AstraZeneca PLC (AZN) Earnings Call Transcript & Summary

Welcome, ladies and gentlemen, to AstraZeneca's Capital Markets event, Meet AZN management ASCO 2022. 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 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 at this event reflect the knowledge and information available at the time of this event. 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 the event. And with that, I will now hand the conference over to Chief Executive Officer at AstraZeneca, Pascal Soriot.

Good evening, everybody. Thank you so much for joining us tonight. It's really a pleasure to have you here. So my name is Pascal Soriot. I'm the CEO of AstraZeneca. And it's been really a fantastic experience to be back here in Chicago, not actually the wait for immigration, but meeting people here over the last few days has been really, really nice after 2 years of video conferencing. And of course, as you can imagine, we are also excited with the data we presented and the response we got to those data yesterday. So thank you again for joining us tonight. This is our forward-looking statement. These are the speakers for tonight. So first of all, we have Dr. Aleix Prat, who is joining us and he's one of our principal investigators. He is an oncologist that [indiscernible] in Barcelona. Dave Fredrickson is our EVP of the Oncology business unit. Cristian Massacesi is our Chief Medical Officer and the Head of Oncology Development. Mika is there. Mika Sovak is our Vice President in charge of the global franchise for Enhertu and she is the person who has carried this program over the last period of time and did a fantastic job. Susan Galbraith, many of you know, Susan is Head of R&D for Oncology. Some of you may not know Sunil Verma, who is there. And Sunil is the head of -- SVP, Head of Oncology Medical. Before this, Sunil was in charge of clinical development for Enhertu in breast cancer as a whole. And Sunil is one of the great opioid leaders in breast cancer. We have in this room, actually so many opioid leaders in lung cancer, in breast cancer. So it's really been a great couple of years having really fantastic people working with us building our Oncology portfolio. And finally, a newcomer Andy Barnett, who is our new Head of Investor Relations. Andy, if you want to stand up so people can see you. Andy joined us a couple of years ago from Genentech and was in the Oncology business before taking this new role as Head of IR. We believe it is important to understand our products and the science, if you want to be head of IR. So I'm sure Andy will do a great job. So this is the agenda. I'll share with you a couple of slides, then Susan will tell you the story for Enhertu as we see it so far. Dr. Prat will take you through the DESTINY-Breast04 trial results. Dave will take us through the consequences of the implications of those data and what we intend to do in cancer, in particular in breast cancer. And finally, Susan will come back to give you an outline of what we have in Oncology over the next few years. And of course, we have the Q&A. So if you look at our company in Oncology, what we pursue in Oncology is very similar to what we try to do across the overall company. First of all, we have, across the company, a strong last-stage pipeline, but certainly very much in Oncology. In oncology, we have 7 NMEs, we have 90 projects, including NMEs and LCMs in development in Phase II and Phase III. In fact, we have, in total, about 18 products either in development or in the market in Oncology including older products, of course, as you would imagine, Faslodex and Zoladex. But there's certainly a very broad, very strong portfolio of products. These are older products and new and, in fact, very new products that some of them are in Phase II. We have a very innovative early-stage R&D, and we're working on a number of technologies, ADCs in particular, and of course, in HER2 and Dato-DXd are the leading 2 agents that we licensed, and we collaborate with our friends at Daiichi Sankyo, but we have our own program of antibody drug conjugates that we have been developing over the last 2 years. We also have a portfolio of bispecifics. We are working on products. We are now officially a cell therapy company because we started a cell therapy program in the clinic a few days ago, Susan, I think. And we're also, of course, looking at combinations. Oncology is becoming incredibly exciting, but also incredibly complicated, as you can imagine, the combinations in particular and how do you combine agents and also how you find the right patients is becoming certainly a very, very sophisticated and very complicated exercise and it's really good to have the talent we have with us and people who are really expert in their respective cancer types to help us guide us through this. We're also working on digital health and certainly looking at multiomics as we consider our research programs. Then we have continued working on business development. As you can see here, there's a few examples of deals we've done recently. And finally, and importantly, I think the message we try to tell people across the whole company and the whole pipeline, but it's very much true in Oncology is that we have -- as a company, we have -- sometimes say, we've already given to this challenge of patent expiries. So moving forward, our patent expiry profile is actually pretty good. And in Oncology, we have one product that is losing patent protection before 2030. And it's Lynparza, and we are working very hard on the PARP1, which I'm sure you will have questions about. So a very, very good patent protection or profile over the next 8 to 10 years. It's also true outside Oncology. We lose patent protection for Farxiga in 2006 -- 2026, 2027, depending on the geography, and we're working on combination products. And we are also losing patent protection for Soliris in rare disease, but we are switching this to Ultomiris as you know. So overall, as a company, patent profile is pretty good. And that's why we believe that for all of the reasons you see here, we can continue to grow at what we call industry-leading growth rate post 2025. Focusing on the ASCO. Over the last 4 years, we've had a plenary presentation every year, if you look at it, 2019, '20, '21, '22, and the progress we've made as a company coming to the ASCO over the last 8 to 10 years has been tremendous. And this year, we had 108 abstracts. I think it's important to remember that compared to a few years ago, companies actually present some of their abstracts at different congresses, ASCO GU, ESMO, et cetera, et cetera. So you don't have this mass of abstracts that certainly I was used to, when I was at Roche coming to the ASCO. But certainly, 100 net abstracts is a very large number. We had 9 old presentations and of course, the 1 plenary presentation that we all know about. And we presented the data on HER2, Calquence, MEDI5752 are bispecific CTLA-4 in renal cell cancer and also an update on -- safety update for Lynparza and prostate cancer from the PROpel study. So a very busy ASCO and certainly, of course, the highlight for us this year was DESTINY-Breast04. So with this, I'll hand over to Susan. Thank you.

