AstraZeneca PLC (AZN) Earnings Call Transcript & Summary

A warm welcome, everybody, to this 2023 ASCO event for investors and analysts, and thank you greatly for your continued focus and interest in our company. Just for your information, the presentation is now online. You can access it for those in the room by the QR code on your tables. You'll see all of our AR materials also listed in the same section. Here's our forward-looking statement. Please finalize yourselves with if you haven't done so already. Here's the agenda for today's call. Our prepared remarks will be delivered in two sections with Q&A at the end of each section, breaking up some of the content. It will give you a chance to ask questions about particular subject matters in more depth. The first section will focus predominantly on lung cancer. And we're actually delighted to have Dr. Roy Herbst here for that section. So I know it will be, hopefully, very informative for all of us and useful to you in the room. For those dialed in by Zoom, please you raise the hand function for you to ask questions or if you want to type the questions in via the Q&A tab that will also work. We'll be trying to get to as many questions as we can during the allotted time. ASCO is an incredibly important meeting for our company, as you know. In fact, Pascal was just telling me it's this 22nd ASCO in a row. So it gives you an idea of just how committed this leadership team is to this event and making sure that we have key data showcased at the best Congress as we can. This slide details the speakers, but I also would like to introduce you to on the right-hand side of the slide, a number of R&D leaders that are helping us to redefine how cancer is managed across a range of different tumor types. One of the -- the depth and breadth of the talent we now have in the company is proving to be exceptionally useful as you think about how cancer might change and evolve in time and how we might be at the forefront of that and the way that we're designing our trials and bringing forward new medicines. So it's an exciting time for oncology. It's an exciting time for this company. And with that, Pascal, I'll hand over to you. Thank you.

Thank you, Andy. Good evening, everybody. As always, a very exciting ASCO. And as Andy said, it's my 22nd and it gets more exciting every year. And when you compare with what we had 20 years ago, it's just amazing. The number of options that are opening and becoming available for patients; and the science is just amazing. So let me start by maybe a general comment on our company overall because we have, of course, oncology. But as you know, we also have, still, a very large presence in what we call biopharma, which is essentially cardiovascular, respiratory, and we are now starting to look at how do we build immunology within this. And our rare disease division generates about 17% of our sales and actually growing quite nicely, as you saw in the first quarter. The important part of this graph is at the bottom there. And it is important because it shows you our ability to actually commercialize -- develop and commercialize our products globally. And it's also important because it enables us to build partnerships and tap into innovation where it is. And I'm sure you've noticed it's hard to miss that Chinese companies are innovating at an incredible speed. And being present in China gives us both the opportunity to actually partner with them and help them potentially develop and commercialize their products globally. But it also gives us a first-hand experience of what is to come in terms of commercial competition. So far, we've done quite well. As you know, TAGRISSO is competing with local companies, and we still have dominant market share with TAGRISSO in China. So we really are training ourselves in terms of getting the right competitive spirit, because this competition is going to heat up everywhere around the world. And as you can see on the right-hand side here, we are growing in every geography. The emerging markets are doing extremely well. China is now growing again. But importantly, outside of China, we are doing well everywhere; in Latin America, in the Middle East, South Asia. So really, the company is growing in every geography very strongly on the back of a strong oncology growth, but not only oncology. Now this is really a very special ASCO for us as a company because it's our fifth plenary presentation. And so we were trying to think about who else, as a company, has had 5 successive plenaries and I try to remember whether Roche had one, but we couldn't find of anybody. So we are really quite excited and a lot of abstracts. And the important piece is ASCO is not, today, the only venue where companies present their data, their results, their clinical results. In the old days, it was mostly the ASCO. But today, the ESMO is about the same size, and we have a very busy ESMO this year for us. And as you can see here, we are presenting important data at this ASCO. We continue to work on building our leadership in a number of tumors that we have selected as priorities. We're looking at new modalities, and I'll cover this in a second. And of course, and importantly, we're trying to leverage the pipeline, the oncology pipeline to develop innovative combinations. The future is going to be about combinations. It really makes it very exciting but also very complicated and very challenging. I think the future of oncology is going to be challenging for newcomers in particular because combinations are going to be really fundamental to strategy. And so you need to have the products, you need to have the talent. You need to have the team to design this, then the cost is going to go up, because typically standard of care now cost a lot of money. In the past, it was chemo. So it was okay. Now you more and more are comparing to standard of care that is protected by patent and expensive. So the future development in oncology will become complicated and expensive. And of course, probably more lengthy because we're moving to earlier stages. In terms of these modalities, we've built -- continue to build on our historical strength in DNA damage response and tumor drivers, as you can see here. But we've also introduced in the last few years a new focus on antibody drug conjugates in particular and not only Dato and Enhertu that, of course, are great products and everybody talks about, but we've started developing our own pipeline of ADCs and I see Puja here in the room. She's done an amazing job, and we have quite a number of products that are exciting. In particular, the EGFR/cMET ADC, that B7-H4 is also interesting. So we're really now building our pipeline of ADCs. We are also, on the immuno-oncology front, we're working on bispecifics, as you know, and making rapid progress with those. We've started working on immune engagers. And there's one compound listed here that is looking quite exciting. And finally, we started making inroads in cell therapy. It was a very different approach. In the near term, we're looking at CAR-T, but in the long run, we have new technology, we hope will help us make a difference. So as you can see here, we're really looking at a number of technologies. And the last slide I will show you is this one, we've had seven readouts in oncology in the last few months, and we were joking there the day 7 out of 7, so that's quite nice. But I wanted to attract your attention a couple of points here. The first one is if you look at the left-hand side of this graph, you have TAGRISSO, ADAURA, of course, on top, and then you have AEGEAN. And then, those two studies, plus, of course, the Merck study, what they show, I think, is the need, and they create is enormous, I think, call for action to screen and diagnose people early. Because we now are starting to do in lung cancer, what we've been doing in breast cancer for many years. Now we can't only transform lung cancer, like breast cancer treatment has been transformed. If people are diagnosed; and that's not the case today. So there is a big call to action, I think, for the community, for us as a company, but the community at large to diagnose those patients early. Maybe another message I wanted to leave you with is the fact that we have a number of study with Imfinzi here that I've read out positively. So I think it really confirms the good activity of Imfinzi and some of you, of course, know the -- there was a long debate about PD-L1. Imfinzi is it as active as PD-1. I think those studies show that it is definitely a very active compound. And the final comment is if you look at on Enhurtu, the data that we presented this morning in various tumor types, and then you look on the right-hand side, the DUO-E study. I think what happens here is, we're going to see an earlier treatment of endometrial cancer with this different regimen, as you can see here. That will create an enormous unmet need in second line. And that's where Enhurtu can actually come in and fill this unmet need, because patients are going to be treated earlier. Endometrial cancer is expected to grow quite a lot. In fact, it's expected to double over the next 7-8 years in the U.S. because of obesity. It's a consequence of obesity and it probably will be the same everywhere around the world. So a big opportunity here for Imfinzi, but also Enhurtu to make a big difference in the treatment of endometrial cancer across seven line -- several lines. So with this I'll hand over to Roy Herbst, who is going to present the ADAURA results, which, of course, were the most exciting results for us at this ASCO. I'll hand it over to you. Thank you so much.

