AstraZeneca PLC (AZN) Earnings Call Transcript & Summary

Welcome, ladies and gentlemen, to AstraZeneca's Meet the Management Event ASCO 2025. Before I hand over to AstraZeneca, I'd like to read the safe harbor statement. The company intends to utilize the safe harbor provisions of the United States Private Securities Litigation Reform Act of 1995. Participants on this call may make forward-looking statements with respect to the operations and financial performance of AstraZeneca. Although we believe our expectations are based on reasonable assumptions, by their very nature, forward-looking statements involve risks and uncertainties and may be influenced by factors that could cause actual results to differ materially from those expressed or implied by these forward-looking statements. Any forward-looking statements made on this call reflect the knowledge and information available at the time of this call. The company undertakes no obligation to update forward-looking statements. Please also carefully review the forward-looking statements disclaimer in the slide deck that accompanies this presentation and webinar. [Operator Instructions] Now please stand by while I connect you to the auditorium in Chicago.

Good evening, everybody. Thank you for joining us. Welcome, and we'll start sharp on the hour because Dr. Tolaney and Prof. Turner need to leave us at 7:50. So again, welcome, and I hope you have enjoyed the ASCO as much as we did so far over the last couple of days. So this is our forward-looking statement, nothing special here. You know it. This is our agenda. And we are privileged to have, as I said, Prof. Turner and Dr. Tolaney joining us. And you will have a chance to hear from them what they think of this data, but also answer your questions. And of course, you have a number of us from the leadership team. Let me just start with reminding you what our ambition is. Our ambition is to become a leader in cancer care and -- but also in cardiovascular disease, respiratory disease. We're building immunology. And of course, we want to remain a leader in rare disease. Our goal is to achieve $80 billion by 2030 through science and through innovation and building a pipeline of products we have. But we're also working on what we call the day after tomorrow. And you've probably seen some of the early data we have on some of our new technology products. As you can see here, we are making good progress across all our key priorities, whether it is about weight management and the risk factors, we have now 3 Phase III studies that have either started or in the process of starting, supporting our oral PCSK9. We also have the Phase II program supporting the oral GLP-1. Now if you look at oncology, whether it's ADCs or IO bispecifics, cell therapy T-cell engagers also, we're making very rapid progress. We now have 6 specific AstraZeneca ADC in the clinics. We're also making progress with our radioligand programs. We have a large Phase III program supporting our IO bispecifics. And finally, as you can see on this chart, we're making also very good progress with our cell therapy and our T-cell engagers. We -- I should have mentioned one thing important, very important that Cristian would not be happy if I didn't remind you this. 50% of our trial -- more than 50% of our trial are enrolling ahead of plan. And it's important to keep this in mind for 2 reasons. First of all, it's good news. We are really executing very, very well, and Cristian and his team are doing extremely well. But the second aspect of it is you have to expect, of course, our R&D expenses are following this trend of accelerated enrollment in our clinical trials. Now we are still staying with our guidance for the year, of course, and just wanted to keep you informed of our progress. But for the full year, you should expect us to land at the upper end of the lower 20s. As we always said, we are targeting lower-20s percentage of turnover in terms of our investment in R&D. But this investment is actually a good investment. If you look at the progress we're making, we've had a whole series of positive studies and important ones since the beginning of this year. Of course, 3 of those we're presenting here, 2 plenaries and 1 special session, and there's more to come. More to come this year, as you can see on this chart, more to come next year. So it's really important to keep in mind that we have a very heavy readout programs over the next number of months. And that's why I've said many times, and I want to say it again tonight, by the end of this year or early next year, you'll have a good sense for whether we are on track with our 2030 goals of $80 billion. First of all, you'll see the momentum in the existing portfolio, but also you will see how we deliver on the pipeline. We're also investing in both manufacturing and R&D sites. So an important message here for you relates to tariffs. The biggest part of our manufacturing footprint is in the United States, and we have been building it over the last few years because we always thought we have to have a specific footprint for China, but we also have to have a specific footprint for the U.S. That allows us to manage the potential tariffs if they were to be implemented for pharmaceuticals. There's a couple of products, as we have communicated before, that we believe would still be impacted by tariffs, and we are moving the manufacturing of the U.S. supply to the U.S. And if any impact of tariffs we would have, it would be only temporary. So on the manufacturing side, we are well covered. We have invested recently in a cell manufacturing plant in Rockville, Maryland. So the tariff is something we can manage. Just very quickly because today is a meeting about oncology and the data we presented. For those of you who are interested in MFN, we don't have a lot more to say than what we've said before, which is we actually agree that there has to be some form of equalization. Europe and the wealthier nations around the world need to pay a bigger share of the innovation. And probably there has to be a rebalancing of the cost of this innovation and the pricing of our products across the world. But there's not much more we could talk about today about MFN. Let's try and focus on the great data that we've been developing. And we're also relying on a very strong network of R&D centers. As you know, we invested in a new center in Beijing, which we believe will really help us continue building the pipeline. And we hope to continue on this great series of ASCO plenaries that we've experienced over the last 5 years. Of course, we remain humble, as always, but we are also very proud to see those numbers, 8 plenaries, and I'm not even counting the special sessions, over the last 5 years since 2020. So with this, I'll hand over to Susan, who will take you through our progress and the themes here. Over to you, Susan.

Thank you, Pascal, and welcome, everybody. So just as a reminder, on the left-hand side of this slide are some of the themes that we are driving in terms of the strategy for oncology drug development that we talked about at the Investor Day last year. So first of all, of course, the opportunity to improve on the backbone of chemotherapy and radiotherapy with the advent of antibody-drug conjugates and radioconjugates; the segmentation of the currently defined IO-sensitive tumor types with the next wave of IO agents; the expansion of that space with the introduction of T-cell engagers and cell therapy; putting these together into key combinations that can have the potential for transformation of outcomes; and then driving the early intervention in the disease where the opportunity for cure and long-term benefit is greatest. I'm also very proud of the consistent delivery of ASCO plenaries over the last several years, which just demonstrates the value of the AstraZeneca team. And you see many of the R&D team that are responsible for this here in the room with us today. So again, in terms of our presence at ASCO this year, it's not just about the plenary sessions that I'll talk -- we'll talk more about. But we have 82 abstracts accepted, and we have many other oral presentations that have gone on, which are just demonstrating our progress against those key themes that we're driving for our strategy; and 8 simultaneous publications during ASCO, including 3 in the New England Journal of Medicine. So what does this do in terms of helping to move us against those themes. First of all, we're going to hear a lot about the SERENA-6 and DESTINY-Breast09 studies, which demonstrate our leadership moving into the first-line treatment of breast cancer in the hormone receptor-positive breast cancer setting with SERENA-6 and in the HER2-positive breast cancer setting with DESTINY-Breast09. We're also establishing improvements in outcomes and transformational outcomes in gastrointestinal cancers, which is something that I don't think everybody pays as much attention to as perhaps is needed. So really delighted with the data from the MATTERHORN study in the perioperative regimen. Again, another example of a perioperative regimen that can have this potential for improving the long-term outcome and the potential for cure, but also DESTINY-Gastric04 confirming a HER2 benefit in gastric and GEJ cancers in the second-line setting. And then again, Phase II data from our GEMINI study in hepatobiliary cancer showing the role of rilvegostomig. And then in lung cancer, where we really have bold ambition and already have a leadership position, we have the ambition to treat 1 in 2 lung cancers by 2030, and good progress against that in terms of further data that was presented here with the NeoADAURA and data with the NeoCOAST-2 in the neoadjuvant setting, TROPION-Lung02 in the first line and TROPION-Lung04 for the combination of rilvegostomig with dato. And I'll talk a little bit more about those. So where do all of these fit in the lung cancer landscape? And you can see, this is the overall map. I've highlighted in the yellow boxes those studies that we're going to talk a little bit more about here. And again, as said, we've established leadership in lung cancer with agents in many different segments of the lung cancer map. And what we're presenting in terms of the data here at ASCO shows how we're moving towards that ambition of treating 1 in 2 patients. So the new data that we have with Datroway and IO in QCS gives both better confidence in how we can use the QCS biomarker in the first-line setting in combination with IO agents. And the data that we've got with NeoADAURA shows how we -- it's really important to be testing patients with EGFR-mutant lung cancer at the time of diagnosis and, together with the data that we've got from the SAFFRON and SAVANNAH studies, really reinforces that Tagrisso is the backbone therapy for EGFR-mutant lung cancer across all stages of this disease. So let me show you a couple of snippets of these data with the NeoADAURA data here. Again, neoadjuvant treatment of EGFR-mutant lung cancer does not, with chemotherapy, does not result in a meaningful pathologic complete response rate. What we hear -- the primary endpoint in this study was the major pathologic response rate, and this was clinically meaningful in this setting. You had a significant improvement in both the Tagrisso monotherapy and the Tagrisso plus chemotherapy arms compared with placebo plus chemotherapy. So again, this reinforces that you really should be testing all patients with lung cancer for EGFR mutations before starting an neoadjuvant therapy because the currently established treatments there of chemotherapy or chemo plus IO do not work. And what the early data for event-free survival from this study showed, as you can see on the right-hand side, is actually that you are seeing a trend to improvement for the Tagrisso plus chemo and the Tagrisso monotherapy arms compared to that standard of care. And what Jamie Chaft, the PI for this, also showed is that endpoint of major pathologic response was predictive of this EFS outcome in this patient population. In terms of the other data that we showed for Datroway and IO, first of all, on the left-hand side, we built on the data that we had shown at the World Congress on Lung Cancer, confirming the pathologic complete response rate of 35%, a major pathologic response rate of 63% for the combination of Datroway and Imfinzi, which is around double the responses for both path-CR and NPR that we saw with Imfinzi plus chemotherapy in the AEGEAN setting. And again, what we also saw there, as you've seen in many other studies, is when you do get a path-CR, that predicts for the event-free survival outcome. So this is very encouraging. In the middle panel, what we showed with the TROPION-Lung02 data is more data on the Datroway plus pembrolizumab with or without platinum chemotherapy in this study. And the most important takeaway from this is that the QCS biomarker that we showed from the TROPION-Lung01 data is also predictive of the outcome in this setting in a first-line setting in combination with IO with or without chemotherapy for both PFS and OS. And on the right-hand side, you see the data from the TROPION-Lung04 study with the combination of Datroway plus our PD-1 TIGIT bispecific rilvegostomig, again, showing deep and durable responses with around half of these patients having ongoing responses at the current time. So again, if you put all of these data in context now of the breast cancer landscape map, what you can see is that we are moving into this first-line setting and changing and redefining how breast cancer is treated in that first-line setting with a very innovative study that is SERENA-6 that Nick Turner will talk about and also the DESTINY-Breast09, which really changes the game for first-line HER2-positive breast cancer. And so with that, I'm delighted to hand over to our key external experts to walk you through this data. And I'll hand over now to Nick Turner to talk about the SERENA-6 study. Thank you.