Thank you, Pascal, and welcome, everybody. Thank you very much for coming and spending some time with us. So let me start by putting a bit of background on the design of Enhertu. So Enhertu is, we believe, a best-in-class HER2 directed antibody drug conjugate, and it's not by accident, it's by design. So the industry has been putting antibody-drug conjugates into the clinic for over 20 years now, but it's an important system design. You got to have the right antibody that binds to a target that internalizes rapidly. We've learned that having a high drug antibody ratio is an important feature with the right kind of potent warhead and a warhead that has the ability to diffuse across the cell membrane so it can generate bystandard activity. All cancers are heterogeneous. As you go down in the level of expression that heterogeneity becomes increasingly important. I believe that's one of the things that's been important in the HER2-low setting. In addition, the design of the linker is critically important. You need a linker that is stable in the peripheral circulation so that whole molecule stays together in the peripheral circulation. So you don't release the free warhead until you get to the tumor, and then it must be cleavable and allow this diffusion across the cell membrane of the cell permeable warhead. So with that design, we've already achieved some key goals that we had at the time that we've done the deal with Daiichi Sankyo. So we've shown already data that's shown that we can improve on established standard of care in HER2-positive or HER2-high metastatic breast cancer. We've been able to expand the activity and see activity across a range of different tumor types that also overexpress HER2 or have a HER2 mutation. And I believe with the DESTINY-Breast04 data that was presented yesterday, we're able to now see an area where we can redefine breast cancer classification and just use this new segment that's targetable with this drug of HER2-low in HR-positive and HR-negative disease. We still got an extensive program that's delivering for Enhertu and our goals are to improve the long-term survival and cure rate in early-stage HER2-high and HER2-low disease. And there are multiple combinations that we think will open up further possibilities for Enhertu including combinations with immunotherapies in combination with our DNA damage response portfolio. So as I've said, if you think about the core strategy for Enhertu in 3 core pillars: the HER2-positive or HER2-high, where we've moved from late-line metastatic into the early line, the HER2 breast -- HER2-low breast cancer setting, where we think with the data that we've seen so far, we can actually offer the opportunity to replace standard of care chemotherapy and potentially also later line endocrine therapy in patients that have no longer highly endocrine-sensitive. And then we have a program beyond breast cancer. So we've already got approvals in gastric cancer. We've got great activity in lung cancer. We've seen activity in colorectal. And I think the discussion yesterday showed some other data from other tumor types that also express HER2. So let's just focus on breast cancer for a second, though, and just say that Enhertu now shows the potential for activity across the spectrum of HER2 expression. So I've talked about HER2 positive or HER2-high. So around 20% of breast cancer. HER2-low represents around 50% of the patient population. And then there's the -- what's currently called IHC 0, which can be HER2-ultra low or null. And on the right-hand side of this slide, you can see the immunohistochemistry images that are associated with these different segments. Now we can also do sensitive mass spectrometry, which gives you a sense of the number of receptors per cell in these different categories, just to put a numeric aspect of it. So in the HER2-high or amplified, typically, you have 2 million or more receptors on the cell surface. In the HER2-low breast cancer, if you're looking at the 2 plus, the range is in the range of 200,000 to 500,000 receptors per cell. 1 plus is around 100,000 to 200,000. And then in this what's called currently IHC 0, you don't have 0 receptors on the cell surface. What we call ultralow where you can see any level of brown you might have in the range of 20,000 to 100,000 receptors per cell and their normal tissue, normal breast epithelium, we'll have around 20,000 receptors per cell. So you can think of -- in these different categories. And that's why you think there's the opportunity to think about what you might do in that ultralow segment. So again, a little bit of context of the data that we had in hand at Daiichi Sankyo and AstraZeneca before the design of the DESTINY-Breast04 study. And it was the data from the J101 Phase I trial in heavily pretreated patients that paved the way for the design of this study. And what they -- what we're seeing was around a 44% response rate with good durability of response of around 8 months, the median progression-free survival of 7.6 months in a really heavily peak population that went across both IHC 2+ and 1 plus. And again, you can see from the spider plot, the durability of those responses. And again, importantly, efficacy was seen in both CDK4/6 inhibitor naive and pretreated patient populations, and there was good safety consistent with the rest of the program. So that was the background to the design of this study. And so with that, I'm now going to hand over to Dr. Aleix Prat, who joined us. He's a senior investigator on the DESTINY-Breast04 trial, part of the Steering Committee and the Head of Medical Oncology at the Hospital Clinic of Barcelona. Dr. Prat?

Thank you so much, Susan, for the kind introduction. It is my pleasure to be here today to share the results of the DESTINY-04 and also provide you my opinion. That although probably my opinion is important to you, it is nothing, I think, compared to what we saw yesterday during the plenary session where we saw a standing ovation for minutes, thousands of patients, thousands of medical oncologists, but not only breast oncologists, other oncologists for other cancer types as well as patients, patient advocates and researchers and many people from the industry. So I think that moment speaks by itself. So not much I can say. However, I would like to bring some thoughts about the study. As you probably know, this study focuses on a population of patients where we have an unmet need. I think this is important to highlight. These are patients that are classified as HER2-low, therefore, HER2-negative. And these are patients that have been already treated with standard therapy. The hormone receptor-positive, specifically with different endocrine therapy strategies and even 1 or 2 agents of chemotherapy. The triple-negative patients receiving prior treatment with 1 or 2 lines of chemotherapy. So in this particular situation, when we are in the clinic, we face a difficult moment. These patients, in general, what we can offer to them is single-agent chemotherapy. Agents that, in general, can have activity for less than 6 months. And these patients usually have a median overall survival of around 1.5 a year. So difficult situation we're facing this year. So the study focused on this population and randomize those patients, as you can see, 221 to T-DXd, standard dose versus treatment physicians choice for the 5 different chemotherapies single agent. The primary objective of this study was progression-free survival in the patients with hormone receptor-positive disease and key secondary objectives were progression-free survival in all patients and also a very important overall survival in hormone receptor-positive as well as all patients. Just briefly, the baseline characteristics, that the patients that were recruited, we're talking about more than 550 patients in international studies you can imagine. I would like to highlight a few things about the baseline characteristics, which are important. One is the levels of HER2-low disease, meaning 1 plus versus 2 plus each negative. You can see the distribution, 58% 1 plus, 42% 2 plus each negative. Also, I would like to highlight hormone receptor status as planned by the study, the vast majority of patients, 89% had hormone receptor-positive disease and 10% around 60 patients had triple-negative HER2-low disease. Also I'd like to highlight that we're talking as expected, the population that has a lot of tumor burden. As you can see regarding liver metastasis, 3 out of 4 patients had liver metastasis and 1 out of 3 patients approximately had lung metastasis. These patients that I mentioned were previously treated with standard therapies. The median lines that these patients receive of therapy in the metastatic context was 3, and in particular, I would like to highlight that 60% received 1 line of prior chemotherapy, but 1 out of 3 patients received 2 lines of previous chemotherapy, which is substantial. And finally, one aspect also of importance because usually, these patients that are hormone receptor-positive are treated today with endocrine therapy and CDK4/6 inhibitors in first line, mostly, in this study, 70% 3 or 4 patients approximately received prior line -- prior CDK4/6 inhibitor. Regarding the results, on one hand, the primary objective, progression-free survival. You can see here the survival curves in hormone receptor-positive disease in the middle intend to treat population or ITT and also the population -- the triple-negative population on the right. You can see that the survival course look extremely similar and the survival of course, separate very early on, which I think something to highlight. I'm talking about increases in median overall survival of going from 5.4 months, as I mentioned, these patients usually progress in less than 6 months, now going up to 10.1 months with T-DXd with a hazard ratio of 0.51, meaning that they have a decrease in the rate of progression of 49%. Intensity population looks exactly practically the same. And I would like to focus on these 58 patients that are triple-negative, very aggressive tumors. You can see the control arm with a median progression-free survival of only 2.9 months now increasing their median overall survival to 8.5 months with a hazard ratio below 0.50 -- 0.46. So clearly, high activity across all patients and independently of hormone receptor status. But also progression-free survival is important. No doubt it was the primary objective. But at the end of the day, we want to increase the chances of patients being alive. Overall survival is critical. We are not so used to seeing increases in overall survival in Oncology and in breast cancer, in particular, in the HER2-negative space. Here, you can see the separation of the curves, we're talking about increases in median overall survival in hormone receptor-positive disease of more than 6 months with a hazard ratio of 0.64, similarly in intention to treat population. And again, going into the 58 patients with triple-negative disease with a median overall survival in the control arm of only 8.3 months, it increased to 18.2 months. So a delta of more than 10 months in this group of patients with highly aggressive disease and has a ratio below 0.50. So very, very impressive results. And not only from a PFS perspective, but also, and more importantly, from an overall survival perspective. One particular question to ask is, if there's any baseline clinical pathological feature that might explain or differences in response? The answer clearly is no. And in particular, for example, if the patient received prior CDK4/6 inhibitors matters? The answer is clearly no. Whether the levels of HER2, 1 plus versus 2 plus matter? The answer is clearly no. And there's not even a hint that there is a gradient there. Prior line of chemo could be also something to evaluate 1 line versus 2 lines or more. And you can see here that the hazard ratios are practically the same. And other features you can see here, including visceral disease, does not matter. So definitely, no baseline feature predicts response to this therapy. And finally, and also very important is safety and tolerability. Here, you can see the summary, of course, patients that received T-DXd were -- took longer than with the control arm. So this is something to account. But even that, I would like to take you to the treatment events, adverse events and specifically the Grade 3 or more. You can see the numerical percentage is lower with T-DXd, 53% versus 67% in the control arm. Also, I would like to point out that in terms of ILDs, interstitial lung disease, this was found to be consistent with prior studies, around 10% had Grade 1 and Grade 2. And there were a few cases only of Grade 3 and only 3 cases with Grade 5 disease related to ILDs. So to me, in my opinion, this drug establishes a new standard of care in HER2-low, hormone receptor-positive as well as hormone receptor-negative advanced breast cancer that has received prior therapy, endocrine therapy and also chemotherapy. Similar magnitude of benefit is being observed across shop groups, including the levels of HER2 do not matter or if the patient received prior CDK6 -- CDK4/6 inhibitor or not. Safety is overall consistent with a known safety profile and showed an overall positive benefit at risk. And here, you can see the summary also of the activity that I mentioned. So thank you very much, and I'm very happy later to discuss and give you also my opinion. Thank you.

