AstraZeneca PLC (AZN) Earnings Call Transcript & Summary

Good evening, and welcome. Very happy to be here today with you at the ESMO. It's been a very busy ESMO, as always, and a very exciting one. Let me start with the forward-looking statement and the agenda. We want to take you through some of the data that we have presented today, and Dr. Lisberg, Dr. Bardia who were the presenters today will actually go over the data again. And then Susan will also add a perspective on our R&D portfolio. Cristian and Dave will share our views on our leadership in immuno-oncology. And finally, Dave will explain how we continue to build Tagrisso over the next few months. And of course, importantly, we are here for your questions. So we have -- over the last few years, we have shown actually our ability to build this oncology business, as you know it today, starting from very, very little a few years ago. And as you can see on the left-hand side of this graph, developing a number of products that have built a leadership position in their category. Tagrisso, of course, Lynparza, Imfinzi and Enhertu, more recently, in Imjudo together with Imfinzi and Calquence, and now we are in the process of launching Orpathys. And the way we want to build leadership in oncology is first of all, by building leadership in a few key cancer types that we selected as our priorities. As you can see here, lung, of course, where we have a very large presence, and we continue to build on this. Breast cancer, hematology, GI and gynecology and GU cancers. We also do that by expanding our pipeline. So we've been building on our history in small molecules, in tumor drivers, in DNA damage response. And more recently, we've been adding new technologies, ADCs, with, of course, the partnership with our friends at Daiichi Sankyo, and I will show you in a minute a few more of our ADCs, but also immune engagers, cell therapies and, of course, immunology and our bispecifics. We're also building our leadership through shaping the future of cancer because really the future of cancer is about diagnosing patients early, intercepting the disease and trying to cure them. So we can eliminate cancer as a cause of death. In doing this, we also will alleviate cost in the health care system and, of course, save lives. We are using -- we are investing a lot in diagnostics and biomarkers and also we're using, of course, digital technologies to help us manage the patient pathway better than in the past. So as it comes to -- as it relates to our pipeline, I just want to highlight here that actually, we are progressing with some of the priorities that I've shown you a minute ago. We have 5 ADC now that are in the clinics, and you see some of them here on the left-hand side. You can see we have EGFR-cMET ADC. We have a folate alpha ADC. We have B7H4 ADC. We have a Claudin-18.2 ADC, which is a newcomer. And we also have our immuno-oncology bispecifics. We have 3 of those. We have a newcomer T cell engager, which I have to find somewhere on the graph, but I'm sure you will -- you can see, yes, CD39xCD19 T-cell engager. And we have also a first program in cell therapy. So not only we're focusing on this and developing these new technologies, but we are rapidly progressing products into the clinical stage. And finally, this translates into this tremendous progress and momentum in our pipeline in the last few months. We've had 9 positive readouts this year in 2023. We had -- at this ESMO, 2 presidential plenaries, as you could see -- and it's actually a total of 5 plenaries that we've been able to present this year and an amazing result. And beyond this, we also have data outside of the core tumors that we've been progressing in lung and breast. We have data in gastric cancer. We have data in endometrial cancer injury, and I'm sure we'll be talking about those exciting new data. And with this quick introduction, I'll hand over to Dr. Lisberg, who is going to take us through TL01. Thank you. [Audio Gap]