Thank you, Pascal, and it's great to be here, and it's -- for me as an investigator, it's an honor to be able to present at the plenary twice with data that changed practice. Because, I can tell you, I started in this field as I fell at Dana Farber 1994. And the progress we've made in this disease and the fact that we have results like this, that really makes me feel worthwhile. The work that I've done and collaborating with the whole team here and many others around the world. So I think most of you know the ADAURA, but I'll just briefly go through it again. This study is a Phase III double-blind, placebo-controlled trial. And the idea was to take Osimertinib, third generation, more potent EGFR mutation, selective, less toxic, CNS-active drug and to use it in the acumen setting. And this trial, of course, was possible because the drug could be given for 3 years. The drug had more activity. And the whole idea always makes a lot of sense that patients with lung cancer die from metastatic disease, so we want to bring our best drugs earlier. So this trial, basically, patients have to have complete resection, Stage Ib, IIa or IIIa; they could have had chemotherapy. This was the choice. It was based on Stage and also a choice between patient and physician. There are about an equal number of patients who had chemotherapy in both arms. It was somewhat stage-dependent. And the patients were randomized with 3 stratifications by stage, the 3 stages, Ib, IIa or IIIa, about an equal number of patients in each stage. But the EGFR mutation type, we only included the two most canonical mutations in this trial: exon-19 deletion, exon-21 point mutation, L858r, and race: Asian versus non-Asian. Patients receive standard of care plus or minus Osimertinib. And the -- we used a placebo in this trial because there was no standard of care to use an EGFR inhibitor before this. As I shared at the plenary yesterday, there had been a number of trials that maybe showed a hint for DFS, not significant, nothing showing survival. So this trial truly broke a new ground. And it was TAGRISSO, 80 milligrams once daily versus placebo for 3 years. You can see we randomized 682 patients, a worldwide effort. Plan treatment direction was 3 years. The primary endpoint of this trial was disease-free survival. We expected this drug would be active, but it was powered for a hazard ratio for DFS of 0.7, a 30% benefit. I still remember getting the call, it's a little bit more than 3 years and a few months ago. The trial actually reported, it didn't unblind, but it reported 2 years earlier than expected because the results, as I'll show you in a second, were so much better than -- even exceeded already high expectations. And that was the first endpoint DFS that we had at ASCO 3 years ago. So here are those data. And I still remember when we got these data, it was I think about -- good Friday that year and actually ended up on the ASCO plenary because the program committee just was -- this is lung cancer, look at these data. On the left, you can see the primary analysis stage Ib to IIIa disease. And you can see that the hazard ratio was 0.2 meaning highly significance, meaning an 80% improvement in disease-free survival. We -- primary analysis that was stage IIa to IIIa disease. This is adding in the Ib's, a group of patients that actually do even better on their own. So really, just a very impressive DFS, which resulted in the drug approval in many places, but not all -- reimbursement in many places, but not all, and that's why these new data are so important. And then you can see on the right, the updated data, we did at 50% maturity, meaning 50% of patients having DFS events. We updated it. It's in JCO this last January. And there you can see that the hazard ratio remains quite good. It's 0.27. The curve is coming a little bit together after all patients have stopped, maybe a little bit, but still 0.27, 73% improval in disease-free survival. So these are the data we had before the current ASCO, and quite impressive. And I think on the next slide, we add it in the brain because that's what makes a difference to patients. I can tell you every patient I care for is always worried about the brain and recurrence in the brain. And if you look at the right, what we can see is this drug has activity in the brain. And you can see this is the DFS curve for brain metastasis as the first site of progressive disease and, look, the hazard ratio is 0.24%. So a 76% decrease in [indiscernible] of that and other metastatic sites as well. So here, you're taking the best surgery, and the best chemotherapy, if needed, and then you're adding this drug in and you're preventing metastatic disease. Now we have all thought that, that would have an effect on survival, but there were naysayers. You all know, you read the same Twitter I do, and the same posts. And I have surgical colleagues who would often tell me, where is the survival data? So, well, we did not want to disappoint. So here it is. And again, we all waited anxiously for this. This was event-driven. Of course, you never want to see this happen because it means people are -- 20% of the patients had to have events, but this happened earlier this year, and then there's a process that occurred for data review analysis. But I still remember sometime in March, the first day I was able to see this. Again, I expected there were some leak-to-leak-to-leak curves, of course, from Japan. But still, I expected this would hit survival. But I wasn't sure. And -- but to have it with a hazard ratio of 0.49, look at 60 months and the median follow-up is about 5 years in both arms, no medians reached, by the way, as far as survival yet. But look at the landmark, 85% versus 73% of patients alive. The hazard ratio across the whole course of the curve, 0.49, that's a 51% improval -- improvement in fighting death. So 51% fewer patients died across that curve. It's just really a phenomenal result. It's, again, high expectations, but we even exceeded those. And this is in the stage IIa and IIa patients. Then when you add the Ib's in, still hazard of 0.49, p-value 0.001. And you can see, again, the 51% decrease in death. So really very significant results. I can tell you, I see patients myself. I do research, I do administration, do a lot of things. But I can tell you, everyone, that I've seen in the halls, I've not seen any one at this meeting of my clinical colleagues who hasn't said, "Wow, this is really impressive." They -- even the most harsh critics and I invite you to talk to them, too. So, then, this is an exploratory analysis. So this is a secondary analysis, a secondary point, so this is an exploratory in the secondary. So take it at that, and some of the groups are small. And of course, in the groups of the earlier disease, there are fewer events. But for the most part, the point estimates are all to the left of one. And they range between 0.35 and 0.68, but pretty much -- and this is in our paper, it's in the supplement. If you want to get a copy of that, in the paper, the New England paper that came out yesterday. You can see that pretty much all the different groups, whether it be sex, age, smoking history, race, stage, EGFR mutation status. Active in chemotherapy, that's an important one. I showed those curves yesterday. Whether you had it or not, there's a benefit. So I think this really shows the broad impact of this drug in the acumen setting. There are toxicities. No drug is going to be less than a placebo. But it is interesting that if you look -- I highlighted it yesterday, I wanted to highlight one number for the plenary. I think I did the grade III events. And if you look at any grade III -- first of all, there were no deaths on the study, no grade V events, but it was 23 -- any grade III was 23 to 14. So there were a fewer grade III events in the patients who got the drug. Patients -- even though this is an EGFR selective agent, patients still might have some rash, some low-grade rash and GI side effects, skin changes on the nail beds, maybe the eyelashes. But these are things we've learned to manage through. I can tell you, for the most part - and I've had patients I've cared for myself - patients can make it through 3 years. But I think, now, having the survival data, that will be important for the patient who might be having a challenge through that. And then here's the conclusions. So in the primary analysis, we had already demonstrated a statistically significant and clinically meaningful DFS benefit versus placebo, along with improved CNS, DFS and a tolerable safety profile. That existed before today. But now the survival data, both in the primary population of stage IIa and IIIa patients, you can see with high confidence overall survival of 0.49 and p-value 0.0004. And the overall population, adding in the IBs, 0.49, P0.0001. As I said, generally consistent across subgroups, including, most importantly, by disease stage. I showed those curves yesterday. They were in the paper and prior action in chemotherapy. So, conclusion, and I believe this strongly, it's the first global Phase III study to demonstrate statistically significant and clinically meaningful overall survival benefit, with targeted treatment in this population. It reinforces TAGRISSO as a standard of care for patients in the Ib and IIa setting who have had resection. I think it's also a new paradigm in the entire field, that a targeted agent, that's been honed over 25 years, first used in patients in the refractory setting, then used in patients in the frontline setting. Now actually a trial that with a lot of foresight started before the drug is even fully approved in the frontline setting. Now we see that we're bringing targeted therapies earlier in disease. As Dr. Soriot said, I think we have to screen more patients. We have to profile more patients. One of the things that -- certainly through the societies I'm involved in, the [indiscernible] and others, I think we need to get the word out. We need to screen patients. We do a very poor job of that in the United States and around the world. We need to do profiling on patients because now at a tumor board, this is very relevant. And I think that with these data out there, with the push it's gotten from this meeting, I think we'll see more of that. I think that's my last slide, and I'm happy to take questions later. Thank you.

Thank you, Dr. Herbst. So I'm just going to now summarize some of the other data that we presented at the ASCO in lung cancer. And then Dave is going to set the scene of what that means, and we can take some Q&A on the lung cancer portfolio. So let's start with Dato-DXd. You can see the design of this drug on the left-hand side. Design is a best-in-class TROP2 targeting antibody drug conjugate with a high potency topo payload with a short systemic half-life, optimized drug antibody ratio of 4, in this case, the same tumor selective cleavable linker as we have with Enhurtu, and then the bystander effect of the warhead once cleared, because it crosses cell membranes. There's three areas of focus for the development program. We're examining it as a potential to replace chemotherapy or a component of chemotherapy as a backbone. And we've got further outcomes that we're looking at in terms of novel combination regimens, including with IO combination agents. And we're working actively on assessing the predictive value of ROP2 biomarkers because, of course, the data that we've seen so far with this molecule is to date in an all-comers patient population in contrast to the data we have with Enhertu, which is all in HER-2 selected patient populations. So we already have 7 ongoing Phase III trials across non-small cell lung cancer and breast cancer, and you can see those highlighted here, and I can go through those designs in a little bit more detail. So, at this meeting, we've presented some updated data from the TROPION-Lung02. This is a Phase Ib study, investigating DATO-DXd and pembrolizumab, with or without platinum-based chemotherapy in a mix of first-line and second-line metastatic non-small cell lung cancer patients. And I think the other thing I'd draw your attention to is there's two different doses that were examined in this study, 4 mg/kg and 6 mg/kg, and you can see that across the cohorts here. So the dose expansion was in patients with less-than-or-equal-to 1 line of platinum-based chemotherapy and treatment naive across those different cohorts. So this is now released under the embargo, so I can show you the spider plots that will be presented tomorrow. We've got, on the left-hand side, the doublet therapy, on the right-hand side, the triplet. These are not randomized cohorts and there's a little bit more maturity on the triplet than under doublet arm of here. But again, you can see really good waterfall plots with the vast majority of patients having some tumor shrinkage. And the other thing I would point out is that there's good durability of response on those spider plots, including across different levels of PD-L1 expression. So in the first line, the response rate was 50% in the doublet, 57% in the triplet. The median progression-free survival is immature at this point, which is why I would focus on the durability of response on the spider plots. And these data are supportive of the ongoing pivotal Phase III first-line trials, TROPION-Lung07 in the non-squamous population without actual genomic alterations. In the less than 50% PD-L1 TROPION-Lung08 in the -- greater than 50% PD-L1 and the AVANZAR study, which is also testing a predictive marker for TROP2. Here's some of the safety data. As you would expect, if you're adding platinum-based chemotherapy, you're going to have some difference in the safety profile. And we've called out some of the rates here of anemia, decrease in platelet counts, neutrophils and neutropenia at the bottom. Grade I-II in the darker shade and Grade III in the lighter shade. So we didn't see any new safety signals here. We didn't have any grade VI or V ILDs attributable to DATO-DXd. The rate of Grade I or II ILD was 20%. And that rate of Grade III or more ILD was similar to what you expect to see from pembro plus chemo from the KEYNOTE-189 study. Hematologic toxicity was manageable with lower rates observed, obviously, in the doublet arm. And again, the safety is supporting the 6 mg/kg dose in the first-line studies that I've referred to. So again, this is a summary of some of the trials that we have ongoing. You've got the Phase III TROPION-Lung01 study, and I'm sure that you're all confident that we are eagerly awaiting those data. We've got the Phase II study, the TROPION-Lung05 study. We were looking in the EGFR mutant space in the Phase I ORCHARD study in combination with TAGRISSO. And then as I've described, you've got the Lung08, the Lung07, and the AVANZAR studies. So we have different opportunities here, as I've said before, we can replace current chemotherapy with the TROPION-LUNG01, looking at novel combinations in the ORCHARD study, and we're wanting to move DATO-DXd early in treatment paradigms in combination and look at that level of TROP2 biomarker. So finally, I'll end with just a review of the ARTEMIDE-01 study. This is our PD-1 TIGIT bispecific Rilvegostomig. Rilve rhymes with silver, Rilvegostomig, is how you pronounce that, if you're interested. We've got some preclinical data showing some differentiation for this molecule versus co-administration of two agents together. The data that we're presenting so far is a data set in patients that have had prior checkpoint inhibitor lung cancer. You wouldn't expect to see really high response rates in such a patient population, based on that mechanism of action. But now that we have the safety and the PK well established in this setting with no surprises there. We're now opening -- we've got cohorts ongoing in checkpoint inhibitor-naive non-small cell lung cancer patients. And we're planning to initiate Phase III trials for this drug this year. So now I'm going to hand over to Dave, who's going to walk you through the landscape maker.