Thank you very much, Dr. Prat, much appreciated on that. So just really now to talk about what the implications are from these data and how we see them having an opportunity to truly transform cancer treatment. And where we have an opportunity to then after this, think about where we can go with Enhertu. So as you heard from Susan in the opening, Enhertu continues to demonstrate the ability to transform breast cancer. We really set out with the first hypothesis being can we transform in HER2-positive disease. And I think that we can very clearly say that with DESTINY-Breast01 and 03 that we've seen, really very, very compelling data from the early study but then confirmed through DESTINY-Breast03 with a hazard ratio of 0.28 and a PFS that hasn't even been reached yet, that we're seeing really incredibly compelling data there. This has also resulted by the way, in approvals in 40 markets across the globe, some with DB01, some with 03 and I think that we're doing incredibly well in terms of the commercial success that we're having with HER2 disease so far. But now today, we talk about DESTINY-Breast04. And again, here, I think you can see that with this, as again, we spoke to and Susan laid out before, with a median progression-free survival of 9.9 months and the intent to treat population, really, really now reaching more than half of breast cancer patients. Susan talked through the dimensions of how we can think about breaking down kind of how HER2-low and what percentage of the population it comprises. We know that HER2-positive disease represents about 20% of all advanced breast cancer. HER2-low, inclusive of hormone receptor-positive as well as the previously identified triple negative or hormone receptor-negative represents about 50%. Ultralow, meaning those that are between 0 and IHC 1+ is around 20%, and then there's about 10% that are in the null category. So that's how we think about this breaking down and as Dr. Prat spoke to, we know that outcomes are incredibly poor, unfortunately within this setting. I think the really important piece to remember is for all the advances that we've made in breast cancer, and we've made many. Unfortunately, many, many women with advanced disease, most women with advanced disease find themselves eventually on some type of classical chemotherapy. And this study is really looking at, can we replace that classical chemotherapy with a more precise better option. And so it's with that backdrop that we put the Enhertu data into context. This is just for hormone receptor positive studies, and it's an effort to try to take a look at across real-world data as well as Phase III studies and available information. What is the expected baseline performance in terms of median progression-free survival that we were seeing and expecting from standard of care? And then how does that actually get affected by Enhertu? What you can see is whether it's with chemotherapy or endocrine recycling that many of the studies really were between somewhere 2.5 to 5.5 months of progression-free survival. We also saw the TROPICS-02 data with SASE, where we saw in a fourth line plus population, 5.5 months of median progression-free survival. And so it's within that context that we do think that you begin to get an understanding for one of the reasons that I think that Dr. Modi got the response that she did yesterday, 10 months within this third line POS population. I think an important other thing just to put this into context is what we do know is that this establishes a new standard of care. And I think that what I'd like to orient you to on this chart is there's a couple of important dynamics that do take place. So while it is true that women with advanced breast cancer have relatively poor 5-year survival rates, they do cycle through many lines of therapy for their advanced or metastatic disease. The pink or purple represents endocrine therapy. The blue represents chemotherapy with the darker blue that sits at the bottom being their first -- Sorry, with the lighter blue representing their first chemo and the darker blue representing their second chemo. So the couple of dynamics I want to show you, most women with in their first line that are hormone receptor-positive are being treated with endocrine therapies. Very few of them are being treated with chemotherapy. As we progress through line by line, we do see endocrine recycling happening, but we also see the introduction of chemotherapy taking place. The light blue represents -- or the light blue represents that first chemotherapy, which will have an opportunity to look at that in DESTINY-Breast06. The DESTINY-Breast04 population is this darker blue. And I think that the questions that you'll hear physicians begin to ask is, based on 04, can we start thinking about actually moving to first chemo? Can we think about moving lower? These are questions that are anticipated by 06 and that we're looking forward to reading them. So it's really with that, that we see ourselves as having a great opportunity to be able to really have a lot of commercial success as soon as we're able to work through to approval. And with that, I turn it over to Susan to talk about what's next in Oncology. Thanks, Susan.