Thanks so much, Dr. Lisberg. So now I'll review the results from TROPION-Breast01. In terms of brief background, so if you look at breast cancer, the most common subtype of breast cancer is hormone receptor positive, HER2-negative breast cancer. HER2 negative essentially includes both HER2-0, as well as HER2-1+, 2+ or HER2-low. So this is about 70% of all breast cancers. For a patient with metastatic breast cancer, in general, we start with endocrine-based therapy. And after a patient has endocrine-resistant disease or you've exhausted endocrine-based options, we move to chemotherapy. And chemotherapy is used widely for the management of endocrine-resistant hormone-receptive positive metastatic breast cancer. But the problem is twofold. The first is the response rate, the survival, progression-free survival, overall survival with chemotherapy is quite low. And second, chemotherapy can be associated with significant toxicity, particularly myelosuppression and neuropathy. Now there are ADCs being developed, including TROP2 ADCs -- one that's already FDA approved, others in development, some of which have shown good efficacy, but the challenge with a number of those ADCs includes the toxicity profile because those ADCs can have toxicity profiles that are a bit similar to chemotherapy with neutropenia, thrombocytopenia and diarrhea. So clinically, in the field, there is a need for better agents, both to improve efficacy and to lower the toxicity profile. And in part, that's where Dato-DXd comes in because it's a TROP2-directed antibody drug conjugate. But based on the linker and the payload, it's a bit more tumor selective as compared to a number of other ADCs. So the off-target toxicity with Dato-DXd is much lower. TROPION-Breast01 study looked at Dato-DXd versus standard chemotherapy of physician's choice. This is what we would do in clinical practice for patients with metastatic hormone receptor positive breast cancer. Two points to note here. First is the schedule. Dato-DXd is given every 3 weeks, while standard chemotherapy, be it eribulin or vinorelbine or gemcitabine is given day 1, day 8, every 21 days. So it's a bit more inconvenient as compared to a Dato-DXd. Capecitabine is the other agent that's used in this setting, but it's given daily day 1 to day 14, every 21 days. And the second point is the endpoint. The endpoint of the study was progression-free survival as per independent review and overall survival. The study had 2 primary endpoints. But as per protocol, it was specified that if either of these endpoints are statistically significant, the study would be considered positive. In terms of baseline characteristics, they were well balanced between the 2 arms. This is what we would see in metastatic hormone receptor positive breast cancer, usually median age in the 50s, majority female. This was a global study. That's why 40% of patients were Asian, 49% white. 2/3 of the patients had received 1 prior line of chemotherapy. For hormone receptor positive breast cancer, CDK4/6 inhibitors are recommended as first line. Some patients get it in the second line as well, but majority of patients get CDK4/6 inhibitors. And that is what happened in the study as well. Majority of patients had prior CBK 4/6 inhibitor use. The chemo agents that are often used early on in breast cancer include taxanes, anthracycline, so we would expect that majority of patients who go on this study would have received prior anthracycline, taxanes, and that's what we saw. 90% had prior taxanes and anthracycline. So it represents the population that we would see in this setting in the endocrine-resistant setting. In terms of outcomes, so progression-free survival, this was the primary endpoint. The study met its primary endpoint. The results were statistically significant and clinically meaningful. If you look at a couple of metrics, we can start with hazard ratio. The hazard ratio was 0.63, so that's 37% lower risk of disease progression or death with that of DXds compared to standard chemotherapy in this setting. The second is median PFS that was about 7 months with Dato-DXd versus 4.9 months with standard chemotherapy. If you look at the curves, you see the small drop and then you see separation between the curves that appears to increase over time. So looking at landmark analysis becomes important in this setting. So if you look at 9 months landmark analyses, at 9 months, 37.5% of patients who are still on Dato-DXd, as compared to 18% with standard chemotherapy, so that's almost double. And so this potentially would have an impact on overall survival given the way the curves are looking. In terms of subgroups, all the subgroups derive benefit regardless of age, ECOG performance status, prior use of chemo, prior use of CDK4/6, or prior use of taxanes. This is, in general, what we would expect when the hazard ratio was 0.64, and there was no anticipation of cross resistance based on any of the subgroups. So that is what was seen. In terms of response rate, the response rate was higher with Dato-DXd, 36% including complete responses versus 22% with standard chemotherapy. And then finally, in terms of overall survival, the results are not mature, only a median follow-up of 9.7 months. At this early look, the trend is in favor of Dato-DXd, with hazard ratio of 0.84, but we need more follow-up and the study will continue to plan overall survival analysis. In terms of safety, I mentioned this in the presentation today as well. In general, we see the intervention having more toxicity than the control arm that was seen in a couple of other presentations today as well. But here it's the opposite where treatment-related AEs, Grade 3 or higher were lower with Dato-DXd, as compared to standard chemotherapy. And this is what was seen in TROPION-Lung01 as well, where the incidence of AEs were lower with Dato-DXd as compared to docetaxel. In terms of dose interruptions, reductions again, those were lower with Dato-DXd as compared to standard chemotherapy. There was 1 death from standard chemotherapy because of febrile neutropenia. In terms of specific AEs, the 2 most common AEs seen with Dato-DXd, and this was seen in the Phase I, Phase II study as well, so it's not a surprise, includes mucositis, as well as dry eye. Most of the AEs were Grade 1 and Grade 2. As a reminder, Grade 1 ocular toxicity would be something that's detected on routine exam. The patient does not have any symptoms. And TROPION-Breast01 required patients to have a baseline ophthalmological exam, and then every few cycles, they were required to have eye exams. And that's how the grade 1 ocular toxicity was captured. Patients were not having symptoms. It's because of the exam it was captured. The side effects that we're seeing more with chemotherapy included chemo-toxicity so that would be myelosuppression, anemia, neutropenia, common chemotherapy side effects. In terms of ILD, that's an important consideration with Dato-DXd. In breast cancer cord, the incidence of drug-related ILD was much lower, 3% all grades. There was 1 patient who had Grade 5 ILD which was adjudicated drug-related ILD. But as per investigator, that was Grade 3 and the patient died because of disease progression. So in summary, this study met its primary endpoint. It demonstrated both improvement in efficacy and overall safety, as compared to standard chemotherapy for patients with endocrine-resistant metastatic hormone receptor positive breast cancer. The PFS benefit was consistent across all subgroups, higher response rate with Dato-DXd. Trend in terms of overall survival with Dato-DXd. And also in terms of safety, the safety profile was favorable, manageable, no new signals, most AESIs were grade 1, grade 2. And if anything, patients had lower Grade 3 AEs with Dato-DXd as compared to standard therapy. So based on totality of the data, both efficacy as well as toxicity, we feel that Dato-DXd is potentially a new therapeutic option for patients with metastatic disease in the endocrine-resistant setting. So, I'll pass it on to Dr. Galbraith.