Thank you, Susan. All right. So if we just take a look at two slides that look at the program first. Overall for TAGRISSO, then I want to talk overall for the lung cancer program. I think that, after we had an opportunity to listen to Dr. Herbst, what you can really see here is that we've got a very, very comprehensive program with TAGRISSO to really establish as the backbone of therapy to treat as soon as possible in EGFR disease. Obviously, with ADAURA, this really solidifies within the Stage I - III population, within those that are resectable. I think what's important to note about ADAURA, and Dr. Herbst mentioned this, I mean I think that despite the fact that the DFS data were positive and have been available for quite some time, and we do have a number -- I mean, many, many countries across the globe, where approvals have taken place. There are still many care gaps that exist. Care gaps in the sense of only 80% testing rates, maybe only 75% adjuvant treatment rates. Referrals to multidisciplinary teams, also not being as high as they can be. And so those sets of care gaps add up and create a number of patients who really -- we see these data as creating an imperative and urgency to be able to go out and make sure that we're able to get treated and to use this, obviously, to grow the brand across the globe. Reimbursement, we've got a number of countries that while approval may have taken place, reimbursement hasn't taken place. A number of large ones, still ADAURA in South Korea, Brazil, Taiwan, Hong Kong. These are places where there's opportunity now to use these data to secure reimbursement for ADAURA and creates opportunity. FLAURA, of course, remains the standard of care in frontline. We look forward to sharing the FLAURA2 data, which we'll do later on this year. FLAURA2, as we've talked about in the past, we really do see that this is going to offer a chemo combination, exactly which of patients will want to be treated with that combination. Obviously, we're going to want to engage with the community and see how they see the data and look at those data. We continue to believe that monotherapy is likely to be the standard of care for the majority of patients, oral benefit risk profile as the way that it is. But we are really excited about FLAURA2 for what it represents and also for the competitive aspect, in light of anticipated Mariposa data. And then lastly, I guess what I'd point out is that in the second half of this year, we look forward to the LUARA study reading out, which is in the unresectable Stage III -- or excuse me, the unresectable Stage III population. And then lastly, NeoADAURA. NeoADAURA will read out in 2024, and we have an opportunity there. We showed this at the last quarter. This is the extent of all of the work that we're doing within lung cancer. I think what's really important about this slide is that we are really developing a scale, a breadth in lung cancer that I think is pretty unique and unparalleled, relative to the others that are within the oncology space. And what you can really see here, we just talked about TAGRISSO as a backbone of treatment within the EGFR setting. IO with Imfinzi really clearly been well established with PACIFIC in the unresectable setting, and we've got a series of various specific kind of II through IX studies that are really looking to build on that backbone and serve as an opportunity for us to be the one that beats PACIFIC, if anybody beats PACIFIC. Of course, we're advancing the ADCs to replace systemic chemotherapy. We've got approvals already within HER2, in the HER2 mutant population. But of course, we're all looking forward to the data readouts, which we anticipate the first of soon. Novel combinations then becomes the opportunity, together, not only with obviously durvalumab, but also novel combinations as we move into, with our bispecifics. So hopefully, you can see within this, we've got an incredibly extensive program in lung cancer. We hope to be able to reach -- the overwhelming majority of patients to be able to have eligibility for our medicines in the future, and we've got a significant investment in a great number of assets in order to be able to get there. So with that, we're going to take a pause at this minute to do Q&A. I think I invite you, Pascal, up to moderate the Q&A. So we will ask in this session, if we could, we'll do a lung-based Q&A, since we've got Dr. Herbst here to be able to go through that, and then we'll present some other elements from ASCO afterwards. Thanks, Pascal.

Thanks, David. [indiscernible]. So if we have mics, there's one coming.

Andrew Baumcity. A couple of questions for Dr. Herbst. So when your patients ask you, how long should I continue after I've been on the drug 3 years? What do you tell them and how do insurers react? The second question, I'd be interested in the views on the CHRISALIS2 data that was presented at this conference, particularly in MET-negative patients. And I'm just trying to reconcile the CHRISALIS data with the CHRISALIS2, it's confounded because obviously, one is first line, the other second line, but one is also Asian versus another population, which is non-Asian. So, to what extent can we think about potential differences in ethnicity as it may impact mutational profile of EGFR, or the frequency of MET-mutations as we think particularly with regard to Mariposa and the competitive dynamic with TAGRISSO-based therapies. Sorry for the long question.

No, I can answer the first, the second, I might need some help from my colleagues on the panel since I've been sort of tied up with other things. And I haven't seen that presentation. But I would say that the trial was for 3 years. And we chose 3 years of therapy because we knew that previous studies with older generation drugs that 2 years had failed. And could it have been longer, yes, but we also wanted a trial that could get completed, and we wanted to have the results. At 3 years, we stop. And sometimes patients are concerned about that. They're doing well. But we stop, and that's -- we have samples, by the way, on those patients now, and we'll be analyzing them at some point to look at CTDNA and so forth. But at this point, it's 3 years. There are some hints that perhaps in some patients, it might be better to give longer. We saw that in the DFS update, where after stopping, if you look at the CNS relapse rates, they do tend to slightly go up. So, I think at some point, we might want to give longer therapy, and there is a trial known as TARGET, as you saw, that is looking at 5 years of therapy. I also think some patients might need less, that maybe we've cleared everything in there. They're done with 1 or 2 years. So we're -- right now, the treatment is for 3 years. I think there will be studies in the future that will address longer treatment. Of course, if someone fails retreatment, these are all things that will be addressed as time goes on. But right now, I think the discussion with the patient is, this is the period. Many are very happy to stop. They've been on and they don't have to worry about it anymore. As far as the other question, I don't know, Susan, do you have...

Maybe, Dave, if you want to add something. And then for the second question, we could ask Leora or Christian, unless. . .

Yes. I would like to add on the first question. And I think that, Andrew, you asked an important element around how long will patients want to stay on therapy. But then your second question for insurers. What we are seeing is that the -- in the U.S., patients are able to stay from a payer perspective on therapy through to the 3 years. What I'll also note is that we've talked in the past about how Part D reform, while it has elements of cost that it comes with it, that the affordability elements and the smoothing elements as it relates to co-pay capping is something that we see as beneficial. And I think that in the ADAURA setting, this is particularly going to be an important spot because as you take a look at kind of the cumulative impact of 3 years of co-pay for a Medicare Part D patient, who has to be able to make the co-pays. I think that this is an element where persistence as it relates to getting duration of therapy out to that 3 years. And in fact, we talk to patients, and Dr. Herbst, I don't know if you want to -- I mean we talk to patients who they say, "Well, gosh, I'm not sure I even want to stop at 3 years" at least from a patient perspective, in terms of what their emotional sentiment is. So I do think that Part D reform is beneficial to the economic situation for patient affordability here.

I'll agree. There are some people that it's very -- with all these drugs, whether it be targeted therapy or IO, they very reluctant to -- some people are already concerned about that. So it would be nice to have the action.