Okay. So we got the -- HER2 has got some great data there, and we've got the great promise of that, but it's not the only antibody drug conjugate we've got in the portfolio. We think we're also very excited about datopotamab deruxtecan. In particular, we think that this is a drug that can complement the range of patients that can be treated that are where you're directing a [indiscernible] versus directing HER2. So let me address some data that was presented recently at the ESMO Breast meeting with the BEGONIA, which is a platform study in triple-negative breast cancer, which looked at the combination of datopotamab deruxtecan plus Imfinzi and reported a 74% response rate, again, with good durability of response. Now giving you some of the context about the numbers of receptors per cell for HER2 across the range of cancers, just to put it into context, the TROP2 to expression, in HER2-low and HER2 null tumor cells could have around 5x more TROP2 receptors per cell than HER2 using, again, that kind of mass spec technology. So again, I think you can see that there's potential complementarity between these 2 antibody drug conjugates. Obviously, the prevalence of low HER2 is lower in the HR-negative group than it is in the HR-positive and in that HR-negative group that currently called triple-negative. We have the TROPION-BREAST02 Phase III study, which is already ongoing, which is targeting to -- for this patient population. And again, comparing with datopotamab deruxtecan versus investigator's choice of chemotherapy in patients with locally recurrent inoperable or metastatic breast cancer who've had no prior chemotherapy to the first line setting and who are not a candidate for PD-1 or PD-L1 inhibitor therapy. And we expect the readout of this beyond 2023. So when we look back at that overall map of breast cancer, how do all of the molecules that we've got addressing breast cancer fit into this spectrum? So first of all, let me address you to the left-hand side of this slide. So I think the current classification of breast cancer, which has had the HER2-positive, hormone receptor-positive, HER2-negative and triple-negative best cancer, we need to change our terminology. So I would suggest the terminology that we could think of as HER2-high, which is, again, that 20% of patients and you see in HER2 in that segment. In the hormone receptor-positive group, I think we're going to have to further classify patients by other characteristics, whether they're HER2-low or ultralow, whether they're TROP2, endocrine sensitive or resistant or DDR targetable. And you can see the portfolio coming through here. So you can see, first of all, the box, second from the right on the slide for the blue box of Enhertu in that DESTINY-Breast04 setting. And then just adjacent to it one to the left is the DESTINY-Breast06 setting. And again, I think datopotamab deruxtecan will have activity in this HR-positive setting as well. But again, one of the things that we need to think about is the patients that are best selected for TROP2 directed therapy versus a HER2-directed therapy. Also within the HR-negative group, we can also subdivide by HER2-low, potentially TROP2-positive IO or DDR targetable. And again, you can see the potential to move both monotherapy Dato-DXd in this space and also potentially in combination with IO, as you've seen from the BEGONIA 7 study. We do think that there's still a large segment of endocrine-sensitive breast cancer, where the oral served drug camizestrant can play a big role. We've got, obviously, Phase III trials ongoing in the first-line metastasis setting looking at the combination of camizestrant with CDK4/6 inhibition and the potential to move this molecule into the early stages of disease. And we look forward to the readout in the next several months or years of the capivasertib studies in both HR-positive and HR-negative disease. So just again an update on the datopotamab deruxtecan program. And again, let me start with the non-small cell lung cancer setting, where you can see we have already got the TROPION-Lung01 and 05 studies that have been ongoing in the second line plus population. And again, just to -- we assure you there's no changes that we're anticipating to the timeline readouts for TROPION-Lung01. We've also got the TROPION-Lung08 study ongoing in combination with pembrolizumab. And then in breast cancer, TROPION-BREAST01 in the HR-positive and the TROPION-BREAST02 in the first-line triple-negative, HR-negative patient population. We do anticipate starting 5 new Phase III trials in the next 18 months for datopotamab deruxtecan based on the profile that we've already seen in the extensive program that we've got ongoing, both as monotherapy and in combination. So I want to just spend a couple of minutes now on capivasertib, our AKT inhibitor. It's a potent and selective pan-AKT inhibitor. And we presented at the ASCO updated overall survival data from the Phase II study that supported the initiation of our Phase III study, CAPItello-291. You can see on the left-hand side, the updated overall survival in the intent-to-treat patient population that goes across both pathway activated and pathway nonactivated patients. We showed an improvement in overall survival with a hazard ratio of 0.66 of -- from 22.4 months to 29.3 months. And again, it was an even better hazard ratio when we looked at the updated pathway activated patient population using next-gen sequencing which identified a further 20 patients that were identified in the pathway altered group and actually had a hazard ratio of 0.44 for overall survival with an absolute increment of around 19 months in that segment. So again, we look forward to the data reading out from the Phase III trial, which you could see the design of -- on the right-hand side. Again, I've talked about the advances that we've made in the antibody drug conjugate space. I'm delighted to say we have our own proprietary antibody drug conjugate technology, which is now in the clinic with our own linker and TOPCAT summarize warhead. The first target that we're addressing is B7-H4, and this drug is in Phase I. And I want to draw attention to 2 things. One, on the left-hand side, an illustration of technology that we're developing for our antibody drug conjugate portfolio, using computational pathology, artificial intelligence, use of analysis of immunohistochemistry stained pathology slides. So with this, the AI algorithm can identify the cancer cells within the slide. They can identify the periphery of the cancer cells and can quantitate pixel by pixel, level of staining in the cell membrane. What that then produces is a histogram distribution of the Pixel entity across that slide. This gives you a fully quantitative continuous score of the receptor expression. And using that, we've identified a cut point that we're using to select patients in the Phase I study from the start. And we've seen, again, on the right-hand side, very encouraging preclinical data in a range of PDX models, including in triple-negative breast cancer. This target is expressed highly in triple-negative breast cancer, ovarian cancer and also cholangiocarcinoma. And we've also shown that we've got a great combination activity for this with our selective PARP inhibitor, AZD5305. So we intend to use this kind of approach and this kind of technology across our portfolio. Another drug that we have some updated data on at ASCO is MEDI5752, which is a novel PD-1/CTLA-4 bispecific 2 clinically validated immuno-oncology targets. And we know that with CTLA-4, there's real benefit in terms of the long-term survival that you see with this target. The challenge has been the toxicity. The design of this molecule is designed to only bind CTLA-4 in the presence of PD-1 on activated T cells. So you get preferential binding within the tumor and less binding within the normal tissues. We've been exploring this across the dose response data, and that was presented at the AACR earlier this year. And what we've shown here is an updated data from efficacy expansions from that study in renal cell carcinoma across a range of doses. So these are the dose escalation data and again, very good durability of response that was seen in this setting. But at the 1,500-milligram dose expansion, we've seen, again, highly durable responses, median duration of response of over 16 months, median precession-free survival in the same range. We've also seen similar efficacy at the 750-milligram dose expense with a similar response rate. What is different about the 750-milligram versus the 1,500 milligram is improved toxicity with fewer discontinuations, which lower hepatic toxicity and at both doses, we've seen substantially lower colitis and diarrhea than you see with the combination of CTLA-4 and PD-1 antibodies, when given separately. I would also remind you that we've also updated the data for Calquence at this meeting, again, a best-in-class BTK inhibitor designed to be -- to bind BTK inhibitor, BTK receptor more specifically. And we've had updated data from both the relapsed/refractory setting to CLL with the ASCEND study. And on the right-hand side, the ELEVATE first-line treatment-naive setting for treatment-naive CLL. Again, very impressive separation of these curves, 62% progression-free survival benefit of 4 years in the ASCEND study and the combination of Calquence and obintuzumab with 84% progression-free survival benefit at 5 years. And again, this robust clinical safety profile and deep efficacy profile that we have seen is supporting the successful commercial activity that we're seeing with Calquence because they've seen been doing a fabulous job with this in the marketplace. So with that, I'm going to stop here. I'm happy to turn over to questions and answers and invite Pascal back. He's going to moderate the Q&A session. Thank you.

Thank you, Susan. All right. So we have a panel here. And then of course, we have also a few people in the room who could add -- Simon, do you want to get started?

Thank you, Pascal, and thank you for the -- all of you for the presentation. Going back to DESTINY-Breast04, I just wondered if you could get your thoughts on the basis of the ILD toxicity. It doesn't appear to be a major issue, and it seems to be getting better with time. But one of the suggestions in the presentation on Friday, is that it could be the result of activity by the drug on lung metastasis. I just wondered if you could give your thoughts on that and the potential that as you move into earlier lines of therapy and hence slightly less severe disease, this problem becomes smaller rather than greater in potentially sort of adjuvant settings. And also related to the safety issue on DESTINY-Breast03, there was a recap of the safety data on exposure-adjusted incident rates, taking account of the fact that the 2 arms have not seen equal drug exposure. That's definitely the case in DB04. So I just wondered if you've got that data to show the EAIRs for DB04.

Thanks, Simon. So Mika, maybe you can take those 2. And if Sunil, do you have anything you want to add, please jump in.

I'd be happy to take that. So I think with the emerging data, I think as you pointed out, we're actually very reassured that the ILD rates, especially the high-grade ILD rate is actually decreasing as we move into earlier disease. We actually have looked at -- we looked at a meta-analysis that was published last year. You might have to help me Sunil with the exact -- which congress it was presented at. But we've done a meta-analysis of the safety database of all ILD cases, and it does not appear as though lung metastasis results in an increased rate. So is -- we do continue to sort of look to try to understand the underlying mechanisms of action, but lung metastasis does not seem to be a risk factor. Your second question was around the exposure-adjusted rates of adverse events, which we presented for DB03 at this Congress, we don't yet have those data for DB04, but we will be looking at that, we'll present it in future congress.

If I can just add one more thing. I think the key thing that we learned was prompt identification of ILD and also giving clarity on how to give corticosteroids and how long to give that for. So that was the finding that we really worked on to make sure those guidelines were implemented, and we worked on those guidelines in 2019 and those guidelines were implemented in 2020 across the program and across all of the trials. And that, I think, is really what the clinical community has found to be really helpful to reduce the rates of higher-grade ILD. But I think the question that you're asking as to what is the mechanism, we continue to explore that further, but certainly not related to a presence of lung metastasis.

Thnak you. But because it's an important point, I think it would be useful if Aleix -- DR. Prat, if you could actually give your take on this from a clinician viewpoint and your own experience.

Sure. I think what Sunil mentioned is critical, right? The guidelines must be followed, and we do that. We've learned how to manage this toxicity because it can happen. The percentage is getting better. But following the guidelines being proactive, giving the right dose of cost steroids, I think it's completely manageable. And I'm -- now that I'm also participating as an investigator in studies in early disease, the experience of ILDs with my patients, so I can say is that it seems to be as safe as in the metastasis, I think. But again, my personal experience.

We've got one here and then another one here and here.

