy. Xaluritamig is what we call XmAb 2+1 T cell engager designed to facilitate T cell-mediated lysis with STEAP1 expressing cells is unique because it has specificity to STEAP1 and also attract the T cells through the CD3 component. And in preclinical studies, xaluritamig showed a broad anticancer effects, both in prostate cell lines and xenograft models. So it was moved into early phase development. Next slide, please. And this -- we presented the first global -- first-in-human open-label study in patients with advanced prostate cancer. The primary objective of this study was really to understand the safety and tolerability and find the maximum tolerated dose of xaluritamig. Secondary objectives included looking at the pharmacokinetic profile but also look at the preliminary anti-tumor activity. And it's just important to note in a Phase I study like this is that -- usually, you see responses in less than 10% of the patients. Other exploratory objectives include progression pharmacodynamics and immunogenicity of the drug itself. Patients that were included, where patients with metastatic cancer, resistant prostate cancers. These are patients who have progressed on most of the standard therapies, hormonal therapy, chemotherapy and the novel anti-antigens such as enzalutamide/abiraterone. This was a dose exploration, which means that we started a low dose and increase the dose over time. We went through 12 cohorts to really define the optimal dose board, which actually included giving premedications, hydration and also doing what we call step dosing. And we finally discovered the safe dose or the tolerable dose in this population, which was a 0.1 milligram on day 1 cycle 1 to 0.3 milligrams on day 8, day 15 was 1 milligram, and day 22 was 1.5 milligrams. These were given with premedications and it was well tolerated. The patients that enrolled here was typically what we see in a Phase I study like this. It's important to note these were very heavily pretreated patients. Number of prior lines of therapy was an average of 4, 27% of them had 5, 85% of them had prior chemotherapy. But most important is this is large proportion of these had visceral metastasis 53%, but of those, 37% of them had liver disease, which is the most aggressive component of this advanced disease. Next slide, please. Again, just get to the point here, this is what we call a PSA waterfall plot, which actually shows the decline of PSA, which is the marker we used to see the clinical activity in patients. We looked at it in 2 ways: high dose and low dose cohorts. The high-dose cohorts was above 0.7 milligram, which was a therapeutic levels that we saw preclinically that we suspected versus the low dose. Important is that we saw responses in all cohorts across the board. And again, that it was high, response rate was higher in the higher dose cohorts. When we look at the PSA, less than 50% was done in the local cohorts, around 40% in a high-dose cohort was 59% -- but patients whose PSA went down more than 90% was 19% in the low-dose cohort and 36% in the high-dose cohort. Very impressive PSA responses for this very advanced population. Next slide, please. Again, so we also looked at patients with measurable disease on the CT scan. And again, that we looked at high dosing, low-dose cohort, and again, is that partial responses for overall was 24%, 3% in low-dose cohort, 41% in the high-dose cohort, which again, is very remarkable for this population. Next slide, please. Again, it's very much like tarlatamab. Responses occurred very quickly. If you look at the blue stars there, that's when you get the PSA 50 response and PSA 90 is as you can see it there, is this is on the purple triangles. It usually occurred within the first 4 weeks. Again, that the patients have been on a long -- some patients have been on it long time. It's a little difficult in the Phase I study because we are pushing some patients to toxicity. But the 19 patients from the high-dose cohort made on treatment at data cutoff, those 13 patients from the high-dose cohort remain untreated for over 6 months. If you look at those patients who actually had measurable disease and confirmed partial response, the median time with 9.2 months. That's also a very, very encouraging data there. Next slide, please. This is just an example of a patient who responded. On the left is a 65-year-old who was heavily pretreated with metastatic castrate-resistant prostate cancer. Again, what you see in the liver on left large metastatic lesions into liver after several cycles. These have virtually gone away. And this correlates with this PSA dropping down to less than 90%. On the right-hand side is what we call PSMA PET scan. What you can see is diffuse disease, not only all the way through the liver, but throughout the bones. And what you see on the right is this a complete resolution of the liver and most of the bone disease. Again, is that this was also confirmed the PSA90 response. It's impressive here because very rarely we see responses in the bone. And here, we are seeing responses in bone, which is also a very encouraging response. Next slide, please. Toxicity was overall manageable. The major toxicity, we did see was low-grade cytokine release syndrome. We did have -- initially early we had some grade 3. But with the adjustment of the premedications, afterwards, we did not see any grade 3 CRS. We did see fatigue and anemia, which was related to more to the disease. And again, that's really no grade 4 or 5 events. We did have treatment-related musculoskeletal connected disorders, reported 14% of the patients. Again, etiology is not known since STEAP1 is not known to be expressed highly in the muscle. So that's ongoing investigation to better understand these musculoskeletal and connective tissue disorders. Next slide. So overall, xaluritamig is the first T cell engager, STEAP1. The MTD was established with a step dosing premedications. The safety profile overall was manageable with CRS generally low grade, and we saw principally only in cycle 1. We saw very encouraging antitumor activity in heavily pretreated patients with metastatic castrate resistant prostate cancer. Again, PSA90 response in the high-dose 36% with overall RECIST criteria of 41%. And again, ongoing is dose expansion and optimization are currently ongoing, both as a monotherapy and combination. Again, this abstract actually really -- or the study really gave a lot of buzz around ESMO because -- this is really the first T-cell engager that has shown activity and its first-in-class in prostate cancer in a tumor that they -- most people thought would be a cold tumor and could never be targeted like this. So I think it really breaks a new barrier for immunotherapy in prostate cancer and other solid tumors also that there is really hope that these types of therapies are going to really be meaningful and make a great impact on patients. And I'll turn you back to Dave, and thank you for your attention. I will take questions later.