Thank you. So what I'd like to do now is put these exciting data into context. So with these 2 positive Phase III trials, it builds our confidence in the future potential for Dato-DXd. So when we started this program, we had 3 goals. The first -- and aligned with the vision about the potential for antibody drug conjugates to replace backbone chemotherapy in many settings -- the first goal is can we beat conventional chemotherapy. And we've done that now in both breast and lung cancers. The second goal is, can we have a combination profile with IO because we think there's great potential for improving the efficacy there. And with the data that we've already shown you in TROPION-Lung02 and TROPION-Lung04, we show a great combined ability and a step-up in response rate. The data we presented here at ESMO with BEGONIA, shows an unprecedented 79% response rate in triple-negative breast cancer and a median PFS of more than 30 months. So these data in totality help build the broad potential in earlier lines and other tumor types. We've already told you that we've got 3 ongoing Phase III trials in lung cancer with the TROPION-Lung07, TROPION-Lung08 and AVANZAR, in combination with IO with Avanza being in combination with Imfinzi. And we have now 4 Phase III trials in triple-negative breast cancer, TROPION-BreastO2, TROPION-BreastO3, and we're just going to post onto clinictrials.gov 2 additional trials, TROPION-Breast04 in the peri-operative setting in triple-negative breast cancer in combination with Imfinzi and TROPION-Breast05 in the first-line triple-negative breast cancer setting, again, in combination with Imfinzi. And then the PanTumor trials are also ongoing, which I think will provide some exciting data for the potential for Dato-DXd in other tumor types. We'll probably share some of those data in the coming months. Okay. So as Dr. Bardia and Dr. Lisberg have already shown, here's a summary of the data showing that we can beat conventional chemotherapy in both breast and lung cancer with clinically meaningful improvements in both trials. What they've also shown is that the adverse event profile is better for Dato-DXd than the standard of care chemotherapy, with a lower rate of grade 3/4 adverse events. In terms of a particular adverse event of interest, which is ILD, what we told you before -- and these data are consistent with that -- is that the rate of ILD overall in the Dato program is lower than what we saw with HER2. And as you remember, what we've done within HER2 as we've gone into earlier lines, we've seen lower rates. We also see lower rates of ILD in tumor types like breast cancer, which have lower predisposing baseline characteristics to that. So in the non-squamous patient population, the rate of Grade 5 ILD was 1.7%. And just to put that into context, it's similar to the 1.1% rate we saw in the updated DESTINY-Breast04 data that Dr. Modi presented at ESMO earlier this week. So we believe that Dato-DXd has a best-in-class profile based on its design, which supports IO combination. It's designed with a stable linker, allows for lower bone marrow toxicity, and that means we've got better combinability with this asset with platinum chemotherapy. And as you can see from the chart on the right-hand side, that leads to improvements in the overall response rate when you can add platinum chemotherapy to Dato-DXd and Imfinzi in the TROPION-Lung04. It's also a convenient dosing regimen, once every 3 weeks, 1 IV infusion per cycle. And I think that matters for patients in terms of how frequently they come to the chemo suite. And what the overall profile mean, with that lower toxicity burden, means that we can keep patients on treatment for longer and that is important when you think about that 13.8 month median PFS that we've seen in the BEGONIA data. And again, in terms of the tolerability profile, it's reassuring that we've now treated a large number of patients -- hundreds of patients in the first-line setting in lung cancer in combination with IO therapies with acceptable combinability. I'm just going to go to the next slide. So this slide puts into context the broad potential of Dato-DXd in earlier lines. You've got the addressable patient populations on these slides, again, in lung cancer in yellow at the top in first line and second line plus settings. And then in breast cancer with the studies that I've described -- with the different studies in breast cancer. And then again, I think there will be potential for Dato-DXd in a number of other tumor types that we're exploring in the PanTumor study. So in terms of what we've done with the TROPION-Lung01 data, we've taken some learnings from that, as I've described earlier. So I want to describe a little bit more some of the actions that we've taken. So first of all, given the data that's different in the non-squamous versus the squamous histology populations, we are enriching our program for the non-squamous population and capping the number of squamous patients that are in those trials. We also have confidence in a biomarker for Dato-DXd. And I think that confidence in the biomarker was based on the ability to predict both progression-free survival and overall response rates. And also, it helps to explain the difference in efficacy that you've seen in this histology because the prevalence is greater in the non-squamous population, which is in the majority of patients there versus a minority in the squamous population. And so we think this biomarker approach gives us the opportunity to have a differentiated profile for AGA in the first-line settings, and we can consider how to use the biomarker in other trials. So now I just want to go on to how we're going to continue to lead in the antibody drug conjugate space. And I just want to call out Puja Sapra, who's leading our ADC discovery portfolio and recently got a lifetime award at the World Antibody Drug Conjugate for her contributions to this field. So what we've learned over many years of trialing these agents is that you need to have the system design right. You need to have the right target and antibody design, and you need to have a linker which is stable in the peripheral circulation, cleavable in the tumor microenvironment, and you need to have a warhead, which is not only able to deliver a bystander effect in the tumor microenvironment, but also it's matched the relevant biology of the tumor type that you're targeting with the antibody. So we now have 5 wholly owned antibody drug conjugates in the clinic, 3 of which have come out of Puja's discovery organization, and we've got several more that will be coming into the clinic over the next year or 2. So our leading, internally discovered program is AZD8205. It targets B7H4 and you can see that some of the indications that we've got the potential for that. We will likely share data on this next year, but I can tell you that we've got encouraging PK, proof-of-concept, with this agent, and we intend to accelerate its development. We also have an EGFR/cMET bio-specific ADC, AZD9592, which has potential not just in EGFR-mutated, but also EGFR wild-type non-small cell lung cancer, as well as head and neck cancer. And we also have a folate receptor alpha-targeted ADC. And again, just as a reminder, in ovarian cancer, folate receptor is the highest expressed surface receptor that you can find, 10x higher than some of the other receptors with targets that you've seen before. And I think with the topoisomerase warhead, this is something that can be very differentiated and best-in-class in that context. We have licensed 2 programs, the first of which is AZD0901, which is a claudin-18.2 directed ADC. This one -- this target is really important in gastric cancer and in pancreatic cancer. And actually, the data from this program will be presented at the ASCO Virtual Plenary next month. We've also licensed a GPRC5D program, again, with an MMAE warhead, which is our first foray into multiple myeloma, and we'll have more programs entering into that space in the coming time period. And then we're also investing in the next wave of antibody drug conjugates. So we're seeing great activity for a number of different ADCs with topisomerase warheads, but that's not the only warhead that we think we need. We've got a diversified warhead portfolio that Puja's team is building, with improved therapeutic index for microtubule inhibitors, improved therapeutic index, the DNA intercalators. Imagine, for example, an anthracycline without the cardiac toxicity of the current anthracyclines. We have PROTACs that we can add on to the ADCs. We also have bispecific potential for our ADCs and also dual payloads that can help reduce the total complexity of combination therapy. And I'm pleased to say that we're making progress on radio conjugates as well. We actually have an open IND now for EGFR/cMET bispecific ADC with an actinium warhead that we're excited about as well. And that, we'll probably dose the first patient before the end of the year. So I think all of this adds to the potential that we see for all of these agents to replace some of the backbone chemotherapy that we have. And so with that as background, I now want to hand over to Cristian Massacesi, who's going to walk through some of our immuno-oncology data that we also presented.