All right. Anybody wants to help with the second question, otherwise, maybe what we could do is move to the next question and come back to [ Chris Aristo ]. I think you were second. Go ahead.

Viktor Sundberg from Nordea. So I've seen data here at ASCO, perioperative studies in non-small cell lung cancer, but all studies are very different from each other in terms of baseline characteristics. And I think a regulatory agency representative voiced their frustration over this here at ASCO. But given the lack of comparability among trials, do you see a risk that AstraZeneca's drug Imfinzi becomes a second or third option among the choices here of drugs? If physicians just look at the data at face value or the absolute numbers between trials.

Leora, do you want to take this one?

If you're comparing KEYNOTE-671 to AEGEAN I think the data that you're seeing are very similar in terms of our path CR compared to their path CR rate. And actually, in AEGEAN, we had a more real-world patient population that KEYNOTE-671 was limited to Cisplatin. I think what the agency was discussing, Harpreet Singh's comment was the benefit of that adjuvant component. And I think Mark Awad had a nice presentation showing that there was more of a flattening of the curve in a AEGEAN -- sorry, in KEYNOTE-671 with that adjuvant, and you're seeing the curve continue to go down in 861. IO-OS maturity is still immature, and we were at lower maturity when we presented at AACR. So we'll -- with time we'll learn about that benefit of the adjuvant component. But I certainly don't think that we come in second or third, I think that there's strong data. And you now have three perioperative studies all showing that benefit in IO-sensitive non-small cell lung cancer.

I think the point that Leora had about Cisplatin [indiscernible] is really important, because if you think about Europe, of course, Cisplatin is more common, but it's possible, likely, we'll see, but that either the label or the reimbursement of both actually is limited to cisplatin. In the U.S., it tends to be a little bit more flexible, but still not having [indiscernible] on the label actually is a limitation. It's very commonly used in the U.S., as you know. I'm sorry?

Maybe just one final question for Roy Herbst also. I was wondering if you plan to do a sub-group analysis that overall survival for patients previously on placebo, but I got Osimertinib a subsequent treatment. Or if that sample at the moment, just too small to see anything meaningful or if you plan to do that in the future perhaps?

We've been asked about that. What do you think, Leor, do you want to comment?

Yes. We're continuing to follow the patients. The good news is not many patients have progressed. So we're continuing to follow that, and we'll be able to update the data in years to come.

Right. That was a long discussion. And that's why I said in the last slide, plans are underway to do that. So again, with more maturity, there will be more ability to make those determinations.

I think there was a question at the back if -- once we get the microphone there, maybe, Susan, do you want to cover the CHRSALYS question?

So yes, Andrew, thanks for the question. So there is clearly data from that -- updated data showing there's activity in the met non-driven group as well as the MET-driven group. I think that's your point. I think the data set for CHRSALYS2 still remains relatively small in comparison with what we had for the OPAL2 data ahead of FLAURA2. So we're obviously looking forward to sharing the FLAURA2 data when we got the chance to do it at a congress, hopefully later this year. And I think what we can say from the OPAL data is that you've got a regimen there that was well-understood toxicity profile. You can stop the Pemetrexed during the treatment if that becomes a problem. Tolerability profile doesn't stop you taking the important medication of Osimertinib and that's a really important point to bear in mind. And so whilst there's activity there, I still think we have to just wait and see what the overall benefit/risk profile is there's still safety side effects that are seen with that regimen that I think represent a challenge. That's what we're hearing for some of the patients that are on it. What we've always said is that there's going to be a choice for those patients that want a combination. And I think the profile of what we've got with FLAURA2, based on the OPAL data, is one that offers a very reasonable choice of patients that want some intensification of therapy. So let's have this conversation again once we've actually seen the FLAURA2 data, I think it will [indiscernible].

What we hear actually -- and from those who have used this regime and the Mariposa regimen is that tolerability is really a big issue. So we'll have to see. And in fact, as Dave said, [indiscernible] from other therapy is likely to retain a pretty substantial position. Actually for a simple reason that it's easy to use and very well tolerated and the impact on people's lifestyle is minimized. Do you want...

Tony Ren from Macquarie Capital. So a quick question about the TROPION-Lung02 data that you guys released this morning. There's nearly 20% pneumonitis ILD. So I just want to ask the panel, in clinical practice, how difficult it is to manage ILD? I hear different conflicting views from thoracic oncologists. Some say it's pretty manageable. Other's say that lung cancer patients are prone to dyspnea to begin with. And also, did you guys see the difference in ILD between the doublet and the triplet? Did you deploy the mitigation strategy?

Leora, do you want to cover this one?

Sure. So the majority of ILD is actually grade I and II. In fact, the rate of Grade III -- we had no Grade IV or V in the Grade III ILD is actually the exact same as what you see in KEYNOTE-189. So for -- and Roy can definitely add in, for lung cancer patients with Grade I, ILD means you see a little fuzzy thing on imaging. And so the safety data that we're seeing is consistent with what we've seen at the 6 milligram, it's a safe regimen. And for the lung cancer patients, it's not a reason to discontinue therapy. So the safety data actually is quite encouraging. And don't forget, it's a mix of first- and second-line patients in there. And we know that the more lines of therapy patients have the more likely or at risk they are for things like ILD.

I was going to say the same thing, just waiting for Leora to confirm that it was mostly low grade. Yes, if it's low-grade, we're quite used to that from our experience with IO and taking care of patients who often will have abnormalities in their X-ray. So I think that would be clearly manageable in the clinic.

If we could also have a microphone here at the front Go ahead.

Mike Nedelcovych from TD Cowen. As it regards ADAURA, Dr. Herbst, you mentioned Naysayers. Dave, you mentioned care gaps. I was wondering if maybe you could help quantify that. Dr. Herbst, is there a number of prescribers that you think might adopt the ADAURA regimen now with the overall survival data, that didn't before? And if so, roughly how many do you think that could be? Dave, same question for you. And then a second question is on anti-TROP2 ADC data from Merck and Kellen, very early-stage data, small number of patients. But I'm just curious of your first impressions, how it compares to DATO-DXd. Dr. Herbst you saw that poster, I'd be curious, but also, of course, your AstraZeneca colleagues could comment.

Right. Well, certainly. . .

[indiscernible] the first question was related to how many percentage of clinicians. . .

Yes. So -- Sorry. You got me. So I think absolutely, it's clear that this will have an effect because 90% of patients are taken care of in the community. And patients have to be profiled. So I think with these data out there, there'll be more profiling. The whole funnel is going to increase, and there'll be more patients profile. You have to have that EGFR number when you first discuss is the patient going to surgery. And I think it's important to have that in advance, especially with other approaches available these days. So I think that's going to make a big difference. I think I've heard from surgeons, I've gone to a number of surgical meetings. One of the nice parts of having presented in the plenary 3 years ago as I became invited to surgical societies and that's important. And what I've heard is many are waiting, where is the survival data? I heard that more and more. And even in my own shop, I've had a couple of surgeons who have made comments, that -- when we've had strategy meetings at Yale. So I think, definitely, this is going to make a difference there. I also think patients -- the advocacy group for patients is quite strong. You saw the discussion yesterday, people -- the patients do worry. So I think now with more reassurance that there is an overall survival benefit there, I think more patients would be willing to commit to that. And then of course, as we've heard the effect on policymakers, the fact that we -- not only the approvals in different places, but the reimbursement and trying to keep it as minimal to the patient cost as possible. So I think that this is definitely going to have an effect. To quantitate it, I think we could probably more than double and even more certainly in countries like the U.S., where EGFR mutations are only 10%-15%. I think this is going to very much increase the denominator and give more patients the opportunity for this therapy.

Roy, can you maybe comment on the Stage Ib patients versus Stage II, Stage III and how that comes into play as well?

Right. Well, Stage Ib, we allowed Stage Ib patients and the staging has changed since we began the trial, but the patients with 3 centimeter or more tumors. And I think that, here in the U.S., I think there will be a large number of patients with early-stage disease, the stage I's that will go on the study. And I can tell you that it's already -- again, in the centers where tumor boards are happening, this is occurring in the community. I think more and more people now are joining tumor boards and discussing these cases. So I think we'll see many more people going on this regimen.

I mean I think just to add to that, and I made some of the comments and the remarks that I made before. It's -- in terms of quantifying it, it's a situation where you get, like I said, 70%-80% are getting EGFR tested. And then 70% to 80% get referred to a multidisciplinary team that includes a medical oncologist, and then 70% get adjuvant treatment. So it's a bit of a death by a thousand cuts. And so in some respects, I think, actually, Roy's suggestion of -- there's actually quite a few patients that are out there that we've got an opportunity to be able to now get on therapy because at all different parts of the funnel, as Roy described it, there are patients that we're missing out on. And so that's a situation where it's well established in the U.S. and Europe. And as I said before, I mean, if you just take a look at some of the countries that I mentioned where we don't, today, have reimbursement. Overall survival, I am confident is going to give a boost to our ability to be able to get payers to have a compelling message in a lot of those are in Asian markets where EGFR rates are pretty high and substantial. And then lastly, driving screening rates is something that I think that there's a greater sense of urgency around than I've ever seen before, and that's positive as well.