Great. Thank you, Dr. Kelly. I'm sure there'll be plenty of questions. So if we look at the plans going forward for xaluritamig. As Dr. Kelly mentioned, we're rolling dose expansion now to generate additional efficacy and safety data. In fact, we have relatively large cohorts, and I'm pleased to announce that we've completed enrollment very quickly in the dose expansion cohorts. We will be examining those data in through next year to further define the development plan. But again, if you think about the -- what I think of now is the standard paradigm of trying to move towards lower tumor burden settings. Of course, we'll start in later lines of therapy, chemorefractory setting, but also the chemotherapy on naive setting in patients with advanced castrate-resistant prostate cancer. We are enrolling cohorts now in combination with various hormonal therapies, and that will form a foundation for treatment in earlier lines of therapy in combination. And then ultimately, I think the real chance for this molecule should we continue to develop efficacy and safety data that resemble the current clinical profile in settings like biochemical overlaps or earlier stages of disease, which can be fought almost as an analog of minimal residual disease in ALL or a post-induction therapy setting in small cell lung cancer, where you've got minimal tumor burden. This is where I think we really have a shot to change the natural history of the disease. So while it's a little earlier here, I think we again have the foundational data, and we're very eager to move this molecule forward in a broad development program. Well, now let's turn to precision oncology, and I will now turn things over to Jean-Charles, who you all know, who will talk about some of the updates again that we've just presented within the last week or 10 days, and then we'll open it up for Q&A. Jean-Charles?

