Roche Holding AG (ROG) Earnings Call Transcript & Summary

Okay. Perfect. I think we can start on time. And welcome to our Annual Pharma Day here in London. It's, again, a great pleasure to have you with us today and to see so many known faces. We hope we have been able to put together today an interesting agenda, which will cover the areas which are probably of highest interest to you. And of course, we are very much also looking forward for lively discussions over the event. Going to the first slide, I just will take you through the agenda today. Let me quickly take you here through. Our event will start with -- it' will be started by our new CEO, Thomas Schinecker, who will provide a group update, I think, with a clear focus on the topic of R&D excellence in pharma. Second presenter will be our new pharma CEO, Teresa Graham. Teresa will take us through the on-market portfolio, focusing on additional growth opportunities, but also briefly summarize our mid- and long-term opportunities, which then will be covered in more depth in the following presentations. Our third speaker for the morning session then will be Levi Garraway. He's Head of our Global Product Development and as a new member to the CEC, as you know. And he will focus -- start the pipeline deep dives, which will follow and he will cover solid tumors before we go into the lunch break. Following the first 3 presentations, then we will have a 50-minutes lunch break, which will go from 10 past 12, is it -- or 12 to 12:50, sorry. And in the lunch break, of course, you will have the opportunity to meet all of our speakers and also all of the IR officers will be around. After lunch, then we will continue with our deep dive into the pipeline, which will be primarily late-stage, but I think we also will highlight some of the assets, which are on the turning point to potentially go into pivotal studies rather soon. The first presentation in the afternoon then will be given by Charlie Fuchs, our Global Head of Oncology and Hematology Product Development, who will take us through malignant and nonmalignant hematology. The second afternoon presentation then will be by Paulo Fontoura, our pipeline, which will be primarily late stage, but I think we also will highlight some of the assets which are on the turning point to potentially go into pivotal studies rather soon. The first presentation in the afternoon then will be given by Charlie Folks, our Global Head of Oncology and Hematology Product Development, who will take us from malignant and nonmalignant hematology. The second afternoon presentation then will be by Paulo Fontoura, our Global Head of Clinical Development for Neuroscience, Immunology, Ophthalmology, Infectious and Rare Diseases. And Paulo will provide us an update on both our pipelines in neuroscience and in immunology. And I think we have a third session then later, the final session, which is here then by Chris Brittain. He will take us through the ophthalmology pipeline, which is emerging and which is nicely filling up with new molecules, which we will showcase here for the first time. The event -- so after the second round of presentations, we have scheduled 45 minutes of Q&A with all of our speakers being onstage. And please be aware that you can also send us the questions via the chat tool or dial-in, and I will have a look and then pick some of these questions. The event will close at 2:30 British Standard Time. But you're, of course, invited to stay around afterwards for a buffet reception, which will give you an additional opportunity to get in touch with all our senior management. Let me also please point out here that we have prepared, again, a short survey just to collect your feedback. It's 9 questions. The link to the survey will be shown to the participants in the webinar roughly 10 minutes before the closing of the event. Participants in the room, I think, we'll get an e-mail where they can take the questions. And of course, we would very much appreciate if you would provide us here a little bit of feedback as we always strive to improve. So yes, I think that's basically all the housekeeping items I wanted to mention. And with that then, I would like to hand over to Thomas Schinecker for the group update. Thomas, please.