Well, thanks for that introduction, Susan. So it's a pleasure to show the results of SERENA-6. So SERENA-6 was a Phase III double-blind study in hormone receptor-positive HER2-negative advanced breast cancer in the first-line patients who are on aromatase inhibitor and CDK4/6 inhibitor for at least 6 months with any of the 3 approved CDK4/6 inhibitors who had ESR1 mutations detected in circulating tumor DNA and without any evidence of disease progression and testing to identify the mutations was carried out every 2 to 3 months at approximately the same time as staging scans. We randomized 315 patients 1:1 to continuing the AI, continuing the CDK4/6 inhibitor and the placebo for camizestrant or switching over to camizestrant, continuing the CDK4/6 inhibitor and the placebo for AI. And the primary endpoint was PFS by investigator assessment. So we saw a substantial improvement in PFS as the primary endpoint. So we saw median PFS improved from 9.2 months in the control group of continuing AI to 16 months in the switch to camizestrant group with a hazard ratio of 0.44, highly statistically significant, so over a doubling of PFS. And the curve separated early and stay separated. And importantly, at 24 months, only 5% of patients were without progression who had continued on AI, but that increased to 30% who switched to camizestrant, I think with clear evidence of a tail emerging. And I think it's really important to place this in the context with what we've seen with oral SERDs in the second or third line, where we've seen a delta of the median of only 2 months, whereas here, we're seeing a delta of 7 months. So really quite a favorable comparison with the data we're seeing in this setting. We've also looked across a number of prespecified subgroups, clear evidence of consistent efficacy of camizestrant across all subgroups. And importantly, we've boxed out the 3 CDK4/6 inhibitors. So all CDK4/6 inhibitors were represented in the study with clear evidence of efficacy across all 3 of the CDK4/6 inhibitors. Now really importantly, we also looked at patient-reported outcomes in SERENA-6. And so this is the time to deterioration in quality of life, and that improved from 6.4 months in the control group up to 23 months in those who switched to camizestrant with an adjusted hazard ratio of 0.53, and we also saw a time to deterioration in pain. And so I think this is really picking up what progression actually means for patients. It causes substantial symptoms, and it's also a considerable psychological blow for patients. Many patients are on their first-line AI and CDK4/6 inhibitors for years. They're back to normal life, but that first progression event is quite an unpleasant, unfortunately, pointing for them that this isn't going to last forever. So avoiding progression really matters, and I think that's what we pick out in this data. We also looked at PFS2, so time to progression on the second-line therapy. At this point, this data is really quite immature. But nonetheless, there's a really encouraging hazard ratio of 0.52 at this point, and we'll need longer follow-up. And PFS2 is a very important endpoint because it's going to very likely show that, that benefit we saw on PFS1 will carry through. But there are some limitations on the data at the moment, in particular, in the camizestrant arm, many patients are still on treatment. So we've only really had the bad actors that have progressed on the camizestrant arm and are contributing to PFS2. So I think with longer follow-up, we would very much expect PFS2, if anything, to get better with longer follow-up, and we'll certainly look forward to those future analyses. In terms of tolerability, camizestrant was very well tolerated. There were similar rates of serious adverse events in the 2 groups. The discontinuation rates were universally low across the study. Only 1% of patients had to discontinue camizestrant due to adverse events. There's more detailed analysis of the AE profile on the right-hand side. So there was a modestly increased rate of neutropenia in the camizestrant group. But this, of course, is the characteristic side effect of the CDK4/6 inhibitor partner and is asymptomatic. And this modest increase simply just reflects that they were on their CDK4/6 inhibitor for longer. Then if we look at symptomatic adverse effects, broadly, they are very similar between the 2 groups. The one that was higher is photopsia. So that was present in 20% of those on camizestrant, 8% of those on AI. But photopsia is brief flashes of light in the peripheral vision. It had no impact on patients in this study. And bradycardia, we only saw at very low rates. Bradycardia was sinus bradycardia and was asymptomatic. And I think what we really see coming out of this big first Phase III of camizestrant, that the side effects of photopsia and bradycardia were really not clinically relevant. And certainly, that matches my personal experience with my own patients. And then just to look at the conclusions, switching AI to camizestrant whilst continuing the CDK4/6 inhibitor, guided by the emergence of ESR1 mutations during first-line therapy ahead of disease progression, significantly improved PFS in patients with hormone receptor positive/HER2-negative breast cancer. That PFS benefit was consistent across the CDK4/6 inhibitors and all the clinically relevant subgroups. Switching to camizestrant substantially delayed the time to deterioration in quality of life. Camizestrant was well tolerated with a very low rate of discontinuations due to AEs. And SERENA-6 is the first global registrational Phase III study to demonstrate the clinical utility of using ctDNA monitoring blood tests to detect and treat emerging resistance in breast cancer. And we think that this hopefully is a new way of refining first-line treatment for patients with advanced breast cancer that has a significant potential to improve their outcomes. Thank you.