Yes. Thank you, Dave, and good morning, good afternoon to all of you. I would like to share an update on AMG 193, next slide that we share as our presentation a little over 2 weeks ago at the triple meeting AACR-EORTC-NCI (sic) AACR-NCI-EORTC. AMG 193, an oral MTA cooperative PRMT5 inhibitor that demonstrates synthetic lethality in MTAP-deleted cancers. A PRMT5 is a master protein regulator and enzyme that is implicated in cancer development to a great variety of mechanisms. And that -- it's important to remember that the development of first-generation PRMT5 inhibitors was not highly successful. And therefore, I want to draw your attention to the very specific nature of this oral compound. This is a molecule that will only have activity in the specific setting of MTAP gene deletion. MTAP gene deletion will lead to an accumulation of MTA which per se partially inhibits PRMT5. In the presence of MTA, interacting with PRMT5, AMG 193 will be able to bind and therefore, lead to a full inhibition of PRMT5. Next slide. MTAP deletion is a frequent event in the setting of cancer. Its prevalence is around 15% of cancer. So that represents over 100,000 annual MTAP new patients in the U.S. and probably an equivalent opportunity in Western Europe. The top 5 tumors contribute to overall -- 70% of overall MTAP new patient incidents. And you see these are different tumor types where the prevalence is above the 20%. MTAP deletion is detected through NGS and is routinely reported in various tests that have been approved including FoundationOne and Tempus results. If we move to the next slide, this is a busy slide that represents the dose escalation of an ongoing first-in-human study where AMG 193 is administered orally once daily with 28-day cycles of continuous dosing. I would like to draw your attention to the key inclusion criteria, which is to be included on the basis of a locally determined MTAP deletion by NGS or centrally determined anti-protein loss by immunohistochemistry. MTAP deletion and CDKN2A deletion coexist, that's why we also allow CDKN2A deletions. Well, we have enrolled 48 patients in 7 scale [ 18 ] course of once daily 193, ranging from 40 milligrams to 1,600 milligrams. Out of these 48 patients, 18 patients have been treated at active dose levels above 800 milligrams. Let's move now to the next slide. The activity of AMG 193 in the active dose levels above 800 milligrams is summarized in this -- well, it's okay. Let's stay on activity, and I will go to safety later since we've already jumped there. So in terms of activity, this is, as I was saying, that 18 patients treated above 800 milligrams. And what I'm showing here is a 3D waterfall plot, where you can see both the best change of lesions in diameters and the duration of the activity of the drug. Well, it's very interesting to highlight that 5 out of these 18 patients have partial responses in 5 different tumor types. But I would like here to draw your attention to the 3 patients who have non-small cell lung cancer and have been treated at active dose levels. We have one patient who progressed, but we have another patient who is a stable disease. This patient with stable disease has clear tumor shrinking that is amplifying over time. And I'm happy to report that the second patient that was a stable disease when we presented this data in Boston has now achieved a PR. If we move now to the toxicity slide. This is overall a drug that had treatment-related adverse events in around 81% of the patients. They were mostly grade 1 and 2. The most common treatment-related adverse events were nausea, fatigue, vomiting and decreased appetite. And consistent with the tumor selective mechanism of action, there was no clinically significant myelosuppression, which was the main drawback and problem of first-generation PRMT5 inhibitors. If we move to the next slide, which is the summary slide. AMG 193 is an MTA cooperative PRMT5 inhibitors that targets MTAP null-solid tumors. This is a reality that is prevalent in around 15% of solid tumors. We are encouraged because we are seeing now activity across 6 different tumor types, including non-small cell lung cancer. Dose-limiting adverse events and treatment discontinuations were typically due to clinically manageable GI events, and we see a favorable therapeutic window and a clear opportunity to explore potential combinations. We have an ongoing dose expansion cohorts, which will inform the future development strategy. So if we move now to the second precision that's molecule that we have and for which we have data ESMO that's LUMAKRAS. So next is, we are encouraged by LUMAKRAS data that is [indiscernible] in colorectal, and I would very briefly touch on non small cell lung cancer. So for colorectal cancer, I will share with you some of the key slides of our Phase III trial CodeBreaK 300, which combined LUMAKRAS with Vectibix in the third line setting, but I also want to remind you all of the very