The reimbursement part is an important one, right? I mean there was a debate in the community with this PFS benefit drives an OS benefit. So you can imagine a lot of payers we are playing on this to deliver reimbursement. So in China, we don't have reimbursement, because we didn't want to sacrifice the price. Hopefully, this OS data will help us to negotiate the price better. So there are many opportunities and of course, the longer treatment duration. Susan, do you want to cover the second one?

Yeah. Sure. So you're asking about the Kilang TROP2 antibody-drug conjugate. Again, I think it's important to look at the components of the design of the ADC. So I think we probably have the TROp2 ADC with the most stable linker. And I think that affects the toxicity rates that you're seeing in terms of the hematologic toxicity. Which when you then might want to combine with some element of the chemotherapy regime, like we showed that we're doing with the triplet part of TROPION-LUNG02, that percent really raises more of a challenge in terms of being able to do that and the flexibility to do that. So I think the data on other ADCs, as you can see from the whole Congress, there were several new ADCs that are coming into play. Good news is that we've got a lot of data now that we've added on to the data program, and the other programs that we've got, and we will continue to move at pace to develop those in combination, along with the strategy. But we are -- absolutely have no element of complacency here, and we will move at pace in order to help develop that, because I think we have a really exciting construct with DATO, and it's a drug that I think is going to be an important part of many different regimens that we want to go forward here.

Thanks, Susan. We have a question online from Tim Anderson. Tim, do you want to go and ask your question? And by the way, Andy, remind me if we are running out of time. We're okay? Okay. So Tim, do you want to go? [Indiscernible] maybe answer the question. Any other question in the room? We've got one here.

Yes, sure. Chris Uhde from SEB. I thought I'd ask about the CD-73 program. And whether or not you're stratifying PACIFIC9 by CD-73 expression, if so, what's the cutoff you're using? And how did you get it? I noticed the pancreatic poster. It didn't say much there about that. And then, following up on that, do you have a clear idea of why the -- time has gone by, why the control arm in COAST underperformed what you would expect of PACIFIC?

So I can take those questions. Thanks very much. So first of all, just a reminder that the COAST randomized Phase II study was in the PACIFIC type setting. We know that CD-73 and NKG-2A, the other target that was tested in that, are both upregulated following radiation therapy. So what we -- what you would need to have is biopsies post radiation and pretreatment to really understand the biomarkers that might be relevant in that setting. So the answer is we don't have that insight, and we are -- it's an all-comers population in the PACIFIC 9 setting. So depending on the setting in which you might use that kind of target, it might depend on whether or not there is potential for biomarker enrichment with CD-73. So we do intend -- the NeoCOAST2 study is an important place where we're taking these drugs into the neoadjuvant setting, but we have a number of agents in the portfolio that are a priority that are in that NeoCOAST study. So we will look in a neoadjuvant setting as well in due course for Oleclumab. Sorry, for the second part of your question, sorry, can you just remind me?

Yes, sure, it was just the underperformance of the control arm.

Yes. Thank you for that. So we did do a propensity matched analysis of that. When you actually correct for the poor prognostic group that was in COAST versus PACIFIC, you get much closer to the expectation of the outcome from the PACIFIC arm. So it's just a poor prognosis group that was enrolled into COAST.

Thanks, Susan. So we'll take one last question and then move on to the next section...

Peter Welford with Jefferies. Coming back to the -- to some extent, the SKB-264 sort of the calib, but more important, actually, more bigger picture beyond that, given what we've seen with ADAURA. I guess, clearly, we see there that in the EGFR population, we get quite significantly more profound efficacy potentially with the TROP-2 ADC. I think all your studies are right in saying exclude patients with AGA's in the lung studies. Is there any plan to look at data in that population? And equally, perhaps for Dr. Herbst, currently on ADAURA patient or an ADAURA adjuvant patient who then goes on to relapse. What is your current standard of care at the moment now for a patient who has gone through that?

Well, fortunately, not too many have relapsed on the treatment arm. But if they do relapse on the treatment arm, I guess it would depend on when they relapse. Was it on therapy or soon after stopping therapy or has there been a gap, I would probably -- having the resources to do it, probably do a re-biopsy or at least get a liquid biopsy and see if there are any new mutations that have developed at that point, one could see if there's a new target. This is where we have to -- resistance is a problem in this area. This is where studies like the ORCHARD, which as we heard, contains an arm with the TROP2 might be reasonable, certainly for patients that relapse on the control arm, we would give them [indiscernible].

So to address your question about the Phase III studies that are ongoing. TROPION-Lung01 includes a cohort with actionable genomic alterations in it. And of course, we are examining, as we said, in the ORCHARD study, the combination of TAGRISSO with DATO-DXd in an EGFR mutant patient population that has relapsed following a third-generation inhibitor.

Thanks, Susan. So we're exactly on time. Over back to you. Thank you.

Thank you very much. So we had some other exciting data sets that we were able to present here at ASCO. So I'm going to just move on to, first of all, the ovarian cancer setting. I'm very pleased to see the results of the DUO-O Phase III. We've been a leader in developing novel regimens in ovarian cancer. So this is a setting with newly diagnosed Stage III or IV high-grade epithelial ovarian cancer, PARP inhibitor and immune-mediated therapy-naive. And the patients were able to have either primary debulking or planned interval debulking surgery. But importantly, patients enrolled into this study did not have germline or tumor BRCA mutations. So there are three arms in the study: ARM 1 is a standard of care chemotherapy and bevacizumab with bevacizumab maintenance up to 15 months. Two is the addition of Imfinzi, both in -- on top of the chemo/bev component and then Imfinzi and Bev in the maintenance with a total of 24 months of Imfinzi -- and then in Arm 3, you've got the addition of Lynparza on the back of the maintenance period of bevacizumab and Imfinzi. Primary endpoint was PFS, [indiscernible] per investigator in Arm 3 versus Arm 1 in the non-T-BRCA mutant but HRD positive and then the ITT population. With secondary endpoints of Arm 2 versus Arm 1 in the ITT and OS and safety. So here's that comparison, first in the HRD-positive group. So of course, non-tumor-BRCA HRD positive on the left-hand side, which was just under 40% of the patient population. And then the ITT on the right-hand side, and we saw an improvement in the median PFS from 23 months to 37 months. 37 months for this patient population is the longest median PFS that has been seen in many of the trials. And even though the PFS is counted from cycle 2 of the chemotherapy. So you've got 4 months longer than you did, for example, in the PAOLA study, that is still even accounting for that, the longest median PFS that has been seen. The hazard ratio was 0.49 in that HRD-positive group. And then in the ITT group, you can see, as well, an improvement with a hazard ratio of 0.63 in the ITT group. If you then layer in the ARM 2 with the Imfinzi, again, you can see that there's a trend to improvement in the HRD-positive group with a hazard ratio of 0.82 in this setting. And the hazard ratio was closer to 1, it has a ratio of 0.94 in the HRD negative, but there is the tantalizing aspect of a potential tail on that curve, which needs more maturity to follow up. So in terms of how we think about these data, we saw a statistically significant and clinically meaningful improvement in progression-free survival with the addition of both Imfinzi and Lynparza to chemotherapy and Bev. This builds on the data that we've seen from the MEDIOLA study, where we looked at this triplet in the MEDIOLA study and seen 87% response rate with good durability of response in that setting. And we saw that PFS benefit across the subgroups, including those patients with HRD-negative disease. Although, as I've shown you, there was a bigger effect in the HRD positive. There was a numerical improvement in PFS for the addition of Imfinzi to chemotherapy. And there were no surprises in terms of the safety, generally consistent with the known profile. So when we think about what this means, we have data now with Lynparza across the newly diagnosed ovarian cancer space. Obviously, the data that we've seen from this study are more robust in the HRD-positive group. And I think, in terms of the potential impact, we're probably going to need to follow for overall survival, which is immature at this point, particularly for the HRD-negative part of the patient population. So I just want to, of course, remind you that we have announced that we've seen positive high-level results from the DUO-E study in advanced endometrial cancer, and Pascal referred to this already. So again, this combination of Imfinzi-plus-Lynparza and Imfinzi alone, both significantly improved progression-free survival when added to chemotherapy. And you can see the design of the study here. And of course, we look forward to presenting these data in an upcoming congress. But I think if you put the DUO-O and the DUO-E data together, there's a biological mechanism that has been hypothesized about the interaction between PARP inhibition and immune checkpoint inhibition. And we're starting to see examples of where that interaction is potentially playing out into clinically meaningful data in the gynecologic cancer setting. So, now, I'm going to just highlight the DESTINY-Pantumor02 data that was presented earlier today. This is the design of the study, Enhurtu was used at the 5.4 mg/kg Q3 week regimen and across a number of different tumor types with around 40 patients per cohort. Primary endpoint was confirmed overall response by investigator with secondary endpoints of duration of response, disease control rate, PFS-OS safety. The trial criteria was patients that were non-solid tumors, not eligible for curative therapy. It was a second line plus, actually quite heavily pretreated patient population and patients could have HER2 expression by IHC of either 3 plus or 2 plus. Local testing was allowed for enrollment and then central testing was confirmed by Hercep Test. If local testing not feasible and the patients that were enrolled on local testing were subsequently centrally tested. And prior HER2 targeting therapy was also allowed. So you've seen these response rates, which we think are quite impressive, particularly in the gynecologic cancers. In more difficult-to-treat tumors like biliary tract cancer and pancreatic cancer, those response rates are lower. But I would note that in the pancreatic cohort, there were a number of patients with quite durable stable disease. And we also see response rates in bladder cancer. So the overall response rate was 37%. And the median duration of response was close to 12 months in the overall patient population in the IHC 3 plus population, durability of response was 22 months with a 61% overall survival -- sorry, 61% overall response rate. So what you can see here is that activity split out by IHC 3 plus and 2 plus across the different tumor types. And again, I would just emphasize that durability of response, which is really quite impressive. And again, the standard of care expectations differ across these different tumor types. For example, in endometrial cancer, that Pascal has previously highlighted, in this setting of such heavily pretreated patient population, you might expect around a 10% response rate in these patients. And certainly, the durability of response will be much shorter than this. So I think the other aspect of this is how frequent these HER2 2 plus and 3 plus patient populations are across this setting, and Dave is just going to highlight that and put it into context.