Well, now we're going to switch directions and talk about HER2-positive disease. So these are the data from DESTINY-Breast09. So DESTINY-Breast09 is a randomized Phase III study of trastuzumab deruxtecan with or without pertuzumab compared to a taxane plus trastuzumab and pertuzumab for first-line therapy of metastatic HER2-positive breast cancer. So this was a trial for patients who have not had any previous therapy in the metastatic setting for HER2-positive disease, and they were randomized 1:1:1 to receive trastuzumab deruxtecan plus placebo or trastuzumab deruxtecan plus pertuzumab or a taxane plus trastuzumab and pertuzumab. And the trial was really designed to assess if trastuzumab deruxtecan plus pertuzumab is better than THP and was also designed to look to see if trastuzumab deruxtecan alone is better than THP. And so here at ASCO, we presented the results of the interim analysis comparing trastuzumab deruxtecan plus pertuzumab to THP, given that those data did meet the stringent criteria set at the time of the interim analysis for success. The T-DXd monotherapy arm did not meet those stringent criteria. So those patients on the T-DXd plus placebo arm continued to be followed until the primary PFS analysis. So here are the data for the primary endpoint of progression-free survival comparing T-DXd plus pertuzumab to THP. And what you can see is that the combination of T-DXd and pertuzumab led to a statistically significant and, I think, very clinically meaningful difference favoring T-DXd and pertuzumab compared to THP. So the median PFS for T-DXd and pertuzumab was 40.7 months compared to 26.9 months with a hazard ratio of 0.56 and a p-value that was less than 0.00001. And I think what's really interesting is you can see that these curves separate early, but they really continue to widen over time. And in fact, because our follow-up time is about 29 months at this time point, you can see that there are very few events at the tail of the curve. And so it does mean there's some instability at that tail. And so we do still have 46% of patients that remain on study treatment with T-DXd and pertuzumab. So it does suggest that this median is likely to evolve with further follow-up. You can see that the benefit for T-DXd and pertuzumab when compared to THP was really consistently seen across all the subgroups, and that included our stratification factors, which are highlighted in yellow, which looked at prior treatment status, hormone receptor status and PI3 kinase mutation status, again, showing benefit across all these groups. The objective response rate was also numerically higher for T-DXd and pertuzumab, around 86% compared to about 79% for THP. But what I thought was very important, too, was that the complete response rate with T-DXd and pertuzumab was almost twice that of THP. So the complete response rate was 15% for T-DXd and pertuzumab compared to 9%. And when you look at the overall survival data, I think it's important to remember, survival data is very immature and is only being presented descriptively. There was no formal testing for OS at this time. There are only 16% of survival events that have occurred at the time of the interim analysis. But even so with very limited follow-up, you do see that there's already a trend favoring T-DXd and pertuzumab with a hazard ratio of 0.84. When you look at the safety and toxicity, you can see that, in fact, the rates of Grade 3/4 adverse events and the rates of serious adverse events were actually similar between the 2 arms. There were more dose interruptions and dose reductions with T-DXd and pertuzumab compared to THP. And there were numerically a few more deaths with 5 deaths on the T-DXd and pertuzumab compared to 1 on the THP arm. To the right, you can see the most common toxicities by arm with the most common toxicities for T-DXd and pertuzumab, including nausea, diarrhea and neutropenia; and for THP, including diarrhea, neutropenia and anemia. So very consistent with the known toxicity profiles of the individual agents used. So just to summarize, T-DXd and pertuzumab demonstrated a statistically significant and very clinically meaningful improvement in progression-free survival with a 44% reduction in the risk of disease progression or death. The duration of response was also substantially longer with T-DXd and pertuzumab, and the complete response rate was almost double. And while the overall survival data certainly are immature at this time, you already see a trend favoring T-DXd and pertuzumab. And we also looked at PFS2, so looking at the time from randomization to progression on second-line therapy. And you can see that, that also is supportive of the impact on longer-term outcomes. The safety data were consistent with the known toxicity profiles with no new safety signals identified. And so I think these data really do suggest that T-DXd and pertuzumab should be considered a potential new standard of care for first-line HER2-positive metastatic breast cancer. Thank you. I will pass it over to Sunil Verma.

Thank you, Dr. Tolaney. I'm going to keep my remarks short because I think we have some time for Q&A before Dr. Tolaney and Prof. Turner have to leave. I think it's really important to see, we just heard 2 really remarkable presentations looking at transformation in first-line treatment of metastatic breast cancer. So how do we put this into context? So let's take a look at really the meaningfulness of the HR-positive breast cancer segment, as highlighted by the great presentation that Prof. Turner did at the plenary presentation. So the first thing, AI plus CDK4/6, of course, is the standard of care. There hasn't been any substantial changes in that standard of care for many years since 2012 when the first data came out. And SERENA-6 is the first and only next-generation SERD with Phase III data in the first-line setting. There's a lot of excitement, of course, about this field, but this is the first data that we have seen. So I think this was one of the key reasons why this was also included in the plenary. And if we take a look at the data and how it sort of shapes up, there's about 85,000 patients in G7 that are treated with first-line HR-positive breast cancer. About 50% of them receive AI plus CDK4/6. There's another 20% or so that get fulvestrant plus CDK4/6 and another segment that get chemo or ADCs in that setting. Of those patients who get AI CDK4/6, about 3/4 of them do not progress and are able to stay on treatment for beyond 6 months, and that's the patient population that really was enrolled in the SERENA-6. And 30% of these patients will drop ESR1 as point of molecular progression before radiological or clinical progression. And that's this patient population that really camizestrant was shown to be superior to AI in combination with CDK4/6. A few things. There was a healthy debate, of course, discussion that happened at the plenary, but it's important to note that ESR1 testing is already established. And now we're using that established ESR1 testing and using that as part of routine blood test and, as Prof. Turner showed, as part of the routine visits that these patients have every 2 to 3 months. And I think that's an important principle about integration and how this could be integrated within routine clinical care. Now I think the results of this are certainly very meaningful in the first-line setting. But beyond that, they actually support the role of camizestrant. It's a safe regimen, as Prof. Turner showed, and that gives us a lot of confidence as we have a very expanded clinical development program. We're looking forward to SERENA-4 results that broader -- that broaden the camizestrant use in the first-line HR-positive advanced breast cancer setting. And of course, that's an all-comer approach. And then, of course, we have a very significant program in the adjuvant setting with CAMBRIA-1 and CAMBRIA-2, both looking at early as well as late adjuvant setting, and those studies are enrolling very well. And we're looking forward to, of course, bringing this camizestrant in early-stage disease. And again, I want to underscore the safety and the efficacy really positions us very well for this expanded development program. We remain confident with the overall potential of camizestrant, and it's a $5 billion-plus peak-year opportunity with this program that we have. Further moving on to another breast cancer subtype, HER2-positive, as you just heard from Dr. Tolaney, DESTINY-Breast09 really marks another milestone in the treatment for first-line therapy. Now this goes back many decades, as it was presented by the discussant, the standard of care used to be around 4 months PFS with chemo alone in 1998. And we now have exceeding 3 years, nearly 10x the progression-free survival that we saw nearly 3 decades ago. And it's building on the efficacy that was very strong with CLEOPATRA. So this gives us a lot of confidence that this is going to address, of course, the clinical care. Now there's nearly 23,000 patients who are receiving first-line HER2-positive metastatic breast cancer in the G7. And this demonstrates not only activity that is very definitive with a strong PFS, as shown, and supportive evidence for PFS2 and overall survival, but also shows the consistency of treatment effect across key subgroups, including hormone receptor positive and hormone receptor negative. And the reason for really moving it up earlier is there's about 1 in 3 patients that don't go on to receive second or later lines of therapy, again, showcasing the benefit of starting with the best therapy upfront, as Dr. Tolaney highlighted. So we believe and we remain confident that this first-line data further represents the multi-blockbuster opportunity that we have for Enhertu across the lines with AstraZeneca and Daiichi Sankyo. And furthermore, you heard, of course, the release that we shared in May about DESTINY-Breast11, which is looking at neoadjuvant approach with Enhertu followed by THP. And that further underscores our ambition to move Enhertu into early stage. And of course, we are all eagerly awaiting the results of DESTINY-Breast05 as well, where we think that we have the potential to take the Enhertu and the power of Enhertu that we have seen in late line, in first line and beyond. So with that, I think we'll take Q&A. Thank you, Dave.

Okay. Thanks, Sunil. All right. Great. Sorry, just so I can keep track of time here, so we've got about 15-plus minutes to be able to ask questions. Right now, I would request that where possible, we focus questions for Dr. Tolaney and Prof. Turner. You get to hear from us a lot. You've got 12 minutes of precious time or 15 with our 2 guests.

And I think that it looks like, Emily, you've got a microphone. So why don't you start and then Rachel will come by. And if you can just give your name before you ask the question, that would be super. And we'll also take some online, Andy?