Thank you, Susan. So I want a few minutes to update you on capital presentation. I probably have seen just to focalize on some of the key messages. DUO-E, in our view, is a very relevant Phase III readout in the context of the Imfinzi and Lynparza portfolio because the study in a population that is in a cancer that is very common and growing as incident in women. And it's a study that is testing a new hypothesis, the role of PARP inhibition in endometrial cancer. It is a study that we conducted in advanced stage, newly diagnosed advanced-stage patients or with recurrent endometrial cancer. 3 arms. The standard of care was a platinum chemotherapy standard followed by placebo, of course, Imfinzi, placebo Lynparza. There was a second arm testing platinum chemotherapy plus Imfinzi, followed by Imfinzi, and the third arm with platinum chemotherapy Imfinzi, followed by Imfinzi and Lynparza as maintenance. Primary point of the trial is progression-free survival by investigator, testing arm 2 versus arm 1 and arm 3 versus arm 1. OS was a key secondary endpoint in this trial. On note, MMR status that is important. This indication was a certification factor of course. This is the primary endpoint readout. The study is positive, met the primary end point is statistically, a clinically meaningful benefit. You see both arms showing improvement in PFS with another ratio of 0.71 for chemotherapy plus Imfinzi and 0.55 for chemotherapy plus Infiniti and Lynparza versus chemotherapy. What is important is these exploratory analysis, a comparison that we did to demonstrate the contribution of component of adding Lynparza parts on top of -- on Imfinzi. And you see at the bottom of this table, 0.78 is the other ratio showing that actually Lynparza on top of Imfinzi degrading versus the immunotherapeutics alone. Very meaningful median PFS with the triplet with 15 months represent one of the longest, if not the longest, the median PFS in these indications so far. These are the analysis in the key subgroups, the forest plots for both the doublet and the triplet. And you can see -- I don't go through all of them, but what is important that in all observed key subgroups, the other ratio estimates point is in the favor of the investigational arm. OS. OS was a key secondary endpoint. We tested OS. Of course, OS is immature at this point, but we already see something quite important because there is a clear trend. And the recent trend for both arms, chemotherapy, Imfinzi shown other ratio of 0.77. And again, a better other ratio in this ITT analysis was shown with chemotherapy, Imfinzi, and lynparza, another ratio of 0.59. And again, with an exploratory predefined analysis comparing the 2 investigational arm, we see that is an there is another ratio, 0.77 showing that Lynparza adding on top of Imfinzi. Now MMR status is very important. It's very important because there is a clear predictive factor, potentially prognostic. And because we wanted to understand to interpret these results. What we have observed, there are 2 start -- deficient MMR or proficient MMR. We already knew that in deficient MMR that is a very IO-sensitive subset, represent about 20% of the patients. This outcome with Imfinzi would have been probably relevant. And actually, we have seen another ratio with Imfinzi plus chemotherapy of 0.42, compare standard chemotherapy. And in this subgroup, the addition of the PARP inhibitor is adding definitely less. We already knew that this was a setting where probably IO alone could have performed very well, aligned with other readouts that we had in the recent times. What is really relevant in our view is the outcome in pMMR subgroup. This is 80% of the population. And here, it's clear the PARP inhibitor is adding on top of the checkpoint. You have another ratio of 0.77 with Imfinzi plus chemo and 0.57, again, with 15 months median PFS with adding Lynparza. And again, when you compare the 2 investigational arm, you really see the benefit on Lynparza. This is a subgroup where the medically desire. We also investigated, a predefined analysis but not stratified because we did not stratify for PD-L1 status in this trial, was not well established. It is still not a very well-established biomarker in endometrial cancer. But we, of course, wanted to assess also the outcome of both arms based on PD-L1 status. We use our Imfinzi diagnostic test, SP263, that told us that 67% of the patients were PD-L1 positive. And in this population, we have seen even a more clear benefit of Imfinzi on top of chemotherapy, and also on the triplet, again, showing a benefit. So this is telling us that the PD-L1 in this trial is emerging as a potential biomarker in an ITT population. Now I shift gear I want to spend one slide on MATTERHORN. Again, bringing checkpoint inhibition in the context of peri-operative and postoperative treatment of gastric cancer, resectable gastric cancer patients. This is a setting in which we know that the checkpoint inhibition is working, but it's not been formally endorsed in this specific early setting. And the study was fundamentally testing a peri-operative chemotherapy followed by postoperative chemotherapy, plus or minus Imfinzi and then Imfinzi as maintenance. Three main message here. The first one, Imfinzi improved the pathological complete response of 12%. It's relevant because it's 12% from 7% to 19%, okay? It's not from 50% to 62%. It is a setting where adding having a pCR is not so easy. This was a key secondary point. We spent a little bit of alpha -- just a fraction of alpha -- because, of course, most of the alpha in this trial is for EFS, that is the primary endpoint. The second thing is not only pCR was improved, but also the downstaging was improved by Imfinzi. The downstaging is important because when you are able to bring the patients to T0, so fundamentally, the tumor goes down and 0, you don't have any more involvement or [ lymphonace ], your resect-ability is better. So this can potentially translate in a better survival time. The third point here is the backbone that we used. In different regions of the world, sometimes there are different -- different chemotherapy backbone are used as peri-operative. We decide to use FLOT, that is the best, is the one that showed so far, the better activity and actually is the only regimen that they showed kind of a surrogate -- pCR as potential surrogate for DFS and OS. This is a study, an observational trial. We know the limitation of an observational trial, but it's telling us that -- can give us confidence fundamentally that pCR potentially can translate in EFS. And based on this data, of course, we are now waiting the primary end point readout EFS. And hopefully, we will be able to meet this. With this, Dave, I ask you to go to our IO strategy.