Just wanted to get some information about 8205. Is it a cleavable linker? You said it's a topoisomerase payload. And I guess in that regard, we've heard from some of the physicians that they felt the interstitial lung disease could be related to being a cleavable linker for Enhertu.

Susan? Sorry, yes.

Yes. So we do have a cleavable linker. It's capable in the tumor microenvironment, but stable in the peripheral circulation. So again, we'll be happy to share the initial PK data for that. I do think that the side effect profile that you see with the different ADCs depends not just on the warheads and the -- warhead design and the linker, but also on the antibody targets. And you will see a different distribution even when you've got the same warhead and linker depending on the different target. So I think it's an interaction between those elements that lead to the particular side effect profile. So for example, and Chris, you want to comment with datopotamab deruxtecan, which also has a cleavable linker and target. You don't see the same rate of ILD as we do with Enhertu.

Yes. We have seen some few cases, a little bit higher doses. So there is a dose relation probably also with ILD, but much, much lower compared to what you see with Enhertu. So it is, as Susan said, is a combination of various factors.

So we've got the one here. Just here. Yes. And then next one will be there.

Chris Uhde, SEB. I guess on the same subject, and I -- perhaps it might present and answer as well. Thinking about the fact that you have the HER2-low and then the HER2-null, which is actually not null, but 20,000 copies or whatever versus HER2 mutant where you have efficacy in lung cancer, whereas for some reason, you don't in HER2-positive. Can you share with us some thoughts about first of all, what's going on with the HER2 mutation and the interaction with the antibody? And I guess, does it give us a clue basically about how low you can go with HER2? Does it also potentially explain -- I mean, have you considered whether it explains why you get the ILD where, obviously, I guess, we have some level of HER2 in -- certainly in lung cancer patients and perhaps, I guess, in wild type as well?

Mika?

Yes. So I will attempt to address your question, but I may ask for a clarification if it does -- if I don't address it. So I think within the -- what we know with HER2 mutant, when you have a mutation in the HER2 receptor, it actually affects the internalization of that particular receptor. And as we've demonstrated in our lung studies, highly -- Enhertu is actually highly effective in treating HER2 non-small cell lung cancer with response rates of about 60%. The second part of your question was related to whether that is providing -- sorry, I think the HER2 expression in the lung, whether that was actually having an impact on the adverse event. Sorry, I don't know if I got the second one.

It's whether the ILD is related to the fact that there was HER2 expression in all normal epithelial, including lung epithelial.

Yes. So that is -- that's actually the long-standing hypothesis that we've actually been trying to look at -- others have looked at actually before us. It's not necessarily clear that, that is what is causing the issue. We actually know that the naked antibody does bind but we don't necessarily know that, that is the mechanism of action underlying the ILD. So it's still something that we're looking into to try to understand.

First -- 3 questions. The first one, apologies if this is ignorance, and if it's a silly question, but just with regards to the actual quantification by mass spec at the HER2 receptor. Outside of the academic setting, how many would they actually get the results, community oncologists, do they get that data and are they able to interpret to actually understand whether -- what the HER2 status is? Just to be able to understand the data that they get. And then perhaps one for Dr. Prat. Based on the DB04 data, do you feel comfortable using the drug also in, let's call it, a triple-negative patient, if you like. I guess I'm asking about that, I think there's 58 patients essentially that were, if you like, a -- I guess triple-negative is the wrong definition we've been told now, but let's call it that. you feel comfortable in that population using this drug? And maybe one for Astra. When you've had discussions with the FDA, with the [indiscernible] going on at the moment, is the intention the label will be based on the primary analysis set, so HR-positive, HER2-low? Or will it just be a HER2-low label do you envisage because that, after all, was what the trial included in the first place?

So maybe, Aleix, you could cover both -- the first 2 actually, just give your take also on practice of testing for HER2. And Mika, you could take the last one.

Yes. So regarding your first question, how we measure the levels of HER2. In clinical practice, we don't use mass spec at this point. We use classical immunohistochemistry. I think what Susan pointed out is that there is other ways to measure HER2. And I think potentially, this could be of high value. Going back to how low can we go. Today, this major in immunohistochemistry has limitations. And this category that was mentioned by Susan and others, that called the HER2-0 is still highly terrigenous. We know, given by the definition of the ASCO/CAP guidelines, this definition is heterogeneous. You can have up to 10% of cells with some positivity, but you have to call it HER2-0. So I still think that probably new technologies like mass spec and others could help tease out really what are the minimum levels needed for this drug. And I think there's huge potential. And there's already data suggesting that even in this group, there are responses within HER2 like, for example, the DAISY study from our French colleagues. So that's probably the first question. If not, let me know if I am not clear. The second one is regarding the 58 patients, right, with triple-negative HER2-low disease. Yes, to me, the data is clear. It's clinically meaningful. And despite the low number of patients, yes, I would feel comfortable using the drug in this setting.

Yes. And I think the third question was around our expectations from what the label will look like with FDA. And yes, as our hypothesis was the efficacy of Enhertu in patients with HER2-low disease. And we do -- our base case assumption is that those patients will be included in the label, the hormone receptor-negative.

Pascal, may I just add one answer? I think that to build on the question around -- well, the mass spec question, but just is moving into a piece. I think what you hear in Dr. Prat's answer is that IHC is what is utilized, and I think it's well seen by the field as this is well incorporated into practice in most of patient records, there's the ability to be able to know whether a patient is 1 plus or 2 plus. So I do think that this is an area that, obviously, is we're preparing ourselves for launch, we're understanding on an account-by-account basis, what exists in order for physicians to be able to go into the records and understand for existing patients. How they'll be able to determine that they're HER2-low. But those data in almost all accounts exist and IHC is well understood. So I think we're talking about refinements when we get into mass spec that are important, but I don't think that it takes away from the ability for this to have actually Monday morning impact, and you heard that from the discussant, right? The discussant was actually talking about a woman who she said that she felt compelled to be able to actually discuss this option with. And I think that that's just based on getting some embargoed slides. So I think that as they see the full data set, there's going to be impetus for that.

Couple of questions. Firstly, commercial one for Dave. How much do you anticipate if Enhertu at peak in the U.S. to go through 340B? And just some idea of [indiscernible]. I'm obviously thinking about the value that Roche gifts the 340B network on Herceptin, I know things have changed and evolved in that setting. Second, for Susan, I know that Astra has been a pioneer in the use of ctDNA given the very favorable comparative toxicity profile versus chemo in the metastatic setting, could you talk to what trials you've got ongoing or planned for registration potentially in the adjuvant setting using a ctDNA inclusion criteria? And then finally, just in the first-line metastatic setting, I think you're running a combination with Perjeta as well as a monotherapy arm. Just looking at the strength of the data that you're seeing today. What's the probability that, that single arm, the inclusion of Perjeta arm gets reviewed and continued or not?

So maybe it would be good to give Sunil the last question. And Dave, do you want to get started with the first 2?

Sure. I mean, I think -- so 340B was the first question. And in terms of the second question within that, Andrew?

Sorry. So I was just trying to get a sense of the percentage that you anticipate to go through 340B network in the U.S. And to just review the Apexus pricing of what kind of discount it is, obviously, with the memory of trastuzumab and the money that they leave on the table with that -- with their products.

Yes. I mean I think on 340B, we see all of our products going through 340B, and I don't necessarily have reason to believe, Andrew, on this, that we're going to see anything that's differentially distinct as it relates to Enhertu. I think that we don't go into specific forecast brand by brand on this. And it's something that we take a look at, at the portfolio level in terms of the impact that it has.

Remember, Andrew, also, the -- we have really sort of tried to restrict 340Bs to avoid double dipping or to avoid people getting into 340Bs who should not be in it. And so the large numbers you have in mind for Genentech, they were actually prior to these changes, at least from our viewpoint. Susan?