2 mics.

Oh, 2 mics, all right, very well. Okay. Perfect. Please, Emily.

Emily Field from Barclays. KOL we spoke to this afternoon indicated that the rate of de novo metastatic disease in DESTINY-Breast09 was higher than the real-world setting. And this particular doctor was, therefore, somewhat complex that the ORR rates were not higher than what was shown in D-B03. So just wondering if you could comment there. And then I guess this question is for Dr. Tolaney since we heard from Dr. Turner yesterday. Obviously, a robust debate with the discussion in terms of waiting for PFS and OS data versus the point of the quality of life data presented yesterday. So Dr. Tolaney, I'd be curious on your thoughts on which of those points you would value more.

Yes. So I'll start with the question about metastatic HER2-positive disease. So in DESTINY-Breast09, the number of patients with HER2-positive disease was just slightly over 50% of the population. In fact, if you were to look at the rate of de novo metastatic disease in the United States, it's actually rising. So probably in the U.S., it seems most of our real-world data is suggesting it may be as high as 70% because all these HER2 drugs are doing such a wonderful job at curing our early-stage patients. And with T-DXd, that may be getting even better as it moves into the early-stage setting. So I think the proportion of patients that we see upfront, at least in the United States, is becoming more de novo. And as you saw in DESTINY-Breast09, the benefit was actually very consistently seen both in the de novo subgroup as well as the recurrent group with very similar hazard ratio. So I think it shows it works in both populations. With respect to SERENA-6, I think the data that we saw was really I think it is a paradigm shift in the way we think about breast cancer. We have never really thought about switching therapy prior to anatomic progression, and to be able to switch with molecular progression, I think, is quite innovative. And I will say, it's hard right now because the PFS2 certainly isn't -- is immature. But I think what really got me was the quality of life data that you've not only profoundly improved the progression-free survival in the first-line setting, but these patients are living better in addition to having their disease controlled longer. And I think it goes hand-in-hand, right, that if we can control people's disease better upfront for longer, they're going to have a much better quality of life. And so I think it was very consistently seen. And it seems like PFS2 is also going to hit given the trend. So I think really, it is a really innovative strategy for a patient.

Thanks, Dr. Tolaney. Emily, I just would offer that I had heard, consistent with Dr. Tolaney's comments, commentary that they were pleased to see that there were as many de novo patients in the study because that is a more accurate reflection of what they're seeing than had it been actually a fewer number. Microphone. Yes, please.

It's Rajan Sharma from Goldman Sachs. Just on SERENA-6, do you have a sense of the time window between ESR1 mutation emergence and then progression? I'm just thinking if there's an opportunity to potentially reduce the time period of testing, so I think the discussant talked about that as being burdensome yesterday.

So we don't have that data yet from SERENA-6, and we're certainly going to look at that. We know ESR1 mutations don't emerge before 6 months, and that's why all patients on the study have been on for 6 months at least, and they then emerge most likely in the next 2 years after that. But all our patients come in for blood draws every 2 to 3 months anyway whilst they're on treatment. And so actually, I think adding this on to that blood test could easily become routine in our clinics. And actually, my experience is from all the studies we've done in this and similar that actually, patients find it really reassuring that you're monitoring their disease more carefully. So there might well be a bit of result anxiety for the few days before the blood test. But actually, that's outbalanced by that much longer-lasting reassurance that you're monitoring the disease better.

Simon?

Simon Baker from Redburn Atlantic. A question for Dr. Tolaney. The discussant was talking about the potential for sequencing managing effect with toxicity. Do you -- in how you will implement the results of this study with your patients, do you think there's any role for that? Or does the data support simply following the outcome that we saw in the study?

Yes, no, it's a good question. In fact, it's kind of a good problem to have that we had such a prolonged progression-free survival that we can now even think about these questions, right? I mean, in essence, the combination of T-DXd and pertuzumab is doubling the progression-free survival. And in truth, we don't even know what that median is really going to be. It could be even longer. And so I think that's why this question is being brought up because it could be that patients have their disease controlled for so long. Does it really make sense to continue to keep them on T-DXd and pertuzumab? Certainly, this trial cannot address the question. All we can say is that we know with the treatment the way we gave it, we're having a profound impact on outcomes. But I think it does bake the question, are there ways to optimize therapy? And I think while I find that it's very important, I also don't want to compromise the outcome because we doubled complete response rates, and there is a chance that we could be curing some of these HER2-positive breast cancer patients. So I think we have to be careful about how we think about optimization. And so I think it will be interesting to dig into D-B09 a little bit better because we'll be able to see, well, when are people really hitting CR? Are there opportunities to look at timing of a ctDNA clearance? Are there ways we could think about doing this in a more educated manner? So I hope it will lead to studies to help us to understand that.

Thank you, Dr. Tolaney. Prof. Turner, we'd be interested in your perspective as well on the same question.

Yes, I think it's -- I think we're at a really exciting time in HER2-positive breast cancer, which Sara just brought up, which is I think we're well on the way to starting to cure a significant number of patients with HER2-positive metastatic breast cancer. I don't think we've really been confidently able to talk about that with metastatic breast cancer before, but we start to see it with the CLEOPATRA regimen of taxol and dual-HER2 targeting. We're perhaps around 10% are cured. And I think what really excites me in D-B09 is the tail on the curve to me suggests that we actually might be starting to cure more patients with HER2-positive breast cancer. And it's great then that gives us the opportunity to be more ambitious in the future as well.

Thank you very much. Please, Christopher?

Christopher Uhde from SEB. So I guess this is a -- it's a layered question. But the first part is, I guess, something to be addressed perhaps by both the Professor and Doctor and leadership. So could you please share your confidence to the extent to which you're confident in a positive OS benefit for both SERENA-6 and D-B09? And I guess, perhaps it's worth commenting for -- to give Mr. Market some comfort on the reasons for the gap -- the big gap in investigator-assessed and BICR PFS in D-B09. I mean it's also -- it's clear there's an expectation that PFS will increase with longer median follow-up, but what about duration of treatment as well? And are you more confident today in Enhertu Perjeta than Enhertu monotherapy hitting OS? And to what extent might the Enhertu arm have been hurt or helped by the low concurrent use of endocrine therapy compared to the THP arm? And what does SERENA-6 tell us about the odds of success in SERENA-4 and CAMBRIA?

Those are really lot of good questions.

Let's see if I can help on that. We have issued yellow cards in the past for 5 questions, but we'll let you go away on this one. Why don't we start first with maybe we can have Dr. Tolaney, if you want to talk about the gap in BICR or the differences between BICR and investigator. Also, I think if you could address whether or not you have an expectation of duration of therapy to also increase as the population or the study continues to mature. And then if you want to comment on overall survival thoughts, and maybe we can also ask Susan to do the same.

Sure. Thank you. So you're completely correct that there was the difference -- a slight difference in performance of PFS in the control arm by investigator assessment in BICR. So you saw around 20 months with investigator assessment and around 26 months with BICR. So at first, when you look at that, at least when I first saw it, I thought, well, if it was not blinded, people knew they were on THP. So you could imagine and think, well, maybe there's some bias that patients on the investigator arm may come off early because they want to transition on to get another therapy like T-DXd in the second-line setting. So what we did is we looked at the rate at which investigator-assessed PFS was occurring prior to BICR PFS in both arms, and it's actually the same rate. So that's suggesting it's not due to bias, right, because it's occurring in the same rate on the T-DXd and pertuzumab arm. And so what happens is if a patient comes off for investigator-assessed PFS and they change treatment, we continue to follow them with serial scans until they hit a BICR PFS. So let's say I was on THP and I went on to T-DXd, I'm going to keep getting scans while I'm on my T-DXd until I progress and then that counts as my BICR PFS. So there's going to be a very big lag in time between your investigator PFS and your BICR PFS in that case. And when there are a lot more PFS events in the THP arm relative to T-DXd-P that will drive this difference to make BICR PFS longer. And so we actually looked at this with our statisticians, and that was exactly what drove that difference. So I think it made sense. With regards to the duration of treatment, so at this time with the 29 months of follow-up, we had a median duration of treatment that was almost 22 months for T-DXd and pertuzumab, and our PFS was almost 41 months. And so one could say, well, why is there such a difference there and what will happen to the treatment duration? But you have to remember, when you only have 29 months of follow-up, you cut your treatment duration at that time. So that treatment duration is going to get much longer. It's just kind of where we are at the snapshot in time. And so that will change again with longer follow-up. And then with regards to overall survival, at this time, we only had 16% of survival events. So we just don't have maturity. But I think the fact that you're already seeing a trend is supportive, and there's a lot of data to suggest that PFS2 very much correlates with overall survival. And with a hazard ratio of 0.6 for PFS2, I think it's highly likely that you'll hit a survival difference.