Thank you, Cristian. All right. So I think what is particularly exciting as we think about what we stand and what we have right now with IO and with Imfinzi, we know for the half, and we talked about this at the quarter, that Imfinzi nearing $2 billion in sales with growth of 57% over the previous quarter, sequential growth at '19, and 70% of that growth is coming from HIMALAYA. It's coming from TOPAZ, it's coming from POSEIDON. And what we also see and what's being presented here is that there's an awful lot of LCM and follow-on innovation that's coming behind this with readouts coming shortly that's in the middle panel here across, as you saw, as Cristian highlighted, GU, GI, and within lung cancer. And then, of course, Susan touched on -- and we can talk about more if there's interest in the Q&A -- 3 bispecifics, 2 of which are already in Phase III, another which is progressing rapidly through into Phase I. What I do want to switch now to is out of IO and into Tagrisso and talk about Tagrisso really as the backbone for TKI treatment with an EGFR treatment. And I think that this is pretty consistent with what we've been commenting on actually leading into this really, really important set of presentations that we saw today. What I'd like to highlight within this and just start with as we ground ourselves, and obviously, we all know that FLAURA set the new standard of care for monotherapy with Tagrisso based on overall survival results and PFS results of 18.9 months. But also really importantly, and I think that we heard this from the discussant who was speaking in the last session, oral and a well-tolerated profile are really, really core elements to also go along with that efficacy. With FLAURA2, we had the opportunity to show based upon the BICR analysis, 29 months, 29.4 months median PFS, the longest median that we've seen within this setting, inclusive of the data that was shown today. PFS2 within this, that was a hazard ratio of 0.7, again, inclusive, the longest that we've seen today. And I think also we had an opportunity to be able to share more data and more in-depth on, what about patients with brain mets. And you can see here on the left-hand side here, PFS per investigator, we know on here 24 months, 24.9 on patients who had CNS mets at baseline with a hazard ratio of 0.47. Once again, the longest that we've seen kind of within this setting. And I think that within that, for us, as you really think about kind of the commentary that we've had and the conversations that we've had an opportunity to have with clinicians and with investigators, Tagrisso monotherapy really does remain the first-line standard of care for the majority of patients who really get benefit from the efficacy of monotherapy, together with the fact that it's an oral therapy that's well tolerated. But we do also expect that FLAURA2 to be a standard of care for the subset of patients who can really benefit from a combination approach. And you might recall at WCLC, there was a lot of enthusiasm in L858R also in CNS mets, and I think that as you start to take a look at forest plots from studies that were presented today, I think, again, here, you see that it's quite favorable in terms of how FLAURA2 looks within this. And all of this, I think, is really also underscored by the fact that FDA has granted a priority review for FLAURA2, which is certainly a good indicator of our path forward with them. So with that, I'd like to turn to the Q&A portion of this. I do have, of course, all the panelists who we've heard from, who are up here on the stage, also have colleagues that will be able to join from the table down in front. We have some that are calling in. And so if you've got a question in online, that's something that I'll be able to have an opportunity to be able to come to you and ask for a question. I would like to request, please, that since we have about 40 minutes left, and we've got the team here, but also Dr. Bardia and Lisberg, that we focus in on oncology questions. We are at an oncology conference. I think that would actually be great if we could do today. If there are other questions, we can park them, and we'll have the IR team follow up on those. So with that, maybe we can go to some questions that we have here in the room.

Emily, you were fastest why don't we actually come -- you're in front being a good student, too. So...

Emily Field from Barclays. Maybe just firstly, on TROPION-Lung01, could you confirm whether or not you have filed yet with the FDA? And in your discussions with the regulators, if you are expecting to have an ADCOM for that filing? And then secondly, just in the context of some of the other second-line AGFR studies that we've seen with MARIPOSA-2 and the HER3-ADC, if you could put the Dato-DXd data in context in the AGA population.

Okay. Susan, would you like to start first with the question on filing?

Yes. So as we said already, when we saw the TL01 data, we had a discussion with the regulatory authority. And on the basis of that discussion, they told us that we could file, and we intend to file, and what we will usually do is confirm when the file has been accepted. So in terms of the second part of your question about the relevance in terms of the AGAa, I think, of course, TL01 is a randomized study, and you've got a hazard ratio of 0.38 in the AGA subgroup. You've also got the TROPION-Lung05 data set. I don't know, Cristian, do you want to just describe that in a little bit detail and put it into the context? TL05 and the ORCHARD study?

TL05 is very -- I don't know if you have a chance to see this presentation. The data emerging TL05 actually very consistent with what we see in this trial, of course, in the AGA population. In TL05, we see an overall more than 35% response rate in the population. Even more importantly, 44% in EGFR mutant. The median PFS in that study is 5.8 months. It is 137 patient study, most of them EGFR. This is important because it gave us the confidence that Dato-DXd, as we knew, is confirming, is a drug that can play an important role in EGFR space. And we are also developing a combination with Dato-DXd and Tagrisso in our ORCHARD study. It is a platform study where we test in combination with Tagrisso. I can share with you that this combination is quite feasible. It's emerging promising. And we are planning, of course, to develop further the combination, also in light of FLAURA2 results showing that chemotherapy is added on top of Tagrisso. So if you think we may have a better chemotherapy, this is the buffer.

Thank you, Cristian. And I think also as useful, Dr. Lisberg certainly talked about the fact that the non-squamous benefit also had the AGA patients within there. While that wasn't driving the benefit for the non-squams, it's a pretty impressive hazard ratio. So I do think that, that will be a population that will be of interest. Richard, maybe we can go to you.