Okay. So the second question was about ctDNA, specifically in the adjuvant setting. So we've got a number of trials that are ongoing that are utilizing cTDNA in different contexts. Just as a little bit of context. I know it's not the adjuvant setting, but SERENA-6 with camizestrant is running in patients that are identified with ESR1 mutations on ctDNA. So that's one example. We've got studies that are utilizing personalized assays in the adjuvant setting. But we did announce this week a collaboration with GRAIL because I think there's great opportunity to use the methylation signature ctDNA tools to identify patients from an initial screening setting and also potentially in an MRD setting. So we're excited about that collaboration and look forward to -- think about bringing forward new designs utilizing that kind of tool.

I think just finally, to give a number on this, just to make sure that I'm kind of clear on at a brand level. It's about 20% of our portfolio goes through 340B.

And thank you, Andrew. And the question on DB09. So I think you're absolutely correct that our confidence in our success with DB09 is certainly very high given the results we saw with DB03, we're seeing a median PFS that is succeeding 2 years and expected PFS in the CLEOPATRA, of course, is an 18-month range. We don't see any reason to change the current 3-arm design and we remain very confident that DB09 is going to show and Enhertu doing quite well. And we see what the results are with the combination and the monotherapy, but we don't see a reason to change the design now.

No, sorry, Steve here, maybe. Since you have a microphone close.

Steve Scala from Cowen and Company. You touched upon this, but could you be clear on your expectations for uptake of Enhertu in the HER2-low setting. I assume it's going to be very quick, but maybe you can define very quick? Second question is, if Enhertu works in the absence of HER2 expression, could the payload be administered directly with lower cost of goods? And what would the relationship with Daiichi be in that situation?

On uptake, Steve, I think that once -- I think that we've already got breakthrough therapy designation. I think you saw yesterday a huge amount of enthusiasm for approval timing. I would hope. So I think that once we get approval, I would absolutely expect that we would see pretty rapidly that women with HER2-low disease that their physician believes are ready for chemotherapy that they're going to very seriously consider using Enhertu. I don't know. Was that consistent Dr. Prat, with -- I mean would you look at it that way or differently?

Yes. I mean, I come from on Barcelona, Spain, where at this point Enhertu is not available or reimbursed by the public sector, and we are aiming to get this drug as soon as possible for HER2-positive and especially now in HER2-low disease. So yes, this is a drug that we want in the clinic tomorrow.

Yes. And I think if I can just add something to the uptake. Part of the uptake is how the guidelines are going to be evolving as well. And again, the reception that's been seen by the clinical community that we have high confidence that the -- in addition to the approval that the guidelines are going to shift very quickly because there is a need for these patients with advanced metastatic breast cancer.

It will vary dramatically by country. I mean I met 2 breast cancer doctors, myself, who told me that they already have lined up patients to put them on Enhertu, but it's only in countries where you can get a reimbursement, say, for instance, in Spain and other countries where the product is not yet reimbursed, it's going to be a lot more challenging. And unfortunately, in particular in Europe, you have those countries where patients wait 2 years or sometimes even more. I think the next -- did we cover your -- all your questions, Steve?

There was a last question about the...

There was a second. Yes, there was a -- I can actually take the second question. So I think it's a question about sort of Enhertu delivery. And I think one of the things about Enhertu is that -- this is an ADC that I think is really delivering on the promise of personalized medicine, right? And I think it's also delivering on the promise of ADCs that we've been waiting around for 15, 20 years. The -- and I think, Susan, you alluded to this in your talk about sort of the linker -- every single component of this drug, both the molecule targeting HER2, the linker and the warheads are all critical components that make it such an effective drug. The activity that we've seen in the HER2-0 that's been reported, I think maybe you're referring to the DAISY trial. I think they're one of the questions, and we're in discussion with probably we said under on these -- on that particular data. Our thought is that HER2-0 group actually includes patients that have what we're calling HER2-ultra low. And so we need to take a closer look at those patients to really try and understand. I think this is getting to the question that we're asking DB06, how low can we actually go.

Just to add to that comment. If the expression of the receptor didn't matter, you wouldn't see the difference between the HER2-high and the HER2-low group in terms of median progression-free survival and outcomes and other things that you do. So fundamentally, the antibody is a fundamental piece of the components. So I don't think just the free deruxtecan is going to work nearly as well as the antibody drug conjugate.

Jeff? Yes. Okay.

I'm just wondering what you're learning so far on primary and particularly acquired resistance to Enhertu, which you're doing clinical trials because there was some discussion around it. And I just wondered your thoughts on that partly because things like PEG-like protein expression were discussed? And if there's any argument for mixing warheads or sequencing warheads off this platform? Or do you think that, that's not something you should be looking at?

Cristian, do you want to cover this one because it sort of covers more than Enhertu.

Sure. Work is ongoing. We have not a lot of patients that progress on Enhertu and not in all of them, we have a positive mutation to run the analysis. There are some sustaining collaborator working on this. For instance, the French group that run DAISY will release soon data on some of the potential mechanism resistance. It is probably more complicated than one -- in the defined one single mechanism resistance. Actually, your point on trying to sequencing or adding on top of Enhertu upon progression, additional drugs is probably one way forward. And definitely, when you will have a portfolio like ours with so many ADCs, this will be an important aspect not only upon progression, eventually also to maximize activity in those indications, for instance, where you have a co-expression of 2 targets.

That's one of the big challenges, right? And exciting but complicated and you cycle patients across different medicines. So you combine them. So those are some of the things we're going to have to resolve over time. James?

James Gordon from JPMorgan. A couple of questions, please. Two just about read-throughs. So we've seen a really strong DB04 results, but the extent to which we can definitely read through or not to DB06 and even lower patients. Are you a lot more confident in DB06 working in the ultralow now? Or there are still areas where we need to be a little bit cautious on that study? And the second read-through study was how much do we think it is about the protein you're targeting versus the linker and the warhead? Are you reading through a lot to DS-1062 as well from the DB04 results? Or are there reasons we still need to be a bit cautious there? And maybe if I squeeze in just one on Enhertu question for change, which would be the conference J&J, we're talking about Mariposa and they suggested the next year, they've got high confidence to their combos, so they combo their Tagrisso like products and MET targets in bispecific is going to be better. But I think what you're doing is saying you should use a MET afterwards after you've already felt Tagrisso. So just any thoughts why MET afterwards rather than should we be combining them, please?

So Mika, maybe you could take the first one and Cristian, perfect one for you, the second one.

Yes, absolutely. So I think from -- we have -- the confidence in DB06, I think, has been increased after you've seen the DB04 data, most specifically because if you look at the efficacy in the 2 plus and the 1 plus, it's really highly consistent and that tells us that we believe that we can actually go even lower on HER2 expression and still be able to have efficacy of Enhertu. And sorry, you had a second part of that question that I missed. I don't know.

04 read through to 1062.

Cristian.

The -- when you have a technology that is working, I think you can have more confidence that using the same technology also a different answer can work. And it's not only a matter of technology. It's a matter of clinical data. I think with Dato-DXd, we see consistently activity single-agent activity across indications. Let's not forget that Dato-DXd, we still did not have a former biomarker to select patients with TROP2, but we have a level of activity in non-small cell lung cancer, in triple-negative breast cancer that is quite compelling. Susan showed you the -- how bold is our program with this drug ongoing -- for ongoing pivotal studies and 5 even more to come in -- that will kick in, in the next few months. The second part of your question related to amivantamab and [indiscernible] from Mariposa threaten. I don't know if you have seen the data that have been presented at ASCO [indiscernible] 2, in a later line they are good data, 33% response rate. But there are not data that let us thinking that this combination can bring in even another line a dramatic improvement on what fundamentally you can get with a third-generation TKI like Tagrisso in frontline. In addition, Mariposa is a trial that is using an IV drug, given every 3 weeks with an important rate of infusion reactions. And if you need to show 1 million PFS longer than 18 months, like that we showed in Flora, you need to keep the patient on at least a couple of years. So this is -- will be not a trivial achievement. CNS, we do not believe that when you have an antibody plus a TKI versus the level of activity that Tagrisso show in CNS will be probably improved. And you will have safety that you need to take into consideration. So I don't know. I mean, J&J probably is very, very confident on the trial. There can be even positive, can be clinical practice changing. I have my doubts.