Thanks Susan. You sure? I'm not going to do the waterfalls, but I'll go on to this one right here. All right. So when we did the deal on in HER2, we talked about 3 pillars: HER2-positive disease in breast cancer, HER2 low in breast cancer and really opening up in HER2 and being able to move into other tumor types. And obviously, gastric and the work that we've done within the HER2 mutated lung cancer population has been really important for that. But these data that Susan just walked us through, albeit for a single-arm study, really represent a transformational way of thinking about how to now actually take in HER2 and move it into a population whose disease is and driven by the HER2-ness,but where the receptor actually serves as a great way of being able to target the warhead. And so if you take a look at what this means and you take a look across populations, between 8,000 and 12,000 third-line-plus patients have HER2-expressing tumors in the U.S. Now obviously, this splits out by those that are the highest expressors, IHC3 plus,and you've got the IHC2 plus,and you can see how it works through here. You can see the prevalence rates. And obviously, there's variability in prevalence that's influenced by the timing of biopsy and a number of other factors that come into it. And there's certainly educational needs and infrastructure that needs to be built in order to establish testing across these elements. But with that said, I think that there is an opportunity that comes from here where we've got an opportunity to be able to have a meaningful discussion with health authorities about the possibility of tumor-agnostic label. We also have an ability to be able to use this as a beachhead to be able to launch frontline studies in areas that may very well make sense. And I think that there's very much also going to be interest from guidelines committees in terms of how they think about taking these data and incorporating those into guidelines. So there's a lot that we still need to understand around this before we'll be able to probably properly quantify the commercial opportunity. I guess I'd say that in advance of some of those questions that will undoubtedly be coming up. But I think that we open up a whole area that previously was -- back to my comments before. This was in the upside case of the deal as we were doing it and thinking about bringing it in. What I'd like to do now is invite Anas Younes to present a couple of important slides on the MAIC data that we presented for Calquence. And then after that, I'll come back on stage, and we will do a Q&A, Anas?

Thanks, Dave. Hi, good afternoon. So I have a couple of slides to present. As all of you know, in relapsed CLL, there is two head-to-head randomized trials comparing first generation BTK inhibitors with second-generation BTK inhibitors. The first generation is namely ibrutinib. And the second generation, there is 2 of them, acalabrutinib, Calquence and zanubrutinib. And in our trial, which was head-to-head again acala versus ibrutinib, the data has been published. There's similar efficacy, improved safety in favor of acalabrutinib. IN the trial reported by BeiGene last December, again, it was head-to-head zanu versus ibrutinib. There was improved safety in favor of a zanu, but it was also improved efficacy. And people started to extrapolate. Does this mean the zanu better than acalabrutinib. So you really cannot compare apples with oranges. You really have to do a proper comparison using proper statistical analysis, which is called MAIC. It stands for matched adjusted indirect comparison. This is a very well-established methods even acknowledged by regulators actually for rare diseases also too. And the reason for doing that, you need to make sure that the baseline characteristics is similar before you do the comparison. And that's what we did here to do across a trial analysis. You can look at the baseline characteristics right now, and you can see that they're properly matched. And then you can do the comparison in terms of PFS between the two compounds. And that's what you see right now, exactly superimposable in terms of efficacy -- so -- and if you look at the right side part, which is safety, again, there's very comparable, but there's an edge in favor of acalabrutinib in a couple of spots, mainly for hypertension and a few others. So based on that, I think acalabrutinib remains the best-in-class BTK inhibitor. There's no evidence that anything else is better so far. And I hope it will remain for a long time. And that's all I have to say. Thanks.

All right. Great. Thank you, Anas. Okay. So we're going to open it up for questions now, and we invite you to go beyond lung cancer by all means and ask about anything that you're interested in. So all right as well. So Simon, please?

Simon Baker from Redburn. Two, if I may. Just going back to DESTINY-PantumorO2. You said you're now in discussions with regulators. So I shall ask the unfair question of what comes next, what do you think you'll need to do? And also picking up on a comment from the discussion this morning, you've talked about tumor-agnostic indications. There were suggestions about histology agnostic indication. So taking all that together, how big could this become for Enhurtu, incremental to what we've already got? And then your second question, on to something that you've not talked about but you have presented, but not in an oral. On camizestrant, we had the post a couple of days ago, looking at the data split by ESR1 mutation. And there's clearly a very strong efficacy indicated in the mutant group. Could you just update us on what the incidence of SL1 mutation is both in that second-line setting in first line? Is SERENA2 typical of the incidence we see there? Just trying to get an idea of what the scope is in the -- for the SERD in that setting.

Sure. So maybe, Susan, I'll start on the discussions with health authorities. It's not too much different, Simon, than what I had explained. I mean, obviously, I think that we've got a pretty compelling data set. There's a path that's existing for tumor-agnostic labels, particularly in heavily pretreated later-line populations. And these data are, I think, ones that are -- as we saw from the receptivity here from the community, ones that have the interest within health authorities. In terms of where we go next with it, Susan, I mean do you want to talk a little bit about how we're thinking about some of the development decisions that we'll be making from here?

Yes. So I think, again, these are really exciting data, and I think a paradigm change because there have been other HER2-targeted therapies that have been tried in these kinds of patient populations and haven't worked. And I think that comes back to the design of the ADC and the principles of those designs. So I think it creates a great opportunity. And obviously, the prevalence, as Dave's shown, differs by tumor type, but it is an opportunity, I think, to go into earlier lines of therapy. There is an opportunity to build on the data that you've got with monotherapy through some of the combinations, and I think we'll be looking at those. And you can imagine different combinations for the different settings. So I think this opens up a number of quite exciting opportunities to really change the standard of care and carve out another new area of cancers defined by, first of all, the IHC biomarker. But I also hope that we can continue to improve on the currently available biomarkers and potentially provide both thoughts about liquid biopsies as well as improved computational pathology tools that will better identify the right patients to treat. So as always, when there was a trial with great data, it creates a great set of opportunities of the next things that we want to do.

And then in terms of camizestrant and ESR1 mutations.

So it's -- in answer to your question, yes, it's -- the data that we saw in SERENA2 is roughly the prevalence that you would expect in that second-line patient population. Of course, the prevalence of ESR1 mutations increases with line of therapy. So it's less common in the first-line patient population, but it does increase with resistance. So it was roughly in line with that, to answer your question.

Okay. Thank you. We're going to go next online to Tim Anderson, Tim?

Thank you very much. Just starting off with DATO-DXd. So everyone is waiting for TROPION-Lung01, perceived as being late. It's making folks nervous. Just your latest confidence that this is going to be a positive trial and thread the needle appropriately? And then second question on MEDI5752, which you didn't talk about today, this is your PD-1 CTLA-4 bispecific. You talked before about moving in applied Phase II trials. That's a very broad program. I believe you're sitting on more Phase II data that you haven't released publicly. If that is correct, are we going to be seeing more of that data throughout the course of the year? There was nothing at ASCO, unfortunately.

So Susan, I think both questions for you. One, TLO 1 in terms of timing there. And then obviously, the second on. . .

So my goal in TRPION-Lung01 is to be entirely consistent and totally boring with what I said in the prior results, which is we said that we're expecting the results of this by the first half of this year. Obviously, it's an event-driven trial. We have to wait for the readout, et cetera. The confidence in it being positive is based on the Phase I data from the J101 study, with a high durability of response of 10.5 months. That was seen in the 28% response rate, which, when you take and compare with the standard of care, I think, provides the confidence that we've seen. And of course, the safety profile has been well understood. We're managing the key safety side effects of stomatitis and the other elements well, I think, across the program. So what can I say, watch the space.