Richard Parkes from BNP Paribas. So firstly, on the discussion, I think, of TL01, talked -- asked the question why in terms of the better benefit within non-squamous population. I know you touched on the biomarker and that you're convinced that it, it's obviously going to be important to regulatory discussions. So can you just talk a little bit more about what you'll be able to present to regulators to argue for a biological plausibility of the greater benefit in that population and what data you'll have on the biomarker at that point? And then second question -- if you look at top line performance recently, quite often has been driven by new opportunities in small cancer types that maybe people had overlooked and it feels like DUO-E could be another one of those. One of the questions has been about how much of that benefit was driven by patients with DNA repair mutations. So, can you talk about how important that is to driving that benefit and potential use of that marker in this population?

Perfect. Susan, do you want to start with what we'll be able to share with respect to biology around the non-squams and anything else related to biomarker?

Yes. Well, I think the first point to make is, of course, that the non-squamous population is an easily identifiable population. The study was stratified for histology. And so the analysis by histology was pre-specified. So I think that's an important thing for the context of the population that we think is relevant for file-ability for TROPION-Lung01. I do think that the understanding of the prevalence of a potential biomarker by histology, not just in the TL01 data set, but more broadly is very helpful in that context. And of course, we'll have ongoing discussions with the regulators as we do that. And we'll share more about that when we're ready to publish those data. But I think for the purposes of the TROPION-Lung01, we're confident that we can file and we think the benefit risk profile that we've got is very positive in the non-squamous population.

Great. Cristian, do you want to take a moment to talk about DUO-E and the question that Richard asked?

DUO-E is a great opportunity for us. I think it's validating the concept that PARP inhibition and checkpoint inhibition can work well together in another indication. So this is, in our view, important. And to answer specifically your question, is likely independent of the HMR status or BRCA status, for the simple fact that in this indication, HRR is no more than 20%, 25% of the patients and BRCA is just a fraction of it. So this level of benefit cannot be driven by that population will present a further biomarker analysis later on. We are performing it.

So I think in terms of the point that you make also about some of the top line and the performance being driven by just the indications we move into. I do think, Richard, that we're very enthusiastic about the fact that with endometrial cancer, we've got a population of 30,000 patients in the G7. About half of those would be in the U.S. I think within that, we know that 75% to 80% are pMMR. Now our job is to really understand where can we drive the triplet. And I think that the market opportunity is certainly a sizable one within this. I mean, I think it's a blockbuster plus market opportunity. And then the degree to which we're able to drive triplet into this space, and I do think that we've got a couple of camps. There's some like the PI. I believe that she will want to incorporate it as soon as she has the opportunity to do so. There's other camps that I think are waiting to see more data, believe it's a more heterogeneous group, and I think that we'll have to navigate kind of both of those different groups and populations. Let me just stay at the table and we can go through those.

Eric Le Berrigaud, Stifel. Three questions. Maybe the first 2 to the KOLs. Can you please put maybe for the 2 of you, the stomatitis issue into some perspective because we saw that there were very few Grade 3 plus. But similarly to what we saw on the slide for Mariposa regimen, the skin rash, is it kind of grade 1 and 2, nonetheless, pretty difficult to manage in real-life setting versus clinical trials? The second question, if we compare maybe Dato to Enhertu, the overlapping population of the HER2-low, could we assume that Enhertu stays like probably standards, but how do you think about using Dato-DXd after Enhertu in the subsequent line of therapy? And given the same construct of ADC, are you comfortable doing this or not? And the third question is on maybe can you -- so maybe just for one of you in this case, but put the Mariposa and FLAURA2 into some perspective? And what if Mariposa eats the OS endpoint?

Mucositis in, with Dato-DXd is usually very manageable. If you look at grade 1 mucositis, that essentially refers to mucositis that is asymptomatic. Patient is not having any pain. There is erythema or ulcer that is seen an exam. And grade 2 is some pain, but does not interfere with PO intake. So both Grade 1, Grade 2 does not have an impact in terms of their ability to eat or in terms of their activities of daily living. And this is reflected in the rate of treatment discontinuation. If an AE is really toxic, we would see patients or providers discontinue the drug. But the rate of discontinuation -- out of 365 patients on Dato-DXd, one discontinued because of mucositis. So it just tells you that in terms of mucositis, it is something that's very manageable. In terms of your second question about Dato versus Enhertu, I think that's a very relevant question. And if Dato gets approved in metastatic hormone receptor positive breast cancer, it's a question we'll have to address in the clinic. We've not seen overall survival results yet from TROPION-Breast01, but if you purely do cross-trial comparisons between TROPION-Breast01 and DESTINY-Breast04, both the magnitude of median PFS and hazard ratio was better with trastuzumab deruxtecan. That's a subset of hormone receptor-positive breast cancer because hormone-receptor positive breast cancer is both HER2-low and HER2 IHC0. So for HER2 low, many physicians would prefer trastuzumab deruxtecan over datopotamab deruxtecan, but it's a discussion with the patient. There might be some patients who have baseline pneumonitis or there might be other considerations that come into picture. So it will be a patient-centered discussion. But if everything is equal, we're feeling the T-DXd would be the preferred agent over Dato-DXd, provided we don't have something in overall survival that looks really, really good. That could again change things. But for IHC0, they should be the preferred agent. Now in metastatic setting, we use drugs sequentially. So if a patient is on T-DXd and has disease progression, and after that, I have a choice, I can either use some other chemo agent like gemcitabine or eribulin, or I can use an ADC. And in general, the feeling would be to use an ADC because it's more efficacious and it's less toxic as well, especially with Dato-DXd. In terms of cross resistance, we don't have the answer... [Audio Gap]