encouraging Phase Ib data of LUMAKRAS plus Vectibix plus with FOLFIRI that we have already shown to all of you. In non-small cell lung cancer, we have promising Phase 1b data of LUMAKRAS in combination with chemotherapy in chemo-naive in KRAS G12C-mutated non-small cell lung cancer that we have shared at ASCO, at the [ world ] lung cancer. Next slide. But let me focus now on CodeBreaK 300. This is a trial that feature 3 arms evaluating sotorasib at 2 different dose levels plus panitumumab. And it was then compared to investigator choice of trifluridine/tipiracil or regorafenib, which are standard of care. If we move to the next slide. The baseline characteristics between the 3 arms were generally well balanced. And I want to highlight that 15% of patients received 1 prior line of therapy, whereas 85% received 2 or more lines of therapies. If we move to the next slide, this is the primary endpoint, progression-free survival in the intent-to-treat population. And it's very clear that the median PFS was improved from 2.2 months in the investigator choice for the median to 3.9 months at the 240 milligrams and 5.6 months at 960 milligrams reduction in the risk of progression instead was 51% at a 960 milligram. This is something that was demonstrated by PFS that was evaluated by Blinded Independent Central Review. And it shows something that is clinically meaningful for these patients. Next slide. If we look at the forest plot, the benefit on terms of PFS on the left for the 960 milligram and on the right for the 240 milligram combination is observed in fact, in every single one of the subgroups, and it's observed both at 960 and at 240 milligrams. Even if its nature and the hazard ratio reduction is more obvious at the 960 milligram. Next slide. In terms of response and disease control, the 960 milligram combination had 27% objective response rate and a disease control rate of 72% as compared to a disease control rate of 46% in investigators choice and no responses in that part. Responses were also seen at 240 milligrams but to a lesser extent, with a disease control rate of 68%, that still remains superior to the investigator choice. Next slide. Tolerability was exactly as expected, the most common grade 3 or higher treatment-related adverse events for sotorasib plus panitumumab where hypomagnesemia, dermatitis acneiform, rash and diarrhea. For the investigator choice arms, they were neutropenia, nausea and anemia, and absolutely nothing new to report in terms of new safety. So we can go to the next slide. So in conclusion, CodeBreaK 300 is a randomized Phase III trial that met its primary endpoints for superior PFS of this combination of LUMAKRAS, Vectibix, versus investigator choice therapy in metastatic colorectal cancer. We see a clinically meaningful benefit that is demonstrated by a median PFS that went from 2.2 months in the investigator choice to 5.6 months at 960 milligrams. The PFS favored the combination in all subgroups. We have higher response rate, higher disease control rate and OS, which I have not shown in detail is immature, but trending favorably, notably for the 960 milligram. We believe that these results, along with previous data from non-small cell lung cancer, supports sotorasib 960 mg as the sotorasib dose for use in colorectal cancer. So sotorasib 960 milligrams plus panitumumab is a potential new standard of care therapy for patients with chemotherapy refractory KRAS G12C-mutated colorectal cancer. Next slide. Colorectal cancer discussions with this Phase III data are obviously ongoing with the regulators and we are planning to initiate a Phase III study of LUMAKRAS plus Vectibix and FOLFIRI in their frontline setting of colorectal cancer, KRAS G12C mutation, as you know, represents 3% to 5% of colorectal cancers. Regarding non-small cell lung cancer, I'm encouraged by the activity of chemotherapy LUMAKRAS combination. We have initiated a Phase III study of LUMAKRAS plus chemotherapy in the first-line setting of PD-L1 negative KRAS G12C-mutated non-small-cell lung cancer. We believe that, that population represents around 13,000 patients in the U.S. Next slide. I will briefly remind you next slide that we also have another precision medicine therapy, which is the monoclonal antibody bemarituzumab that is targeting FGFR2b receptor, extremely prevalently expressed in gastric cancer, which is the fifth most common cancer worldwide. Next slide, the phase 2 data of FIGHT has shown a clear superiority of the combination of bema with chemotherapy over standard of care chemotherapy. And if we see next slide, we have 2 ongoing Phase III trials that are currently [ retreated ] in the frontline gastric setting, the combination of bema with FOLFOX 6 versus placebo FOLFOX6 and that's FORTITUDE 101, and we have Bema plus chemo plus nivolumab versus placebo chemo nivolumab in the same setting, FORTITUDE 102. With that, I hand it back to you, Dave. Thank you very much for your attention.