Susan, anything you want to add? I know anything that affects maybe Tagrisso is important to you. Anything you want to add?

Well, I just think -- just a question about the design of addressing a potential resistance mechanism upfront versus when it emerges. The problem is you can't predict what are the proportion of patients that are likely to have met overexpression or amplification as the resistance mechanism upfront. But we know that it's a subgroup of the total and so you're designing a trial with the idea that, that -- addressing that subgroup is going to drive an effect size that's going to be affected across the whole population. And I think that's the fundamental risk in the design as well as what Cristian has said. Whereas if you're addressing the met amplification when it's there with the combination, that's the patient population that's likely to benefit from that combination.

In the net-net, James, as a sort of strategy to deal with as it stands to Tagrisso it makes sense, but as a frontline, first-line agent, it's a bit of a stretch.

Seamus Fernandez from Guggenheim. So just a couple of quick questions actually on the TROP2. I wanted to get a little bit of a better sense of the tolerability differences. So you guys are treating lung cancer patients, quite broadly, you're seeing lower levels. Is that actually a reflection of the lung cancer physicians just being fundamentally better at identifying ILD sooner? Or is it actually something that you think is structurally different in the molecules itself? The second question is as we think about TROP2 expression, relative to HER2 expression, we're seeing quite a lot of a different threshold of access to at least breast cancer tumors potentially to other tumors as we look beyond that. How do you see the evolution of 1062 or data in that context? I'm sure you guys are looking at the molecular dynamics of TROP2 very closely. So very interested to see how broad that molecule may be able to go.

Jeff, you -- sorry. Cristian, do you want to take this one?

I start with the first one. Maybe Susan can take the second one on the diagnostics. So Seamus, the -- actually, the safety profile is quite consistent across indications. J101 is including several tumor types, and we have announced -- we staREIT to have now very large data set in non-small cell lung cancer, and also in triple-negative breast cancer. And actually, there is no evidence at 6 milligram that is the dose that we are using in Phase III of ILD. We have seen in non-small cell lung cancer, not only some cases of ILD at the beginning at 8 milligram, but now is definitely not an issue. So what Susan was stating before, it is not necessarily only the payload or the linker, but also the target that can make a difference. So this is to answer your first part of your question.

And then in terms of how to select the right patients to be treated with a TROP2 ADC. I think fundamentally, I believe that antibody drug conjugates are targeted agents, and it's possible to enrich for the activity that you see if you can utilize the target to select the patient population. TROP2, is a protein that goes through extensive post-translational modification. And some of the antibodies that are used to sustain for immunohistochemistry assays don't actually bind to the part of the protein that is internalized, which is the part of the protein that would be most relevant for understanding the patient population that's likely to be treated with an ADC. So I think there's work to do with the biomarker. I've indicated already that we have computational pathology capability within a company, we're looking at the potential for biomarker development across the program. One of the challenges, of course, is that TROP2 is actually highly expressed in many different cancers. So I think we have to evaluate the value of the cut point and the added enrichment that you can get. That's certainly something that we'll continue to explore.

Elizabeth Walton from Credit Suisse. I'm curious about your ambitions in going even earlier in the treatment paradigm beyond what you're studying in DB06 and whether or not we should think about the result of DB06 as a gating item to make that decision? And would you be able to combine with endocrine therapy? And then more broadly outside of breast cancer for Enhertu, can you talk about the possibility of exploring it in a HER2-low, other tumor types and when we could expect to hear a little bit more on those clinical studies?

Thanks. Sunil, do you want to take this one? I know you've given a lot of thoughts to that.

Yes. Maybe I can take the first one and Mika, you can take the second one. So I think Mika quite nicely sort of pointed out, James, to your question about why we are confident in DB06 and certainly, I think the activity profile that we see in a consistent activity in 1 plus 2 plus. But I think what's also really important is the response rate that we see in DB04. We see a response rate of 52% versus 16% with chemotherapy. First-line chemotherapy gets you a response rate of around 40%. So we're seeing responses in second, third line that far exceed what we would expect in first line. So that gives us a significant confidence of moving into the DB06, but also given that we are much better than traditional chemotherapy and what we would expect with taxane chemotherapy, for example, in first line setting also gives us the confidence to go into early-stage breast cancer as well. And as you can imagine, we have multiple discussions ongoing right now to do that. But furthermore, there are patients who receiving currently endocrine therapy plus CDK4/6 who progressed within the first year. We expect that number is about 20% to 25% of patients who do not drive a benefit beyond 1 year with endocrine therapy and CDK4/6, and they are inherently resistant or have early resistance. So we think that we can benefit such patients with the right trial design as well. So I think there's a lot more to do beyond DB06 to bring it early and for early stage and replace chemotherapy and for some patients, endocrine and CDK4/6 inhibitors.

Thanks, Sunil. Aleix, do you want to give your own sort of private experience and your view on that?

Yes. I mean, based on what Sunil said, I completely agree. I mean, I think this drug should move to earlier lines. There are a lot of possibilities. I do think the first line is there. We are treating all patients today with endocrine therapy and CDK4/6 inhibitors, and we're seeing a lot heterogeneity in the response. Of course, some patients do great, but 20%, 30% of patients progressed in the first years and most likely this tumors need chemotherapy. Yes, probably the trial design here, there could be different options, but I do think that chemotherapy could make a big change in this group of patients.

Thank you. Got another one here.

Pascal, sorry I think there was second part of the question. So I think the question was around sort of do we have interest in going to the HER2-low for other tumor types? I think one of the things about the DB04 data that is so exciting in addition to the unprecedented benefit in women with breast cancer is the fact that we actually have shown that you can target a protein that isn't necessarily a driver of the disease, right? And so that had -- that definitely has implications beyond breast cancer. So we already have some data in gastric cancer and other diseases, looking into the HER2-low space, and this is absolutely an area that we want to continue to look in. I think there's also -- within that question, it was around combinations in endocrine therapy, and we do have DB08, which is an ongoing study we presented -- we had a poster here at this congress. It's in HER2-low breast cancer and looking at different combinations, including endocrine therapy.

Yes. Just a quick question on ILD and the duration of treatment. My understanding from Daiichi Sankyo in Japan is that using the DESTINY-Breast01 regiment. In Japan, they are struggling to treat patients more than 6 to 8 months despite a much longer duration of response and PFS, probably due to their clinical practice with very high index of suspicion. So I just want to see what's your experience treating patients outside of Japan?

Sunil?

Yes. So we have data from a few settings, including the French setting, where we have reported as to their activity, and we are not seeing that issue. And also, of course, in the U.S. use as well, we are seeing that the real-world data mirrors what we have noticed in DB01, where the progression-free survival is in the 14- to 16-month range and we're seeing consistent benefit from the real-world setting in U.S. as well as in France. So I don't think we have seen that data transmitted across as Pascal had mentioned, there's certainly higher grades of ILD that reported in the Japanese patient population, but they are not mirrored in other patient population, even in the third, fourth line setting. We have confidence that as we move this earlier that those rates are going to continue to decrease, and patients do get a benefit beyond. We reported last year the majority of ILD cases up to 95%, even higher occur within the first year. And if you're not having ILD, the rates of ILD drop significantly that allows patients to maintain on benefit. And I think that's what we're seeing in the real-world setting, at least from the French and the U.S. real-world data.