Thank you very much, Bruno, and I'm really happy to be here today with all of you. Some of you I've met over the last 4 years when I was the Diagnostics CEO. But I thought, since I'm now in the role since the middle of March it would be a good opportunity also to engage with all of you here at Roche Pharma Day. So I will cover 4 things. The first one is the new leadership team, which is in place since 1st of April. I'll also talk a little bit about performance and outlook. I'll talk about pharma R&D excellence. One of the topics that's obviously on our mind, but also I know that's on your mind. It's one of 5 key priorities that we've established in March, what the CEC has focused on during this year. And I will highlight also some upcoming IR events. Since April 2023, we have -- we established a new leadership team. Together, we are 12 people. We have a core team with 5 people and to the extended leadership team that has also another 7 people. You see a good mix of people that have been with the company and in these roles for quite some time. I think Alan Hippe is very well known and well respected, a very strong finance CFO, and for example, also Cris Wilbur but also others. What we have done in terms of changes already back in April is that we've actually lifted up the product development organization, which is under Levi Garraway, our Chief Medical Officer, into the CEC. And with that, we have basically an end-to-end view when it comes to our research and development engine, from early stages like pRED or gRED to late-stage but also to partnering. And as you see later, we formed kind of a sub-team out of the CEC that focuses purely on research and development and innovation so that we have a team that actively supports the organization in managing the portfolio. So when we look on the next slide, who is new to the team? It's Teresa, who you see right after me. She is also quite well known, I think, in this community. She's been the Head of GPS, to Global Products Portfolio Strategy. And now since, I think, March, she's been the CEO for the Pharmaceuticals division. Matt is my direct successor for Diagnostics. He's led the U.S. organization before, but he's also spent about 20 years at Roche, both in Pharmaceuticals and Diagnostics and has always shown that he can really deliver. And I think that was also a very important element why he got the job. Levi I already talked about. And Silke is the new Head of Corporate Strategy and Sustainability. She is now filling a new role, a role that we didn't have on the CEC in the past. It's, on one hand, good because now we have a strategy role there to which we can go to and we have a sustainability anchored in the CEC. I think what all of them have in common is they're all down to earth, which I think is an important element. Very approachable for people because I think people go to work and deliver the best work if they're working for leaders that have those attributes. They listen, they empower, but they're very strong in implementation, all of them. And I think that's also something that's very important to me also when I choose leaders. There are these leaders that use a lot of buzzwords. And then if you just scratch the surface a little bit, you notice that actually not a lot of things are happening. I would rather have people that maybe talk a bit less, but then deliver. And at the end, if you think about words are 10%, strategy is maybe 10%, delivery is where you make a difference. And that's a big focus for us. And with sustainability being now on the CEC level, we can also weave that in much more throughout our entire value chain. I think we have a very good story alone with the fact what we do for patients around the world, where we deliver almost 30 billion tests, we treat 14 million patients. But what we do also for the environment or also in terms of the economy. I think in Switzerland alone, I think 40% of all exports in Switzerland are from the pharma industry. And if I have the number right, I think 10% to 13% of that is, alone, Roche. So Switzerland is, I would say, to a certain degree, also dependent on Roche. We have a very strong and good history in sustainability for more than 50 years, and this is due to founding family. The father of André Hoffmann, founded the WWF, for example. And so for this time already, we've been working very hard in bringing down our CO2 emission. For example, when it comes to building new sites, even in the U.S. maybe where sometimes the standard was not at the same level as we had put internally, we always make sure that we think ahead, we put in the highest standards in terms of sustainability. And we don't -- we, of course, adhere to the law, but we were better than the law. And this also translates into the Dow Jones Sustainability Index, where in the last 14 years we were 12 times we were the #1, 2 times we were the #2. And the last time, the only one that beat us was Chugai, so I count this as #1 as well. Now let me take you through a bit the performance and the outlook. First, to Roche, we've delivered 14 million treatments to patients in 2022. We conducted almost 30 billion tests. And we are located in all parts of the world. So we have a very diverse span across the world in terms of R&D. We have strong R&D in the U.S. We have strong R&D in Europe, in Switzerland and in Germany. We have strong R&D in Japan and we have strong R&D in China. So if you think about where all the innovation is coming from, basically with those 4 countries you're already covering 80% to 90% of where the innovation in the world is coming from. If you also think about the potential markets in the future, you're also covering a large part of where the markets are in the future. And all these R&D sites have delivered over the last years. We had, in 2022, 16 blockbusters, so 16 medicines that deliver more than CHF 1 billion in sales. In 2012, it was only 7. We are the #1 in the areas like neurology, hemophilia A and #1 in, in vitro diagnostics. And yes, it's true, we're #3 in oncology. And this is, of course, because of some immunotherapies really taking off like we know with pembro. We are very highly rated by patients, for example, by the rare disease patient groups. But also when it comes to pharma AI readiness, we were ranked #1. And we have a track record to deliver new medicines that are really changing the standard of care like OCREVUS, HEMLIBRA, VABYSMO or Evrysdi. And oftentimes, we've entered in new disease areas where we were not present before and we were successful with that. And now if we look back over the last -- exactly these 10 years from 2012 to 2022, you can see that in 2012 we still had 52% of our sales in Avastin, Herceptin and Rituxan or MabThera. Now we have 14%. You see that we have massively diversified our portfolio, which was our strategy to diversify our portfolio. And this was one of the biggest patent cliffs that the industry has ever seen, and we've