Dr. Tolaney, could I also just ask, there was one other that was in there, which was a commentary on the mono arm. And so just are you any -- do you have any more confidence in the pertuzumab-containing arm versus the mono arm? And is there any extent by which Enhertu might have been hurt or helped because of the concurrent use of...

Of pertuzumab. Yes, it's an interesting question. So I will be honest, I was a little surprised when T-DXd pertuzumab hit and not T-DXd because we'd already seen data in DESTINY-Breast03 that T-DXd has a 29-month PFS in a pretreated population, so obviously, expected to be much greater. But then you look at the statistical design of what's required to hit success at this interim time point and is extraordinarily stringent, and so just a few events could make you not hit. And so when you think about that, then you think, well, it's a little bit of chance that we probably didn't hit in the T-DXd monotherapy arm just, again, given the extraordinarily stringent criteria. And so I think I have expectations that it will hit at the time of the final PFS. And so I would hope that we'd actually potentially have 2 approvals in the future for T-DXd and T-DXd-P. And I do not think there is any detrimental effect. We actually already have data from DESTINY-Breast07, where we actually saw that the 18-month PFS was actually identical in small numbers, but you saw 80% of people were free of progression with T-DXd-P and T-DXd in the first-line setting. So I think it looks very good.

Perfect. Thank you. We're going to take one question, which I think we have time for online, and then we're going to let our guests catch their flights. James Gordon, JPMorgan, please go ahead.

James Gordon, JPMorgan. Two questions on camizestrant and SERENA-6. So the first one was on testing. So the discussant at the conference noted the burden of testing and talked about patient reluctance. And I think the patient side has been addressed just now. But what about funding for the frequent testing that needs to be done? Could that be a challenge, particularly ex U.S.? Is that going to have a lower cost? And is that something that Astra would be able to help with? Or could that be a problem in terms of getting funding for doing this testing? And the other question about the study was just quality of life. So there was some talk about scan anxiety, but what about the benefit on quality of life? And is that likely just because of slower disease progression? Or is that also about side effects for the drugs? Like how burdensome are AIs? And what about the side effects you get with a SERD like cami? Are those side effects likely to be a challenge for patients?

So Prof. Turner, I'd suggest, if you want to answer the second question, we'll take care of the first one.

Okay. So the question was about the quality of life benefit we've seen. I think this is likely driven by the side effects of disease progression of pain that is experienced as well as all the other symptoms that come, and this often does predate the actual confirmation of progression on the scan. And I think we're seeing that in the study where you're starting to see the quality of life falling in the control arm a little bit before you see it on the scan. So I think that fits. I don't think this is driven by differences in side effects. Camizestrant overall we're seeing is very well tolerated broadly. The symptomatic AEs are the same in both subgroups. But we will be able to drill down much more into the quality of life data, the individual sub-domains, and we've got plans to show that soon in a future congress, in a future meeting. So I think that will be really important.

Great. Thank you very much. I will -- then just, James, with respect to the reimbursement aspect of things, now that we've seen results such as these, it allows for us to have conversations with payers across the globe on the opportunity to be able to think about reimbursement. And so we're already engaging in those discussions. Certainly, there's also multiple avenues that are taken. There's the companion diagnostic route, which is the one that in the United States is the best route, lab-developed tests, and there are multiple of those that we're seeing now really in earnest being developed ex U.S. And so it's individual country approaches that are going to be taken, both with respect to access and also reimbursement for the ESR1 mutation testing. And what I would say is that we've got a pretty good track record of working on those things across the globe. Most of the medicines that we've launched have been precision medicines that have had a companion diagnostic, and we'll employ the same here. I want to ask all of us to thank our guests, Prof. Turner and Dr. Tolaney. And you may go. We know you need to catch flights. So thank you very much. So with that, as they leave, I'm going to turn it over to Cristian Massacesi, who's going to take us on to the next section. Thank you.

Thank you, Dave. So let's change the organ, GI. So I will go briefly through the key presentation that we had that was very rich for our GI portfolio. Before going there, I want to show and share with you, like Susan said and showed with lung cancer, breast cancer, a tumor map for gastric cancer. This is an indication that for us is becoming absolutely in strategy. If you start from later line, you can observe that, of course, at this conference, we presented DESTINY-Gastric04 that confirmed in a randomized fashion the role of Enhertu in second-line gastric cancer, HER2 positive after, of course, the approval based on DESTINY-Gastric02 in the context of a single-arm study. We are also running in parallel in a very second important segment of gastric cancer, Claudin18.2-positive, a Phase III study in the same setting, second, third line with our Claudin18.2 MME ADC. The nick name is sone-ve. This is the drug that currently we are comparing against standard chemotherapy. The study is ongoing, is actually recruiting, as Pascal was telling at the beginning, ahead of schedule. We have then moving -- the plan is moving these ADCs in the -- where they can potentially bring even further benefit in first line. In HER2 positive, we are testing Enhertu in combination with fluorouracil, capecitabine in combination with IO in HER2-positive setting. And we decided to run in agreement also with Daiichi Sankyo, we run 2 studies with 2 different IO backbone, pembrolizumab, and this is DESTINY-Gastric05; or rilvegostomig, our PD-1 TIGIT. There are supportive data for both trials, and we believe that this can strengthen the position of Enhertu in frontline study and, of course, help us also to position our PD-1 TIGIT in this important space. Both studies are ongoing and actively enrolling. Of course, there are plans also to move our Claudin18.2 ADC in this space, in this case, will be a combination with fluoropyrimidines and rilvegostomig. Then of course, the line of sight is where MATTERHORN is, what we presented here. So we are creating and generating these very powerful regimens with novel IO backbones and better chemotherapy, the ADCs, and the idea is to bring where we can cure more patients in the early stage where MATTERHORN has been run. Let's go to MATTERHORN. This is the design of this study, is first of all, very important to highlight is the third positive perioperative study that, as AstraZeneca, we are delivering after AEGEAN in non-small cell lung cancer and NIAGARA in bladder cancer. It's a resectable setting. The patient can be resected, but the disease is high risk, so Stage II-IVa. And so the patients are receiving an induction and treatment with a FLOT that is the best chemo backbone, is a cocktail of 4 drugs: fluorouracil, leucovorin, oxaliplatin and docetaxel. And then after the surgery, the patients receive other 2 months of FLOT. What we did in this trial is adding durvalumab during the chemotherapy phase and then adding 10 cycles, 10 months of durvalumab as maintenance. The primary endpoint of the study is EFS. Of course, the key secondary endpoint are overall survival and pCR. pCR is an endpoint that is not very well defined in gastric cancer, resectable gastric cancer like, for instance, in non-small cell lung cancer or triple-negative breast cancer, but it's something that can help to understand the outcome of these patients. We presented, if you recall, pCR data from NIAGARA for MATTERHORN in previously in 2023 because, of course, there was a readout that happened earlier. And there was a benefit. We see a benefit. We saw a benefit with a delta of 12% FLOT D versus FLOT. The results of this ASCO show that also EFS is clinically and statistically significant. The other ratio was at 0.71. The median EFS for FLOT arm was 33 months, not reached for durvalumab FLOT. It's very meaningful because this is an early setting. And if you see the 18 months and 24 months rates, you see that you go from 64% to 73% and 59%, 67%. In an early setting, it's an important delta. What is even more intriguing is this early OS benefit, very strong trend. The other ratio for OS was at 0.78. We already reached the median OS in a comparator arm, it's 47 months, not reached for durvalumab FLOT. This is a descriptive analysis. We didn't plan because of the maturity that we were expecting a formal testing analysis for OS. So this is why this analysis is not statistically significant. But of course, when the right maturity will be reached with adequate number of events, it will be formally tested with full alpha. Safety, very, very important in these specific settings, early settings and very reassuring. What we have seen with -- in this trial once more with durvalumab in early setting is a very consistent safety profile that we were expecting. Actually, I would say, here, if you see most of the adverse events were driven by the chemotherapy backbone. There were not significant differences in terms of all grades and Grade 3-plus events between the 2 arms. And ultimately, durvalumab was very, very well tolerated across all the adverse events experienced by the patients. Let me move now to DESTINY-Gastric04, that is the second presentation -- oral presentation in gastric cancer. This was Enhertu in HER2 positive. This is an important trial because it confirm how Enhertu is effective in these gastric cancer patients, second line, also because it needs to take into consideration that this was the best comparator arm. Ramucirumab, paclitaxel is probably the best standard of care you can have for patients in second line, and Enhertu was able to increase the overall survival more than 3 months. Very meaningful results, safety profile as expected. As you know, in gastric cancer, we have a 6.4-milligram dose. But overall, these data are reinforcing the positioning of Enhertu as a global standard of care in second line. Let me close with some Phase II data in another disease, that is cholangiocarcinoma, biliary tract cancer. We're very proud to have TOPAZ positive some time ago where we were able finally to bring to these patients with metastatic biliary tract cancer a new treatment that double the survival rate at 3 years. But despite that, the medical need is still huge. You see there are less than 50% of patients that are alive after 3 years with the TOPAZ regimen. This is why we want to improve upon it. And we want to do that using, again, a potentially better IO backbone. And these are the data that we presented at this conference with rilvegostomig, our PD-1 TIGIT in combination with standard chemotherapy, gemcitabine and cisplatin, that show a good median PFS, a good response rate with the durability of these responses. So you see the spider plot, please consider the study is ongoing. These are data that patients that continue to receive treatment. One important aspect is, as we learned with the rest of rilvegostomig program, the activity was even better in patients with tumors expressing PD-L1. And we -- the second important learning, the safety profile. Once more, rilvegostomig is very well combinable with standard chemotherapy. We have seen this in non-small cell lung cancer, in gastric cancer and now in biliary cancer. So this is a drug that will represent our backbone for most of the combination that we will run. So based on this data, as you know, we have now 2 Phase III studies ongoing, one in adjuvant with rilvegostomig versus placebo post surgery; another one in metastatic setting that is the combination of rilvegostomig and Enhertu in first-line BTC patient HER2 positive. Dave, I think you're going to conclude this session.