Great. Thanks, Jean-Charles. And thanks, everyone, for staying with us. As we said, we have a lot of material today. A couple of key messages I'd like to leave you with. We move forward with BLINCYTO development, that's the road map for other bispecifics into earlier lines of therapy and now with the development of the subcu formulation. With tarlatamab and xaluritamig, we believe we have foundational data to launch Phase III trials, but also to begin to think about development in earlier stages of disease, again, where we really are going to seek to change the natural history. In the precision oncology portfolio, AMG 193 is moving forward very briskly. And as we continue to accumulate data into the first part of next year, we'll really start to refine the development program there. And then I think you're all familiar with LUMAKRAS in the Phase III program as Jean-Charles just outlined. So with that, I think we will open it up for questions and answers. If you didn't catch the tarlatamab and LUMAKRAS papers in The New England Journal over the last few days that were published concomitantly with presentations at ESMO. I'd urge you to read those carefully. The xaluritamig data were also published in Cancer Discovery. And then a few weeks before that, the AMG 193 data were also published in Cancer Discovery, as well. So there's also a wealth of rich detail in all of these manuscripts. So with that, we'll go ahead and move to the question-and-answer session. Sarah, can you remind everyone of our procedure here and then open up the lines.

[Operator Instructions] Our first question comes from the line of Salveen Richter of Goldman Sachs.

So the tarlatamab results in small cell lung cancer are remarkable in such late-line patients. How do you see the drug fitting in as you move to earlier lines? And then could you comment on potential CNS activity for the drug, given the prevalence of CNS metastasis in this population?

Yes. Thanks, Salveen. Those are great questions. And I'll say a few words and then ask Dr. Owonikoko to also comment on those particular points. We're very eager to move tarlatamab into earlier lines of therapy where tumor burden may be much reduced compared to the sorts of patients that you saw treated in this third line and beyond setting. The -- and as I said, in fact, we expect to launch Phase III trials within the next 6 to 8 months in the sort of first-line induction maintenance setting as well as in limited stage disease. The disease is often treated with induction chemotherapy and the -- a large percentage of patients respond, but the problem is relatively rapid relapse and then aggressive disease. But that moment in time where they've achieved a response in a much reduced tumor burden, we think, is a point where the addition of the bispecific may actually be able to really alter the natural history of the disease. So that will be one of the focuses of the development program. Let me ask Dr. Owonikoko to comment on that and then also the issue of potential central nervous system activity, which is, of course quite important to this patient population. So Dr. Owonikoko.

Yes. Thank you. Those are two very important questions. I believe that tarlatamab would eventually have a definite role in the treatment of all phases of our patients, especially in the earlier stages. And two reasons why I said that. One is the current trial in the DeLLphi-301 study, we enrolled patients to a previously received anti-PD-1, PD-L1 therapy. And that did not compromise the ability to benefit further from tarlatamab. So guessing that these are potentially complementary mechanisms of immune activation and specifically the small cell where majority of the tumors are actually cold tumors. I think the mechanism by which tarlatamab works by binding to T cell wherever they are and then taking them into the tumor immune microenvironment will be an advantage and I will make this work even earlier in the treatment algorithm for patients. With regard to brain metastases, patients who are allowed on this trial, if they had brain metastases treated and stable. We did not have untreated brain metastases, so it's difficult at the current time to say what exactly is the role of efficacy of tarlatamab in intracranial disease control. So in terms of observation on this study, we did not see any excessive evidence as the majority of the patients are progressing in the brain, which could indicate that perhaps it's not working in that sanctuary side. So we did not observe that. The second is borrowing from other studies especially in the frontline with the combination of IO and chemo where patients with untreated brain metastases were allowed. We know that immunotherapy or immune-based strategy could also reduce the incidence of brain metastases. So I would say the jury is still out, but it looks promising based on the aggregate data that tarlatamab might be able to help patients in terms of brain metastases.