What we're hearing in general is that trust is -- with osimertinib, the monotherapy convenience is a big, big opportunity. And really, what the clinicians want is they want the convenience, they want the oral therapy because it's a very patient-centered approach. What I should also highlight with FLAURA2, what we also heard it was there was a defined patient population that can drive a greater benefit with the addition of chemotherapy. And what the clinicians really appreciated is those patients with brain metastases and Cristian marked with some of the data and also patients with LA58R mutation, there's an unmet need, and they feel that there may be an opportunity to add additional therapy with chemotherapy. And the other thing to note, the difference between FLAURA-2, as well as Mariposa is a chemotherapy component, which is given that in addition to osimertinib, it's for a limited duration. It's generally they stop the chemotherapy of 4-cycles platinum and the pemetrexed after 8 months, then you have a prolonged monotherapy maintenance phase, as opposed to Mariposa, which is really treatment to progression. So then you have acute and chronic toxicities that really add up for patients. I think those premises of having a monotherapy approach, but a combination for select patients, and the toxicity that's limited during the chemotherapy portion are some of the reasons why some of the clinicians favor the chemotherapy plus Tagrisso approach for those defined patient segments.

Thank you, Sunil.

Can I add the point here? I want to clarify one thing because there was something in the discussion of the Mariposa that was not perfectly correct, and it's the scanning that we used in FLAURA2 regarding patients with brain mets. So I want to clarify that the scanning that has been used in FLAURA2 is extremely adequate. And every single patient has this can baseline. The patient with brain mets that receive a scan after 6 weeks, then another scan 12 weeks and then every 12 weeks. This is for the patient with brain mets. And all the patients or the patients without brain mets, receive anyway baseline and then add disease progression in neuro-scanning to document what disease progression is happening. And of course, if the patient is symptomatic, neurologically symptomatic, receive an additional scan to assess. So this is very adequate and very close tactics. So this is -- we believe that the -- actually the monitoring and the assessment of the patients in FLAURA2 was more than adequate. Just to clarify.

Thank you, Christian. So why don't we go to Andrew here. I should have had the front end ask for one question per person since we have so many hands that are coming up, and we have folks on it. I'm not sure, Andrew, you're the best person for me to put the one question rule on. But let's see how it goes.

That's a tough one. So could you talk to the time impact on your first-line non-small cell lung trials given you are enriching and whether that applies to both AVANZAR and to TROPION-Lung07. Perhaps also you could talk to whether overall survival may become a sole primary endpoint in those trials, if there's a rationale? And then finally, for Dave, I told you I wasn't good at counting. Finally, for Dave, could you talk to the commercial risk that physicians in the U.S. or rather hospitals may seek to use amivantamab because it's financially more lucrative for them. How does that in the real world pan-out -- or do you think the other factors such as tox management are going to overshadow it?

Susan, would you like to cover off on the time impact for the frontline studies and the endpoint question?

So I'll start and Cristian, because you're running these trials, you can comment. But basically, AVANZAR is going really, really well because there's a huge interest in that. We're actually well ahead of the projected accrual rate. So we are enriching for the non-squamous and actually just capping the squamous. So I don't actually anticipate that we'll have a substantial impact on the time lines.

Nothing to add on this. We don't expect a substantial impact on time lines because we dropped the squamous, so we increased the non-squamous. Of course, the scaling fare are going to be a little bit higher. But there is a very, very high interest over there. I want to answer also your second question on endpoint, OS always been a primary point in the setting of first-line non-small cell lung cancer, together with PFS. So this is not going to change because our primary end point is PFS co-primary points, PFS and OS. And of course, the PFS is already out earlier, NOS will confirm the finding in PFS, no change.

And just one other point. TROPION-Lung07 and 08 are being run by our colleagues at Daiichi Sankyo, but we anticipate that TROPION-Lung07 will have similar modifications.

Now TROPION-Lung07 only in non-squamous. So no modifications. Sorry, Susan. Yes. Yes.

Apologies.

TROPION-Lung08 is squamous and non-squamous. TROPION-Lung08 will have the same modification on AVANZAR.

In short, on the third question, I don't anticipate that the difference between infused and the way that the financials and the economics are going to be a major driver of differentiation here. I do acknowledge that that's certainly relevant within certain subsegments of the marketplace. I think in this particular instance, I'm not sure that that's really something that we see as being a big motivator and a differentiator of a driver of difference here. Luisa?

Luisa Hector from Berenberg. I wonder if you could share with us your sales in lung cancer today. And then given the strength of the pipeline that you're showing us, what long-term opportunity you see in lung cancer and what the key inflection points are for season to deliver on that? And then maybe, Susan, you could follow up with what you need to see in terms of the bispecifics, moving into frontline combinations with data or other ADCs that will push things forward.

Thanks, Luisa. In terms of today, I mean, over 50% of the oncology sales are coming from lung cancer between Tagrisso, Imfinzi with the PACIFIC indication predominantly, but also some with POSEIDON. And then, of course, we have some in Enhertu, as well within the space. Lung cancer is a place where we've got really a leadership position in the most significant scale. We've also had an opportunity to share that as we think about our vision for 2030 that we would aspire to having more than 1 in 2 of patients who are treated with lung -- for lung cancer to be treated with an AstraZeneca medicine. So you can really see that this is an area that we're moving behind. I think just today, and Susan, I think, showed a very powerful slide around this. Susan today just showed the slide that there are over 150,000 patients in the second line plus setting today in the G7 being treated with chemotherapy. That's an enormous number. Then if you add the front line, which is obviously not just as a monotherapy, but it's also in combination, it's another 100,000. So there's 0.25 million patients that are being treated today with chemotherapy, and that's just within a set of marketplaces where there is access already to chemo. So I think that it's really very clear that the opportunity that sits in front of us with Dato is key and important. Of course, the frontline studies are really very, very critical to being able to unlock and to get to where we need to be. And I think that the enthusiasm that we have also for the bispecific platform and the ability to be able to move into that space is also something that allows us to envision how the IO plus the ADC can both be components where those 2 angles are coming together. And I do really think that is the thesis that we have at AZ, which is by attacking the tumor from both of those angles that we're able to really make the best outcome for patients and have the best return. Susan, do you want to add anything to that?