Thank you, Cristian. All right. So just one last slide for me to put into commercial context the data specifically from MATTERHORN that Cristian just walked us through. So we've been working really to build on the tumor presence that we have in lung and in breast cancer to really move into GI cancers. And we believe we've got a portfolio of cancers that really will allow us to advance transformative treatments that are going to change outcomes for patients, and MATTERHORN is really a great example of this. As mentioned earlier, this is our third perioperative opportunity for Imfinzi. This is relevant not just because of the number of times that we've been able to do this, but because of the experience that we're building, working together with multidisciplinary teams, surgeons, medical oncologists, all of those who come together to make decisions on how to best think about neoadjuvant, adjuvant, perioperative approaches to treating these patients that are in curative-intense settings. This has the potential to be the first and the only perioperative IO-based regimen in Stage II-IVa gastric/GEJ cancer. There are a large number of patients, 43,000 that are within the G7. And we know that 50% of those patients undergo perioperative treatment today. So perioperative treatment is the standard of care. I'd also note that it is important that FLOT is the primary standard of care being used. And in fact, more importantly, I think these data, and you heard this from the discussant yesterday, really underscore that FLOT should have an even increasing role as the backbone that we see across the globe for this very, very important treatment. And it's 55% today. I would expect that to grow in terms of the percent of patients that are getting FLOT, and that would be true across regions. This is a blockbuster opportunity for us. And really, again, I think that the discussant summarized quite nicely and well when the discussant speaks about D-FLOT and talks about D-FLOT because of the fact that the EFS and the OS is being seen across subgroups, where OS is obviously still at this point only a trend, but that's quite important. DESTINY-Gastric04, it supports the existing Enhertu indications that we have. It's already established, that is Enhertu is the standard of care in second-line HER2-positive gastric and GEJ in many countries. This reinforces confidence in Enhertu, converts conditional to full approvals in markets where we pursued an accelerated path to market. It enables a move into earlier lines. We've heard this consistently. More patients, therefore, have an opportunity to benefit the earlier you move up durations of therapy because of time to progression are longer, and this is good for patients and an important opportunity to, therefore, to make sure we're moving our best therapies as early as possible, which support reimbursement. And we're moving Enhertu into frontline gastric cancer. DESTINY-Gastric05 will come along with ARTEMIDE-Gastric01 in the 2026 time frame. So with that, we will move now to Q&A, same format that we had, had previously within this.

And without further ado, we can move straight into it. And please, a question in the back.

It's Ben Jackson from Jefferies. Two questions from me. The first, just coming back to the SERENA-4 discussion afterwards talking about the no crossover to the Enhertu -- sorry, to the camizestrant plus CDK4i in SERENA-6, talking about whether that causes any problems for assessing whether the benefit is actually seen when they progress on the ctDNA, the emergence of ESR1 versus actually the radiographic progression. Can you talk to us and just whether you see that as either a problem to uptake or a problem to whether that actually could be an issue with the FDA approval of the indication? And then secondly, just really quickly on SERENA-4, you're obviously looking at the combination with palbo there. Could you just talk us about the rationale and the belief in the -- a broad indication across all CDK4/6s?

Susan, so maybe you can speak to discussions and whether or not you anticipate the crossover is having any impact with agency. You can also then talk about palbo and SERENA-4. And Sunil, if you have anything to add on uptake for SERENA-6, that would be great. Susan?

Okay. So in terms of crossover, obviously, camizestrant is not yet approved anywhere. So you can't have it crossing over to a second line when you don't have it approved in the second line. It was possible for patients to obviously go on to second-line therapy as appropriate, and you saw that broken down in the trial design. But I don't think it's a relevant discussion. What you've heard from what Prof. Turner said is there's 2 things that really underpin the confidence in the approvability of SERENA-6. First is a very strong and very clinically meaningful improvement in the PFS 1 endpoint. Second is a very strong trend to PFS 2, which shows that you've got the carry-through to that second-line scenario of the effect that you saw in the first line. And the third is in terms of what patients feel of function, which is a key part of the regulatory approval thing, you're actually seeing an improvement in the quality of life, which is very durable because you're avoiding that progression on to symptomatic disease progression, and that is meaningful in this setting. I think it's further reinforced by the fact we have announced that we've got breakthrough therapy designation by the FDA. That is a recognition of the impact of the potential for this new method. And I think the final point that I would just say is if you -- that hasn't been discussed at great length is the safety profile that we saw in SERENA-6 is excellent. You've got a similar discontinuation rate to the discontinuations that you see on an aromatase inhibitor, some 1%. You really don't have any meaningful toxicities in terms of GI side effects or the other things that can cause issues. And I think that predicts well not just for the potential for the broader first-line indication, but also in the adjuvant setting. And it's really gratifying to see when we saw that about that profile because it really was at the top end of our set of expectations. So in terms of the relevance of SERENA-6 for SERENA-4, I'd just remind you that out of all of the second-line studies, the profile that we've seen in the SERENA-2 study, which was a Phase II study, looks really good, competitive with all of the others. We did see an overall benefit in the overall population, not just in the ESR1. And in the endocrine-sensitive wild-type population, this drug is also active, as you would expect, because it is a degrader of the estrogen receptor regardless of whether it's wild-type or ESR1. So again, we have to wait for the results of the SERENA-4. But given the profile that we've seen, I think we've got confidence going into the SERENA-4 in terms of the safety profile and the potential for efficacy. So I'm delighted with the camizestrant program.

Just specifically to answer your question on why palbo in SERENA-4, I think, because it was part of the question. I think when SERENA-4 was designed, palbociclib was the standard of care. And we were generating the combination data with ribociclib that we're not ready to go directly in Phase III. This is the reason why we were able to introduce ribociclib, abemaciclib and, of course, palbociclib in SERENA-6. But SERENA-4 would have been too late to amend and to introduce another CDK4/6 inhibitors.