Great. So we'll move on to the next question. Let me just make one clarification. I got a little bit ahead of myself, my team just reminded me on AMG 193. That manuscript is just about to be submitted, but is not yet published, but the others are, of course, out and available for you. Sarah, what's our next question?

Your next question comes from the line of Jay Olson of Oppenheimer.

Congrats on the update. Can you tell whether or not there's higher response rates in patients with higher DLL3 expression for tarlatamab? And also if there's any potential correlation between DLL3 expression levels and CRS or ICANS. And then separately, are you planning to develop a subcu form of tarlatamab or xaluritamig similar to what you're doing with BLINCYTO. And then finally, can you talk about any CNS activity with AMG 193?

Yes. Let me take the first part. I'll ask Jean-Charles to discuss the AMG 193 portion of your question. We are looking at expression response data now, i.e., DLL3 expression versus response and should have those data forthcoming. The -- our general sense though is that we saw responses, including major responses across all levels of DLL3 expression, including in a patient or 2 with so-called negative tumors. And I wanted to address that because this is a question we've got quite frequently over the last few days. Recall that immunohistochemistry has a threshold of detection of typically a couple of thousand receptor molecules per cell or more in order for a positive result to be returned based on a wealth of data that we've accumulated in the laboratory over the last decade. We believe that the BiTEs are active, probably a threshold of a few hundred target molecules per cell surface and maybe even fewer than that. So my strong bias is that the DLL3 negative tumors that were reported is responding are, in fact, DLL3 positive but just below the limit of detection. And everyone really needs to keep in mind the assay limitations here. Likewise, we're looking at CRS versus DLL3 expression levels. So far, I don't think there's a major correlation there. There's probably a slightly better correlation with tumor burden as opposed to target expression levels, especially when we saw target expression in almost all of the tumors here, 96%. Finally, in terms of subcu development, we are thinking about that across the BiTE portfolio, that would vary molecule by molecule, depending on various characteristics, and I'll provide guidance as that work proceeds. In terms of CNS activity with 193, let me ask Jean-Charles if he wants to comment for a moment.

Yes. Thank you. We know preclinically that we have a very good CNS penetrant agent. Our master protocol for 193 will explore a cohort of lung cancer patients with brain mets, and we hope to have that started next year.

Right. Thanks, Jean-Charles. Yes. And specifically, we have demonstrated CNS penetration in various rodent models preclinically, and we attempt to do relatively rigorous experiments here. So that formed the basis for exploring this in patients with brain metastases, as Jean-Charles mentioned, beginning next year. Sarah, why don't we go to the next question?

Your next question comes from the line of Gregory Renza of RBC Capital Markets.

Congrats on a productive conference. Maybe just keeping with the AMG 193. Just curious if you and the team, Dave could comment on 193 in the context of just the broader PRMT5 landscape. And as you've described the development program, just interested in your comments on the case for combinations, namely with MAT2A, 397 and where you see perhaps the biggest impact from combination therapy and perhaps that which could be most transformative here in these patients?

Yes, thanks for the question and you opened up, I think, a very important avenue, which is what is the development path for a molecule like AMG 193 that has now demonstrated activity in at least 6 different tumors. So one of the things that we will be thinking about as we generate additional efficacy and safety data and finalize dosing, is there a tumor-agnostic pathway potentially supplemented with specific indications, will we proceed in specific indications. These are all of the things that are top of mind, but I think they are all possibilities based on what we are seeing here. In terms of the MAT2A combination with IDE, this is based on the biologic rationale that 2 basically are orthogonal hits on this pathway may be additive or synergistic. We're just beginning to generate those data. And obviously, looking at that very -- we'll be looking at that very keenly. I think the clinical profile, including tolerability that we have observed so far does set this molecule up for potential wide range of combinations. And as we get into '24, we'll start thinking -- talking about some of our thoughts there. As you mentioned, in many of the specifics, solid tumor indications are ultimately, combination therapy may in fact be the best use of a molecule like AMG 193. Thank you. Sarah, we'll go to the next question.