And then for Rilvegostomig?

On Rilvegostomig -- so we are excited about the potential of the opportunity for the bispecific molecules to go into different segments of the diseases. Obviously, with CTLA-4 as a mechanism, we feel that we've gotten the design of Rilvegostomig something that mitigates the key -- what has been a limiting factor for CTLA-4 currently, which is the toxicity. We all know, based on the prior data, that the more you can give of CTLA-4, the more efficacy you get, but it's the tolerability profile, which is the challenge. And so the other thing that we also know with the CTLA-4 mechanism is that you need to see the -- you need to wait to get the benefit at the tail of the curve. So the key data that we have been generating across our program is what's driving our internal decision-making. And that's our priority at the moment is the internal decision-making and the build to enable us to move the Phase III program at pace, and we're all highly focused on doing that.

Right. Thank you, Susan. Can we get a microphone down here, as well, in front.

Chris Uhde from SEB again. So my question, I guess, following up on two of these here. So Pantumor. So in terms of the label effort, would you be thinking second line or third line plus and all levels of HER2 or not? And to what extent does the low efficacy in PDAC and BTC complicate an agnostic strategy from a regulatory perspective? And if there's any help you can give us, I guess, fair to assume that the prevalence of HER2 biomarker increases with lines of therapy? And then for the second question, coming back to the CTLA-4. We saw some very nice PCR data from the STRIDE regimen in HCC. And it made me wonder, which I'm sure many others are, whether or not patients who did not achieve PCR should be getting another pulse of [indiscernible].

All right. Super. So perhaps I'll address the first part of the Pantumor question and then Osama, ask you maybe for your point of view on the HCC question. Again, here, we'll take the data. You've seen the data, and we'll have conversations with health authorities in terms of where there's an opportunity to consider those data as being sufficient for a label. And I think that where you know that the conversation is most likely to resonate is where the highest unmet need exists, and that is in these later lines of therapy. And I think that there'll be a discussion around how to think about expression levels. And so at this point, it'd be really premature to be able to make a prediction about how that will net itself out. I think that the data that we have are ones that we'll take forward. And there'll undoubtedly be needs for us to continue to supplement those data, and we'll do some of our own work to bring it into earlier lines. But I think that what we have here is even more than a proof of principle. I mean I think that we've got something that's very meaningful to be able to drive forward. Introduce somebody who you may not have all met before, but who certainly, you see up here, Osama Rahma, who, as you see, is a clinical strategy head in our GI cancers. And we certainly are enthusiastic about HIMILAYA, TOPAZ, the things that are happening within GI. And obviously, we spoke about the Matterhorn high-level results. Osama, do you want to address the [indiscernible] and answer the question?

Absolutely. Thank you so much, David, and thank you for the question. So for the STRIDE regimen, we know with 1 dose of tremelimumab, we can really get the tail of the curve. And your question about whether rechallenging patients is actually an important question. As Susan highlighted, we have a few platform clinical trials that we are actually optimizing CTLA-4 for [indiscernible]. And we are looking, actually, to answer some of those questions to move to late development eventually. For the PTC part, we are very excited, very happy that we are the first in immunotherapy indications in BTC in frontline. And HER2 data really make our position very strong in BTC and this very, very heavily very poor prognosis patients, and we're seeing actually activity with the HER2 even in very poor prognosis, called bladder cancer, where 50% of those patients were all bladder. So you can only imagine what we can do when we move in earlier setting.

Right. Thank you. Osama.

Yeah, Victor Sundberg, again from Nordea. In the Pantumor study, I was just wondering, we saw very low ILD rates in this -- across all indications. Do you think that's mostly due to biology? Or is it also a part here where physicians are perhaps doing a better job managing ILD in general? And could we extrapolate that maybe to early lines of treatment in, let's say, breast cancer with HER2?

Susan?

Yes. So first of all, we do understand some of the risk factors for ILD. And absolutely, it differs by the disease that you're treating. So people have said before that if you've got other comorbidities that affect lung [indiscernible] symptoms, there's a higher risk. If you've got multiple lines of prior therapy, particularly prior lines of chemotherapy, there's a higher risk. So it differs by disease and it differs by prior treatment. This was a heavily pretreated patient population, but not one that necessarily associated with some of the lung comorbidities that you might see in other diseases. So that's point number one. And I do think that will likely make a difference. I think the second thing is that, absolutely, we've learned across the program as we've been -- got more experienced, and we spent a lot of time educating investigators physicians about management of ILD and the ability to identify patients when they're at the early lower grade evolution of this adverse event so that you can actually prevent the onset of the more severe growth. So I think it is both elements that are contributing to the lower rates here.

Thanks, Susan.

Emily Field from Barclays. Just two, kind of on the back of some of the presentations from the metastatic breast cancer session today. Firstly, the PATA1 study update. And at the end of that, they said that, that underlies the rationale for the SERENA6 study. But in that, they converted patients to a [indiscernible] upon detection of ESR1. So, how does that compare to the design of your study, were I know the included criteria you have to have the ESR1 mutation, but does sort of time matter? And then the study on sort of the dynamic nature of HER2 low status sort of stress that, give biopsies when practical. But I guess sort of how does that translate -- or do you see that translating into the real world to perhaps expand some of those triple-negative patients into the HER2-low population so that they could receive in Enhurtu?

Perfect. Ingrid? Got it. Sunil, please for both of those.

Yes. No, thank you, Emily. So I think the PATA1 was really the proof of concept that was behind the SERENA6 study design. And what we learned from PATA1 was when does the emergence of ESR occur? And what we know is that it doesn't tend to occur in the first 6 months after exposure of endocrine and CDK46. It tends to occur later on. So the way the SERENA6 design was created was really with the thought of when the ESR1. So I think we had that consideration. And with the data that was presented by Francis Coma Bedard and the team certainly reinforces our belief for the SERENA6 study and the rationale behind the SERENA6 and how we're thinking about camizestrant. The HER2-low evolution is certainly an important topic for discussion. So, what we do know is about 60% of patients who are hormone receptor positive have HER2 low. The question is that was -- that you're referencing is, how does HER2 evolve over time. And that was, I think, related to what was asked previously as well is that we do see that there is an evolution of HER2 over time. And generally, in triple-negative breast cancer as a result of treatment adoption that you may tend to see previously IHC0 patients who now express IHC1 or 2-plus. So the clinicians now are biopsying patients upon recurrence, certainly, and also patients who are now progressing to look for that IHC1/2 status. So what we have shared before is that if you think that 60% of all HER positive about half of all currently classified HER2-negative or IHC1 plus/2 plus, we expect that number to evolve now that there's more biopsies happening, more follow-up and more consideration of what IHC represents. So what that number will equate to in the future state is still unknown, but certainly clinicians are rebiopsying patients to look for the IHC1 signal.

Thank you, Sunil. Sure, we can go in the middle there. That's fine.

Harry Gilles from Berenberg. Just another long question, if that's all right. So the overall response rates in TROPION-LungO2 look pretty comparable with KEYNOTE-189 and 407. So I just wanted to sort of ask how you think about these relative to one another, of course, with the usual caveats of cross-trial comparisons. And then perhaps how reflective these KEYTRUDA first-line data are of what's seen in the real-world setting? Because we've seen a number of sort of trials recently whether comparator or active arm sort of seemingly underperforms compared to this.

Sure. Leor?

So I think it's important to look at the responses by PD-L1 expression, and the waterfall plot doesn't quite break that down. And so I think that's where you're also seeing the data in the spider plots with that durability of response. So I don't think you can just look at the response rates in totality. We are seeing deep responses. And if you look, for example, that triplet, that's a great response rate when the majority of those patients are PD-L1 less than 50%. So we're continuing to watch that data. That data is maturing. But it's not so much you can just look at the response. You've got to really take note of the PD-L1 expression there as well.

Thanks, Leor. I think this is an important point, and hopefully, that starts to come out in the way that we begin to describe the slide that I showed on the lung cancer map and how we're looking at just the sub-segmentation, the fact that we're really moving away from this belief, obviously, of a homogeneous population, of really understanding patient selection and the importance that that's going to play as we think about tailoring different combinations, but also the sequencing that needed to be gone through. Lung cancer is definitely an area that is becoming, I would say, more sophisticated and nuanced in terms of patient selection than it's been so far, please [indiscernible]

Yes. James from JPMorgan. Just one question on DUO-O. When I went to the presentation, the discussion wasn't actually that excited about the HRD positive, they're more excited about the negative, and they were suggesting for the positives, you could just use PAOLA1 and you might not need the extra [indiscernible] from giving someone a triple therapy because they already get quite good efficacy with the doublet. So, is there any risk that you don't get the HRD-positive approval? Or if you do, you don't get much use there, people just want to double it, and it's only really an opportunity in the HRD negatives?