So the first thing I would just say it's exciting to see that we've actually dosed the first patients in our first Phase III trials for volrustomig in -- with the eVOLVE-Lung02 and eVLVE-cervical. That's initially in combination with chemotherapy or so there's work ongoing to look at combinability, too. To answer your question, combinability and safety is the first question that we're answering. And we're gathering data across our bispecifics about the patient populations that we think are going to be best for the different members of the bispecific portfolio that we've got. As a reminder, all volrustomig PD1/CTLA4 designed to have lower CTLA4-based toxicity. We also have a PD1/TIGIT, which might have some benefit from having the 2 targets in some cooperative binding. And then we have a PD1TIM3 as well in development. So there's a lot of work that's actively ongoing. Matt Hellman is leading the early oncology program and helping to drive forward those programs.

Thank you, Susan. We'll take one question from on the line. Steve Scala from Cowen. If you would please go ahead with your question.

Several years ago, AstraZeneca set Tagrisso guidance at $5 billion at peak. Sales are already above that and growing. All things considered, what is AstraZeneca viewing as likely peak Tagrisso sales now? Based on your confidence, it would seem to be well above that, but maybe you can please tell us.

I mean I do think that going back to the conversations that you had said, we were enthusiastic very much at the time to be able to get to something near that $5 billion number, and we talked about the importance of really getting duration of therapy on ADAURA. We -- excuse me, from FLAURA, we spoke then also about the opportunity to be able to move from there into ADAURA and to be able to do that across the globe. I think what's been really great to see is the progress that we are making across regions with ADAURA has been important and continues to be part of the growth that we have going forward. We also are continuing to get continued duration of therapy out of FLAURA. And then, of course, the FLAURA2 data represent an opportunity to really, I think, be able to in a certain subset of patients potentially have a longer duration of therapy and offset any share losses that we might see from the other combination. I guess what I would point out, too, in terms of the other growth opportunity here. We look forward to the LAURA study readout. I think that this is another segment within the Stage 3 unresectable that's certainly sizable. And I think that, that would be a real catalyst to allow us to be able to move there. In terms of a specific number, I won't give in to long-term guidance on Tagrisso, but those are the elements that I think that contribute to the opportunities. Maybe another question back in the room -- why don't we stay here at the table here?

Tony Ren from Macquarie Capital. So a quick question, first of all, on the ILD. I think Dr. Bardia mentioned that ILD is much lower in breast cancer. Is it a matter of patient selection or mitigation strategy that's being implemented earlier there? Or is it -- do we have some pathophysiological rationale behind that? And the next one, again, I want to go back to the FDA application for TROPION-Lung01, the -- so my question is, will they approve the drug based on a subgroup analysis, basically, by definition, lacking statistical significance? Would they give a conditional approval, then, waiting for OS outcome, would they require a new study powered for the non-squamous histology? Because obviously, the 0.7 month PFS win in the all-comer population is below the 6-week scan interval, right? So again, if you could provide some clarification.

Super. Why don't we start first with Dr. Bardia answer the question on your experience with ILD within breast cancer, and then Susan will come to you to address the second question.

There are 2 components to the question. The first is the incidence of Dato-DXd ILD in breast cancer compared to other cancers like lung cancer. And the second is Dato-DXd as compared to, say, T-DXd in breast cancer. So if you look at the first component, I think the reason could be the natural history in the treatment patients receive in breast cancer. In lung cancer patients often get radiation, which can impact lung parenchyma, which can increase the risk of pneumonitis and Dr. Lisberg speak to that as well. In breast cancer, in general, that is not common. So we've not seen that degree of pneumonitis with the same drug, which is Dato-DXd in breast cancer as compared to lung cancer. As compared to why it's lower as compared to trastuzumab deruxtecan. That could be a couple of reasons. First, HER2 is expressed in lung parenchyma. And if we look at any anti-HER2 agent, even TDM-1 that can cause pneumonitis. It's just that the incidence is lower. And the second is drug antibody ratio is different. In T-DXd, there are 8 molecules as compared to Dato-DXd therefore.

Was there something you want to add in?

I think it's just one thing that's really important to note is that the clinical community does adapt and learning how to manage and identify those patients in a much better way. If you may recall, when we had an Enhertu first come out in late line setting, the higher rates were much higher for higher-grade toxicities. And you get better, you recognize better, you mitigate much better. And I think that leads to a much better adoption, a much lower rate of higher-grade ILD. And I think that's going to happen, with Dato, it's going to happen. With TROPION-Lung setting as well with Dato-DXd lung cancer.

Great. Susan, in terms of the second question?

Yes, sure. So just as a reminder, of course, TROPION-Lung01 on met its primary endpoint with statistical significance in the primary endpoint. And I think in terms of OS, I think there is a precedent that OS is not required to file, but particularly showing no detriment. And again, in the non-squamous patient population, the point estimate for the OS at this point, 0.77. Confidence intervals only just crossed one in that subgroup. So I think there's a strong trend to OS benefit. But there is precedent for having approvals in a subgroup that's driving the benefit. There have been a number of trials recently where it's shown that. And I think within lung cancer, pemetrexed is a good example of driving the benefit within a non-squamous subgroup and that leading to an approval within that histology-based indication. So I think whilst we -- the precedent is you have to file on the ITT, we'd expect that doing the regulatory conversations, we'll focus on the subgroup with the positive benefit risk ratio. From our perspective, that's in the non-squamous population.

Great. Thanks, Susan. So I think we'll continue to take questions here in the room. We have also some online. We'll go till 9:40. I know we're originally planning to go to 9:30, but we'll do that just in an effort to try to get to the questions that we can. Maybe we can go to the table in the back, just right here.