Your next question comes from the line of Umer Raffat with Evercore ISI.

This is [indiscernible] on for Umer. Just want to ask a question on tarlatamab Phase III data. So there is a higher number of patient deaths before post-baseline scan on the 100 milligram dose. Can we get more color on this difference between 2 dose arms and why these patients on the 100 milligram seems to progress faster? And also just a follow-up on the DLL3 negative patients. The number of patients in this trial is small, but what is the expectation for this subgroup of patients?

Yes. Thanks. So in terms of the progression, I suspect on -- one of the things we noted on the 10-milligram arm versus the 100-milligram arm is that patients because of tolerability, patients were able to get on average a couple more cycles of therapy at 10 milligrams. And that may, in fact, have led to the difference in overall response rate for example. And always in these patients, dose selection is trying to optimize the risk benefit in the clinic based on everything that we saw, we really feel that 10 milligrams at this point is at least for now, that the go-forward dose. Second question about true DLL3 negative patients. Here, we saw 96% of patients with some level of DLL3 expression. If you look at the literature, in almost all studies, it's above 80%. And in most studies, it's in the 85% or 90% plus range. So I think that's very consistent. DLL3 is a marker that is present on neuroendocrine stem cells and small cell lung cancer is a neuroendocrine stem cell-derived tumor, and therefore, it's perhaps biologically not surprising that you have very high levels of DLL3 expression. My guess is that we will continue to see that kind of expression pattern going forward. Sarah, next question?

Your next question comes from the line of David Risinger with Leerink Partners.

Yes. So with regard to tarlatamab, and particularly in light of the comment that you just made, could you talk about the market opportunity as you see it? Certainly it seems like it may be much larger than many appreciate. But how do you see the market opportunity for tarlatamab, particularly as you move it into earlier lines of therapy? And then as xaluritamig is evaluated in earlier lines of therapy, how do you see it being positioned in an increasingly crowded prostate cancer treatment landscape?

Thanks, David. I'll take the tarlatamab question, and then I'll ask Dr. Kelly to comment on how you think about developing a molecule like xaluritamig in prostate cancer given his decades of experience in this field. . Again, for tarlatamab, I would go back and look at the BLINCYTO development program. As you saw from the epidemiology figures that we shared, it's 2,000, 3,000, 4,000 patients with third line plus disease, but that's not the end game here. And if we're able to demonstrate good efficacy. And almost certainly, the safety profile will improve as you move to lower tumor burden settings. The real goal here is to be incorporated into that first line and limited stage disease setting where you can then potentially have the drug available to patients across the spectrum. And as I mentioned, they're on the order of roughly 30,000, maybe a few more patients across that spectrum in the United States each year, comparable numbers in Europe. The numbers in Asia are actually quite large where small cell lung cancer is relatively prevalent. So I think that the market opportunity, if we're able to successfully move into these earlier lines of therapy is much, much larger than that a few thousands of patients that you see in the true third line and beyond setting. Now Kevin, perhaps we can ask you to comment on how you would think about the development of an agent like xaluritamig as you move along the prostate cancer continuum, which is, of course, a complex treatment landscape. So go ahead.

Yes, David. It's very complex, and it's getting complex by the day. But the basic premises of most patients with advanced prostate cancer is they do not like hormonal therapy and they do not like chemotherapy. Even in the early settings, they try to look for any alternatives to hormonal therapy. So I think there's really a possibility as you move this up, you could either, I wouldn't say replace it, but actually maybe be able to modify your different therapies upfront, so it's more tolerable for patients. So I do think the patients are looking for alternatives, both to chemotherapy and hormonal therapy all the way through the spectrum and the journey of prostate cancer. So I think there's a huge opportunity to move this forward. Radioligands are also going to play an important role. Again, the question is how can we synergize -- how this new drug could synergize with radioligands and really accomplish eradicating more cancer and have patients free of hormonal therapy and chemotherapy. And that's the way I'd look at it.

Thank you Dr. Kelly.