What you did see in SERENA-6 is that there's a very similar effect size regardless of the -- across that group. So I think that's reassuring as well.

So just I think a brief summary point. I mean I think that the crossover point is really trying to get to a sequencing question. And so I think that it was pretty well addressed today in terms of the value of moving that decision into an upfront and not waiting, and that was at least the view that we heard from both Prof. Turner and Dr. Tolaney that were here. I think I'd also just like to add on SERENA-6, I mean a couple of points that I think are important. If approved, this will be the first next-generation SERD to move into the frontline setting. It will be, we hope, with a broad CDK4/6 label based upon the data that we saw within the study. ESR1 is part of routine blood testing that's taking place today, and it's the first opportunity to move into this $5 billion-plus opportunity that's in front of us. Okay. Please.

This is Zach Dunn from Guggenheim Securities on behalf of Seamus Fernandez. So I'm going to shift over to lung cancer. Hopefully, that's okay. So starting with the QCS data that was shown today, it looks like the NMR+ also selected for more non-squamous, and there was a slight increase in the percentage of PD-L expression high. So I'm wondering, considering those 2 factors, how does this play into your confidence from the data presented today overall in the QCS and using this as a biomarker? And then tagging on to that, what do you think is the bar for success in AVANZAR? And is KEYNOTE-189 a good reference point?

Okay. So Susan, I think those are good questions for you. I mean, again, remembering that we've focused our efforts on looking into non-squamous and refining also the QCS algorithm within that population. But with that, Susan, if you want to build on that?

Yes. Just a reminder, the primary endpoint is in the non-squamous population in AVANZAR. Obviously, the data that we saw in TL02 with the same cut point as we saw with the retrospective analysis from TL01 also showed an enrichment for both PFS and OS in the biomarker positive. And again, you would -- from the data that we had in TL01, you might expect a bigger effect in non-squamous than in squamous, although those numbers are small. I think that overall profile, therefore, predict -- increases the confidence just because you've now looked at data not just as monotherapy, but with the IO combination with and without platinum for both PFS and OS in a first-line setting, right? Those numbers are still small, of course, right? So we have to wait for the AVANZAR data, but that's what gives us confidence. In terms of the bar for success, again, AVANZAR has a primary endpoint in both the ITT and in the biomarker positive part of the population. We increased the sample size following the results of the TL01 and focused on the non-squamous so that we're adequately powered for PFS in both of those analyses, and I think that's what gives us confidence. And yes, we would have -- expect that we need to beat the IO plus chemo standard of care backbone in that setting.

Can I add something, Susan? The bar KEYNOTE-189, of course, is a benchmark, but it's not the only benchmark because it's important to remember that AVANZAR is probably enriched for a population that is PD-L1 medium expressed or low or negative expressed because, of course, in more than 50, here, you have chemo backbone in both arms. So there are -- we expected to have less high PD-L1 positive patients. So it needs to be taken into consideration when you think how you benchmark the comparator arm.

Thanks, Cristian. So just for clarity, when we talk about the ITT population in AVANZAR, it's non-squamous. And so we've got 2 shots on goal in that study. There's the non-squamous and the subset of non-squamous that's the biomarker positive. Right. Okay. Other questions? And we can also take questions online, of course. Right behind you here.

A short one. rilvegostomig, Susan, I'm starting to come around. The BTC data, you previously said the PD-L1 high is kind of an important indicator of where to use it, but the data look very, very good in BTC that we saw here. Is that still a factor? Or are you evolving in your thinking? And what do you expect OS to look like post PD in BTC?

Post PD in BTC, okay. All right. So just a couple of comments on rilvegostomig in general, right? Obviously, it's a bispecific. So the intent is that this antibody will bind both targets on the same cell, which is different from co-administration of 2 different separate antibodies and can produce different biological effects. We did publish a poster at SITC last year, looking at an ex vivo model system that we call the enable system, which basically takes patient-derived tissue from organoids that you can grow for a few number of days. We used interferon gamma as a measure of the potential for T-cell activation in that setting. That shows, in that context, the importance of that bispecific modality. And I think we'll continue to build out that data set and show that. I think it also shows the importance of the Fc attenuated design of rilvegostomig, which I think may also help to explain its differentiated safety profile, particularly in combination with chemotherapy agents. So the data that we've then shown, I mean, the relevance of PD-L1 high as a biomarker for predicting the response to IO differs by tumor type, I would think. And of course, we do have the durvalumab indication in biliary tract in the first line based on the TOPAZ study. So what we were looking for in the trial here was something that could improve on that. And I don't know if you want to make any comment, Cristian, about the...

Yes. I think rilvegostomig require a little bit more discussion. I think we are -- Susan explained very clearly the differentiation that this bispecific bring versus when you put the 2 antibodies together, okay? This drug has an Fc that is reduced. This is also can explain some of the efficacy and safety that we are observing. We have now 4 data sets. We have ARTEMIDE-01 that is the monotherapy data in frontline non-small cell lung cancer. We see very compelling activity, more than 50 in terms of response rate in PFS. We see good activity, very good activity, superior what you expect to a PD-1 in 149 response rate in PFS. We see good activity in gastric cancer in combination with standard chemo in biliary, as you see before. And we will start to generate data also in combination with our ADCs. So these are the data set that, of course, along the year, we will release when they are ready to be presented. But this is reinforcing that what we are doing is supported by data. The drug is very well combinable with standard chemotherapy with ADCs. And currently, we have 8 pivotal studies ongoing with this drug as a backbone.

Thank you, Cristian. I'm going to take 2 calls -- or questions, excuse me, online, and then, Emily, we'll come back to you in the room. So first, Colin White from UBS.

Colin White from UBS here. When the discussant from SERENA-6 was talking about the challenges in assessing clinical utility in the study, she mentioned imbalances in post-progression treatment and imbalances between the arms and the lines of therapy, which could impact the PFS2 measures and OS data. And not thinking about in terms of approval, but I was just wondering if you could comment on whether or not that might -- you think that might impact the perception of the robustness of those measures when we get the data. And then the second question I had was on DESTINY-Breast09. The discussant seemed to be suggesting that it might be -- it could be used in patients that might do less well on THP. Is that your expectation of how Enhertu will at least initially be used in the first-line setting? Or do you anticipate rapid uptake?

Thank you. Cristian and Sunil, maybe I ask you both to answer those questions.

Okay. Maybe I can get started and, Cristian, please, jump in. So let me take the D-B09, and it reminds me of last year's late-breaking abstract with D-B06. If you may recall, we had a very similar discussion about the discussant saying that there may be patient subgroup and patient segments where D-B06 may be used with Enhertu versus standard of care with chemotherapy. And I think what we have seen is we have seen a strong uptake of D-B06 in HR-positive in a broader patient population. And I think we heard the same in the discussant here. Initially, there is sort of who is the right patient population. And as clinicians get comfortable, that expands. And I think the data that Dr. Tolaney shows was very definitive for PFS, for complete response rate and, of course, for PFS2, suggesting a strong trend for overall survival. So that, I think, is going to lead to a broader uptake. There is a segment of HR-positive where [ patina ] consideration is there. But we again think that upfront use of Enhertu is the right approach, what we're hearing from the clinicians so that clinicians don't have to wait to see whether patients progress or not and they use the best therapy upfront. So I think that definitive data is very strong. I think for SERENA-6, and it's related to the question that was asked previously, you can't really force somebody in the second-line HR-positive setting to have a defined treatment. There's multiple therapeutic options available for that patient, including endocrine therapy, including chemotherapy and, of course, ADC. And what we saw from SERENA-6 is a representative patient demographic and a representative distribution of the type of therapies that are available. So I think that is generally reflective of the current clinical care rather than a forced that I don't think would be well received in the clinical trial setting or in interpretation of this study. So I think both of those factors lead to a much more generalizable result that we have seen from SERENA-6.

Just one more comment on that. Progression-free survival is measured by definition, by progression events, not by the treatment changes. So if you're getting the overall time on before you're actually having that progression, that really matters because you heard from Prof. Turner, it matters because actually patients get sicker and lose quality of life when they progress. And pain is an important part of that. A lot of these patients progress due to bone metastases, and that induces pain. And that affects your quality of life from the moment that you have that. So I think when you're looking at it, I don't think it diminishes the confidence in PFS2. It's still a relevant point. Obviously, SERENA-6 is a relatively small study arm. So in terms of the powering for an overall survival benefit in such a patient population, you're unlikely to have adequate power for that. But what I think is going to be needed is some more maturity on the PFS2 and the trend to OS. And again, what the shift is that we're seeing in that curve on PFS2 gives you confidence that you're going to be able to achieve both of those.