Can I comment first on maybe one of the commercial aspects, and Mohit, I'll invite you to comment on this as well, if you'd like to, within the U.S. experience. But even within ovarian cancer with the success that we've had, obviously, with the SOLO1 allowing us to move into BRCA. And with PAOLA giving us a compelling promotional narrative to be able to have around that HRD-positive population, there is still a number of ovarian cancer patients who are not being treated with the PAOLA regimen who are HRD positive. And so, I think that if we take a look at the opportunity, that even with these data, which is the longest ever PFS that we've seen in an HRD positive population, I think that even if that was to result in a belief that there's an imperative to be utilizing Lynparza plus Bev within this population, even with or without a triplet. I think that, that's something that's a really important opportunity for us to be able to pursue. Now that's obviously distinct from the other part of the question, which is the HRD negative element, but I do think that there's a great opportunity here even just to reinforce the role of PARP plus Bev within that population.

Well, just let me add a comment because I agree with what Dave has said. So 37 months is the longest that we've seen in this is a non-BRCA mutant HRD positive. The equivalent group in PAOLA gives you a 28-month median peer first, just to put it into perspective. So even though you're now counting from cycle 2, so you've got to discount that element -- there was a worst prognostic population included in DUO-O because you were allowed patients that had stable disease, not just partial response after the initial chemotherapy as well, remember. So this is a worse patient population with the longest median PFS even in accounting for that difference. And that is meaningful. I mean I think that is a meaningful piece. Obviously, what we want to see is the evolution of those cars and see what we're getting is plateauing even more than we saw with PAOLA as well. And as we said, the OS is immature and other things. So I would -- I'd echo what Dave has said. I think this reinforces that we can raise the bar further in the first-line setting in the HRD positive. And of course, comparing with PAOLA, we didn't see the trend for PFS benefit in the HRD negative, but I think in that patient population, getting the mature OS data will be important to understand what that trend looks like.

Yes, thanks. I mean I do also think that just with the IO component and the tail aspect, the mature OS is even more important as a result of being able to see under the contribution of that to the regimen.

Can I just add something, Dave. I think just talking to clinicians, investigators, having that bevacizumab backbone in this regimen, there's still a lot of clinician investigators who are on the fence of the value of bevacizumab in ovarian cancer. And I think this sort of reinforces that, that combination is actually showing that robust efficacy. And I think that's really helpful for the HRD-positive patient population as well.

Agreed. Thanks, Sunil. Yes, please.

Seamus Fernandez from Guggenheim Securities. So two quick questions. So first on TROP2. Can you just help us understand the importance of testing as you advance into earlier lines of therapy? Obviously, you're starting to incorporate that into some of your clinical trials. So just trying to get a better understanding of that. At Merck's meeting just now, they were sort of trumpeting their expertise in developing biomarkers, et cetera, et cetera, and kind of characterize that as a vehicle for catching up to Astra and Daiichi. So just hoping you could maybe comment on that. And then separately, as we think about the opportunity to show a survival benefit in the second-line setting, for -- in TROPION-Lung01. Could you just perhaps comment on that and also where we are currently? I believe the final PFS event hasn't yet been achieved, but just trying to understand how far along we're likely to be on overall survival events in that context.

Okay. So Seamus, just to make sure on the second question, this is a question around when we see the data that's coming up, where do we anticipate will be with overall survival? Is that the question? Okay. Leor, I think those are two questions. One on importance of biomarkers as we move earlier and then the second, which we just reiterated.

So the first question, the importance of the biomarker. AVANSAR, as Dave mentioned, is one of the first-line studies that's ongoing, and the TROP2 biomarker is in there. And so, we know how to do biomarkers too. And so we -- and that study is up and running and enrolling. The second question, the OS data will not be mature. So when we get that PFS data, and Susan already spoke to that, the OS data will be immature. It's a good thing for lung cancer patients. They're living longer. And so we'll just have to wait. Again, it's event driven.

I mean I think the only piece that I'd add on this is, obviously, in the later lines, based on J101 certainly, I believe that the biomarker is something that may not be necessary in those later lines. And we'll see in the front line. And I think this is one of the reasons why it's valuable, as Leor points out and as we showed before, that we've got multiple shots on goal in that frontline setting, and the ability to be able to incorporate a biomarker approach, I think that is something that's important within there. And I appreciate that Leor added that we do know a thing or two about biomarkers. I think we've got a few products that we've been able to demonstrate that with. In front, Isabel, here.

On DUO-O, again, I guess a few questions around the data that came out. So in terms of HRR testing, was a negative test really a negative test is one question. And is it possible to tease out evidence of IO activity because there's no Lynparza arm with placebo instead of Imfinzi. So, I mean, perhaps outcomes by mismatch repair status that wasn't shown, yes. And yes, I guess that's. . .

Okay, Susan, please?

So in contrast to endometrial cancer, there's not a high incidence of mismatch repair in ovarian cancer, just to answer the last question first, so that it will be different in DUO-E versus DUO-O. The HRD test that is used is the standard HRD test. It's a genomic scar based test with a defined cutoff. So what was used in DUO-O entirely consistent with what was used in clinical practice and what was used in the PAOLA study. And coming back to the PAOLA study, that is the study that has a patient population that has the lapra and bev control that you're looking for. So I do think it will be helpful to have a cross comparison across those trials matched for the slightly different patient population that we talked about. So I think that will be a helpful piece that can potentially be provided at a later point.

Thank you, Susan.

I had a question on the earlier line ADC programs that you highlighted on your slides, [indiscernible] and B7-H4, any color on when we might be able to expect data? And I noticed that B7-H4, you recently expanded the trial. So is that a vote of confidence?

So first of all, we have now two drugs that have come out of Puchi Supra's discovery group, Pucha's sitting right there can give you [indiscernible], which have our own proprietary linker and topra-summarized warhead. The first was the B7-H4, as you rightly call, which is AZD8205, the second is AZD9592, which is the EGFR-MET bispecific ADC, which is in Phase I dose escalation. And there's going to be a third one with an IND open very shortly, which is the photoreceptor targeted with the topoisomerase warhead. So those are the products of our own internal discovery group. We've also licensed the Claudin18.2 with an MMAE warhead from KeyMed Biosciences. We will expect to share some updated data from these programs within the next year or so, I would expect the upcoming congresses. But we are moving all of those at pace. So I would expect the you would see data sometime in the next year.

Thank you, Susan. Any other questions? Okay. Andy, should we wrap up? Yes. No, I know that. Okay. Great. All right. Well, maybe just in summary and what I wanted to take an opportunity to share here as we come to the end. We really have set out over the last couple of years to build a truly transformative pipeline that looks to build scale across tumor types, really advanced scientific platforms that are meaningfully building upon the standard of care, advancing ahead of the standard of care, and then looking for ways to find how we can use combinations, precision medicine to really be able to make sure that we're able to deliver the best therapy to patients across each of these dimensions. And here at ASCO this year, you can see the number of studies, the number of presentations that we had across each of these different tumor types. And you can see the breadth and the depth of the portfolio, and it really does reinforce the potential to transform patient outcomes across these key tumors. I think that the depth that we have within lung cancer is something that we spent a lot of time talking about today. I want to once again thank Dr. Herbst for joining us today to speak through the ADAURA data. We spent a lot of time talking about, obviously, the work that we're doing with data and bringing that ADC in and the bispecifics, we look forward to really having opportunity to share more about as the year comes along. In breast cancer, I think, certainly, clearly, within HER2, we demonstrated within HER2, both within the HER2-positive as well as within the HER2-low populations, the opportunities that are there. And we see a number of important studies that will be reading out over the course of the horizon with bringing capivasertib. We hope very soon for approval and then later on camizestrant. Hematology, we've been building very actively on what has been an outstanding Calquence foundation within hematology. I think that the one that we'll see the most data the most recently from and that we've already begun to see data from is 0486 in terms of the furthest along. That's the CD3, CD19 T-cell engager and that we've really worked actively to advance that pipeline and there's quite a number of other studies that we've been moving rapidly into clinical studies. GI is the place, that in the last 12 months, we've made the most significant amount of progress in inroads into. It has been the greatest source of the driver of our commercial success with Imfinzi and bringing Imjudo, clear standard of care with TOPAZ-1 across the globe, HIMALAYA, making good inroads. And then obviously, we've got great opportunities that sit in front of us where we are looking forward to readouts of the EMERALD studies, which we think are very important within this setting and building within that. And then lastly, within the [ Gyn and GU ] place -- space, certainly very pleased to be able to see PROPEL approvals across the globe. We've got an opportunity through DUO-O, DUO-E and also through the Pantumor work to really see an opportunity to bring Enhurtu into these areas. So with that, we'd like to just say thank you very, very much for joining us today. We know that there are multiple of these sessions going on. We appreciate your interest in AstraZeneca during this. We think it's been a very successful ASCO for the company. I can't promise that we'll be at a plenary every single year, but we will certainly continue to try to make sure that we make that happen and look forward to seeing everybody at one of the congresses soon here later in the year. Thank you all very, very much in safe travel home. Thank you.