Actually, also in clinical trial. The Japanese data you are citing are coming from the first, maybe the first study where the population was very repatriated, and the treatment management guidelines were not in place. I think today, we are much more aware and also in Japan, the rates are definitely lower than in that specific trial.

Are you seeing similar short duration of treatment in other East Asian countries, let's say, China, Korea, Taiwan? No?

No. And we have -- and just to answer that, and we have looked at that data across all of our -- when DB03 was reported because as you recall, DB03 has 60% of patients from Asia. So it was very heavily represented from Asia. And there was absolutely no difference in Asian outcomes versus rest of the world. So we don't see any limitations in that patient population. And also at San Antonio last year, we reported the ILD rates by Japan, rest of Asia, rest of the world, and we do not see any difference in activity or duration of treatment across those different regions.

You cannot really draw too much conclusion from the initial studies in Japan because the guidelines, as Sunil or Cristian said the guidelines were not in place. And also, those are -- those were very late-line patients who basically derive the benefit and there was nothing else beyond -- after that. So patients tend to -- and the physicians tend to keep their patients on it as long as possible. So given sort of exacerbation of those ILD cases in a way, in it's much better today. Any other questions?

There you go. So I guess there was a lot of talk about tools for assessing which is the right population in a trial setting. Obviously, there are -- well, depending on how -- just how low you can go, I guess and Enhertu may or may not be a personalized medicine. But assuming it is and assuming with the other products that you were discussing the B7-H4 and capivasertib, do you guys need to consider moving into the -- I mean, are you considering moving into the companion diagnostic or, let's say, and/or companion diagnostic software business? And then another separate question that I had. There was a lot of discussion about various IO 2.0 hypotheses and including LAG-3 as well. I heard very little at this meeting about the molecules in the coast, and I wondered whether you can give us an idea of when we could expect to hear more about those?

Susan, do you want to take the first one? We are not a diagnostic company, but certainly we are looking at a number of technologies and partnering with the companies that are dedicated to it. But Susan, maybe you can cover this and Cristian, the second one? Is that okay?

So our strategy has been to partner with the right diagnostic company for the particular companion diagnostic setting, that's happened. And that strategy has served us really well because the technology is rapidly evolving. So if you look at what we first did in the eGFR space. We had a tissue-based eGFR mutation assay, then a circulating tumor DNA-based assay and then refine that. I think that either both -- and I say that it's suitable for the companion diagnostic approval and then we also need more decentralized assays to make sure that the community connect access to the diagnostic testing, and that's a strategy that we share across both R&D part of the organization and Dave's commercial part of the organization. So they're partnering with the right diagnostic partner, for the particular technology has served us well, and that's what we continue to do.

Maybe, Dave, you could say a few words on what we're doing?

Yes, I think just to pick up on that, the complexity of delivering these therapies is really changing rather dramatically. And what it takes to deliver in the real world, as Susan exactly said, the decentralized technologies and the ability to be able to deliver these tests, whether it's in Spain or whether it's in the U.S. or whether it's within Asia is something that we're spending an awful lot of time actually working with multiple partners on in advance of actually getting the launch. I think that we really demonstrated our ability to be able to do this successfully, and we've got a track record with -- we started with BRCA. We moved to T790M. We've learned, I think, some harder lessons on HRD to be quite honest, where this has been an area that we've really, I think, kind of have gotten a lot of lessons on here's what it's going to take to be able to deliver this in a meaningful way. But I would say that the work that we're doing to understand multidisciplinary team engagement, and the work that we're doing to understand what it takes to establish patient identification in a way that can actually be scaled across the globe is one of the core competencies that we've developed, and I think that we're executing against within commercial. And as Susan said, it's with different partners, but we're doing it even earlier in the process. And right now, we're already talking Dato-DXd. How do we make sure that the biomarker work and if it's unlocked through computational pathology? What's that mean for digitization. What's that mean for AI capabilities that are going to need to exist across the globe.

Thank you for your second question. Nice to talk a little bit about immuno-oncology. Recently, I think we have been very excited, I hope yourself as well. We released new data with our Imfinzi and tremelimumab combinations or Imfinzi alone in indications where there was high very medical leaders speaking about Malaya and TOPAZ-1 study with Stride regimen and Imfinzi on top. I think we believe very much in the second way about immuno-oncology. We presented cost data previously. What you know we are -- we launched the Phase III study based on the cost data, it's called PACIFIC-9, it's a trial in which we are comparing standard of care or chemo radiotherapy followed by Imfinzi versus Imfinzi plus monalizumab is NKG2A inhibitor or oleclumab, that is CD73 inhibitor on top of durvalumab in Stage III in non-small cell lung cancer. You have seen the data in the slides that Susan presented with 5752, we are very excited about this asset. This asset can help to deliver a very important mechanism of action in immuno-oncology, like CTLA-4 inhibition but with a more acceptable safety profile. That was the main problem so far. I think this bispecific can help in delivering this on top of standard chemotherapy. And then we have also -- there have been a little bit of disappointment recently, but we have another bispecific PD-1 TIGIT that can play a role because first of all, as the convenience of every 2 mechanism of action on the same drug with the [indiscernible] or PD-1 on top of the digit and can be a drug that we can differentiate compare what is current redevelopment through our portfolio. because we can combine with our ADCs, we can combine with other assets that in our rich portfolio. So we are very much on top of the second way in immuno-oncology.

Thank you, Cristian. So we have time for one more question maybe.

Just a question for Dr. Prat. When a patient fails frontline endocrine therapy, how do you make the decision to try another endocrine therapy or straight to chemo? And the second question is, have you seen any ILD in your patients that you've treated, and have you used the new mitigation strategy?

Thank you for the two questions. Regarding the first question, which is what do we do in clinical practice when patients progress to first-line endocrine-based therapy. It's a difficult situation there. And I think we take into account many variables to decide whether to continue with endocrine-based therapy or to switch to chemotherapy. There's a lot of heterogeneity in this group of patients, right? One we talked about before, which is the time they were on the previous endocrine agent. It's not the same if the patient was 3 years on endocrine therapy versus those patients that were only 6 months on prior endocrine therapy. Definitely, those patients with less therapy -- endocrine therapy and that they progress faster, this is a sign to go for chemo as the second line, whereas those patients have been 3 years with a CDK4/6 inhibitor, we tend to try to give endocrine therapy. I think the data coming in second line after CDK overall is really bad in general. And this is something also to highlight. We know we can increase PFS, overall survival, not that much. So definitely, it's a decision that first, we try to, on one hand, provide the best therapy from a quality-of-life perspective. But at the same time, we are worried that tumor we might lose time and we need to go for chemo. So I think this puts into contradictory situation we're facing. I don't know if Sunil, you want to add something in your clinical practice, But...

I think you covered it quite nicely. I think it's also maybe the pace of the disease and the degree of disease burden that comes into play. Those patients who have visceral disease are very fast, quicker disease pace that you would consider chemotherapy just to get that response and I think where endocrine therapy may not give you that response.

Yes. I think here is potentially where biomarkers, liquid biopsy and other tools could help us tease out better where to go to one therapy or the other one. Regarding the second question, which is in my clinical practice, if I've seen ILDs and how I manage it, I have seen it within clinical trials and outside clinical trials. In my experience, has obviously been Grade 1, Grade 2. I have not experienced Grade 3 or above. And with the current guidelines, again, being proactive. This is very important. We've learned a lot since the beginning, these patients can either continue with the drug, which I have experienced outside clinical trials to initiate again the therapy or sometimes we need to stop therapy, but this is my clinical experience.

Very good. Thank you so much for joining us tonight, and I wish you all a very good evening. Thank you again.