I encourage you to also go back and listen to the discussant from the FLAURA conversation. We also engaged in a sequencing conversation at that point. It's not uncommon for the question to be asked. I do think that what is particularly unique here, and it's powerful because Prof. Turner said it and not us, to hear about the concept of cure being incorporated into advanced breast cancer in the HER2-positive setting is a pretty remarkable place to have been able to arrive to. And so the fact that you don't know who those patients are going to be that are going to get the 15% complete response, I think, is something that's going to drive a great urgency to make sure that the best therapy, the best regimen is being used upfront. And I think we're also, on SERENA-6, going to see that the patient groups are going to be keen on making sure that they have the opportunity to avoid a progression conversation. Just to make this human, when you talk to patients, the worst news that they get is the news that they've been diagnosed with cancer. The second worst news that they get is if their disease has progressed, and they have to get that news alongside all of the other physical realities that they go through. Rajesh Kumar, over to you for the next question, please.

First one is on SERENA-6. Thank you very much for explaining why you're confident about approval. If approved, how do you perceive the adoption curve would be, especially given that you need to incorporate a diagnostic in the pathway? So do you anticipate a slow initial ramp-up followed with a network effect, which ramps up a bit faster? Or do you expect a more linear ramp-up as you already have plans to incorporate diagnostics in the next 2 or 3 years? That's the first question. And second, slightly off topic. Thank you for all the various presentations you've done during ASCO. Just listening to a lot of presentations and recent deal flows, could you help us understand how your next-generation bispecific strategy is altered by all the news flow around VEGF? Clearly, there's a lot of excitement around the mechanism. Would love to understand what your thoughts are on the mechanism and how do you see that space evolve? And how do you intend to play in that space?

Thanks, Rajesh. Much appreciated. So we've got 5 minutes left. We're going to try to answer those 2 questions and 2 more in the room. So first, Sunil, can you comment on expectation for adoption curve and what we'll be focused on in order to drive up that? And then Susan, I suggest for you to handle the VEGF bispecific question.

So I'll keep it short. So maybe just 3 quick points. I think the first one from clinical adoption, there's a clinical perspective and a patient perspective. And I think, Dave, you highlighted the importance from a patient perspective, they want to take control of their disease. They don't want to be controlled by cancer, but they want to take control of their disease. And there's strong interest. We saw that in clinical trial enrollment and we see here from the patients. And that, I think, adoption is coming from the patients. The second thing that I want to highlight is the ESR1 testing is routinely done right now with more than 60% of patients getting that upon progression. So it's already established. And now, as Dr. Turner said, this could be integrated as part of routine blood test every 2 to 3 months. And then the final thing about adoption, you have 2 different populations where the benefit can be seen and they were enrolled. So one is the incident patient population that is newly diagnosed and are on AI CDK4/6 and not progressed in the first 6 months. And then the second population, which is a prevalent patient population, those patients who are 6 months and up to 2, 3, 4 years and have not progressed. And it's important to note that in this study, the first test showed ESR1 positivity in 50% of the patients. So I think that's going to drive adoption as well. So I think those are the key tailwinds that will lead to hopefully rapid uptake.

Thanks, Sunil. Susan?

Okay. So again, our bispecific strategy is focusing rilvegostomig on the more IO-sensitive tumors in some of the combinations and volrustomig as the PD-1 CTLA-4 in the PD-L1 lower or particularly CTLA-4 sensitive. There are some tumor types that have some sensitivity to VEGF. We are testing the combination of rilvegostomig VEGF-directed therapies in HCC, for example, on lung cancer in Phase II studies. It's not yet clear that the bispecifics give you differentiated biology versus the combination of an IO and a VEGF together. And actually, the data that we have from rilvegostomig in the ARTEMIDE-01 study have very similar trend to PFS as what you see with ivonescimab in the PD-L1 high group. So I think there are multiple approaches to improving on the current standard of care. And by focusing the bispecifics on the different segments, we hope to optimize the profile that we have there. And where necessary, we'll look at the combination with VEGF in VEGF-sensitive tumor types.

Thank you, Susan. Yes, please.

My name is Miki Sogi from Bernstein. I have a question regarding the TL02, the QCS analysis. And in the abstract, you shared with us the data from the subpopulation of non-squamous, squamous without actual QCS. And that data really showed a really high promise for the -- based on the duration of response, that was much stronger comparing to PFS. So we were really hoping that to see the QCS analysis within non-squamous to see the actually duration of response actually translating into the PFS in -- with this the further subpopulation analysis. So my question is, first of all, why you didn't disclose that? It was probably too small a sample size. And if that's the case, and now we are, of course, all expecting for AVANZAR results coming in the second half of the year. So which data we should be really focusing on?

Okay. So again, I think the TL02 data set is best looked at really as an overall group because otherwise, if you're sub-setting it, it's hard to interpret endpoints like PFS in 2 small groups. So what you saw overall is that as a doublet or a triplet for both PFS and OS, you're seeing an improvement when you look at the QCS-positive group and enrichment there. And I think that gives us confidence, again, as I said, because it's now a first-line population, it's an IO combination with or without chemotherapy that you still show that enrichment for QCS. Again, we focused on the non-squamous in AVANZAR. So you're right that, that is the more relevant subgroup. But I think that's the effect size that you expect to see. And if you just subset by non-squamous within the doublet and the triplet, you really do get down to tiny numbers.

Thank you.

So -- and also on TL02, another thing that was very striking was really strong doublet data. And I actually asked this question just earlier to Daiichi Sankyo team, whether they are planning to do a doublet in the trial in first-line setting. And their answer was, well, we are already doing this with the TL07, but that doesn't include a QCS analysis. So I just wanted to ask the same question to the AstraZeneca team.

So again, just a reminder, the rationale for including platinum is to increase the depth of response, which may be particularly important in the PD-L1 low group, which is what you'd expect to end up with more of in the AVANZAR study. And also remember the cross-comparison between the doublet and the triplet is challenging because there were actually quite different patient populations that were enrolled. There's a higher rate of brain metastases in the patients in the triplet and actually a much higher proportion of those patients had a lower dose, the 4 mg per kg dose rather than the 6 mg per kg. So I don't think it's reasonable to project that the doublet is superior in any way. I just think that was just the nature of those cohorts that was in TL02.

Susan, there's a difference in platinums as well, yes?

That's also an important point, right, which is that TL02 included a significant proportion of patients that were on cisplatin. We're including the carboplatin, which is best tolerated in the AVANZAR study design. Thanks, Dave.

All right. Perfect. Emily, if we can be fast, we'll fit you in.

Well, and maybe switching gears, I know you haven't had that long, but how has commercial uptake been in the other perioperative Imfinzi approvals of NIAGARA and AEGEAN? And how would you expect this to be similar or different from MATTERHORN?

So on both of those, we've been really pleased with the pace of uptake that we've been seeing. We see, as we would have hoped to have seen, that when you move into earlier settings that there is an urgency to bring on the best therapies possible. AEGEAN is doing well and contributing, as is NIAGARA, and I think that we'll look forward to having an opportunity to share more on sales trends and some KPIs that are there, but both have been moving well. And I also think that I expect that there will be conversations on Tuesday morning about MATTERHORN happening as well in practices certainly within the U.S. So with that, I want to end where we started, which is at the beginning of this year, we talked about this being a catalyst-rich year. We came out of an ASCO where we had an opportunity to talk about 3 of those catalysts that were all from positive studies in really important areas, both for patients and the outcomes, but also in terms of the sales opportunity that they represent with SERENA-6, DESTINY-Breast09 and MATTERHORN, together with some of the other important readouts that we've already seen, high-level results that we know that we've announced from DB11 from POTOMAC. Those are very, very exciting places for us to be as we then set our chart for 2030 and the progress that we want to make. Thank you all very, very much for joining us. Safe travels back from Chicago. We appreciate it. Bye.