grown through this patent cliff and not many have been able to do that. And we've done that by launching a lot of new products. If we look back to 2015, the end of 2015, we launched 20 new medicines in this time frame. In fact, now more than 50% of our sales are from these new products. And all but 2 will not lose exclusivity in this decade. And the 2 that will, will lose it at the very, very end of the decade. So this is the portfolio that will continue to drive growth in the coming years. Now, as you know, we don't give a year-by-year outlook, right? So what I thought I'd just take your numbers and just put your numbers on the slide and just give some context to the numbers that we see in the consensus. So this consensus is built by the 18 different consensus from different models. And this is probably the most, yes, covered or, let's say, the broadest consensus that you can achieve. So the more accurate you can get. On the left-hand side, you see what you think our biosimilar impact will be over the next couple of years until 2026. And you say roughly CHF 8 billion. right? And this, by the way, includes Ronapreve and the Ronapreve fad will be over by Q1 next year. And you see all the other medicines that have an impact where we decline. On the second -- in the middle column, you see all the new medicines, the ones that I've shown on the previous slide and the growth that you expect from these new medicines in this time period in absolute numbers. If you add that all together, you're close to CHF 14 billion, yes? So if these things are correct, then you see we more than fill the gap of the biosimilars erosion until 2026. The other thing is if you look at the Phase III pipeline, only CHF 2.4 billion are represented. And in fact, these are the Phase III assets, which are covered by 50% of the 18. So more than 50% of the analysts cover -- from the 18, so about 9 cover these. Many others are not covered at all. They're not even in the model. In fact, if you look on the right-hand side, and there are some Phase III assets there, none of them are in the model, right? So what you see is that the pipeline is not very much reflected currently in the consensus, which should give us upside, especially when we think into growth in the later part of the decade. And this doesn't even include anything that we would do on the partnering side. Most of these are either not covered at all or may be covered by 1 or 2 analysts. And that's why we didn't pull it into the consensus model because it doesn't have the size that we can fundamentally take it into the consensus. I think this is a very important slide, and you see the slide at the end as well because today we'll talk about all these products over here, but we'll also talk about these products here to make sure that maybe there will be more attention to also products that we have in the pipeline. And many of them are actually Phase III. Now beyond Pharma, I think we have a very strong pipeline in Diagnostics. So Diagnostics, we say that Diagnostics will continue to grow mid- to high single digits for the foreseeable future. There are -- and the market is around mid-single digit. There are, however, inflection points. And these inflection points are the right 3 launches that we'll see in the not-too-distant future. Our base and our strength is our Core Lab and Molecular Lab portfolio, where we have more assays than anyone else in the market. And we have more installations than anyone else in the market, and this will continue to drive growth and will continue to build out our leadership position because we launch more assays every year than any other company. To give you an example, between 2019 and 2022, now I'm discounting the emergency use authorizations. I'm just saying real FDA approvals. In these 3 years, Roche has launched more products in the U.S. than the entire rest of the diagnostics industry together. So you see the power we have in terms of innovation and also productivity in delivering new products. And we spend more in R&D than the #2, 3 and 4 together, but we don't spend more than the entire rest of the industry together. So you can see that we deliver very strongly. Mass spec is going to revolutionize the lab. Today, you have to go in a separate lab. It's a lot of manual labor to do these tests. This system will be fully integrated, fully automated. And at launch, we will already have 40 parameters at launch, and we'll continue to extend that. I think in the second wave, there are another 30 parameters that will come. And again, no one has such a system. And especially because we can combine it with immuno and clinical chemistry, there will actually be a pull-through effect, which means companies that don't have that they will also lose in clin chem and immuno. This is a differentiator. We've seen that now in a number of tenders where people say, we want the system and you get clin chem and immuno because it's the better package. Continuous glucose monitoring, something that we've been waiting for, for a long time because we've been suffering from a technology conversion from BGM to CGM. But over the last couple of years, we worked very hard on this and we have absolutely the freedom to operate to move on to the market with a competitive solution. And finally, next-generation sequencing, where we've also made a lot of progress. Now this nanopore sequencing technology overcomes the hurdle of cost and throughput that other nanopore technologies have. And it's highly accurate. Highly accurate, fast, low cost. And what it also can do compared to other technologies, you can -- the cost is not dependent on getting a plateful, but it's dependent on what you want to sequence. So if you have less to sequence, you can bring it at a cost, which is very competitive. So you see also in Diagnostics we have things to show. So again, looking at the performance this year, I think it's a very strong performance from the group. In Pharma, we are growing 8% and 32% growth is coming -- or 32% growth is from the new products that we launched in 2015. In Diagnostics, the base business is growing 6%, which is absolutely in line with what I've told you before. Now here, we had the COVID effect and this COVID effect will wash out until Q1 for both Pharma and Dia. We had CHF 2.7 billion already in the first half of the year in negative COVID effect. If we would lose every single cent that we had sales in COVID in the second half of last year, which we know is not going to be the case then it would be, at the end, about CHF 4.7 billion or CHF 4.8 billion that we lose. So we are very much in line with what we said. And also in AHR, we see it slowing down. The erosion is slowing down. Now let me take you to the main part of the presentation, which is the Pharma R&D excellence piece. At the beginning of the year, so in March, when I took over, we set 5 key priorities for the Corporate Executive Committee. One was around strategy and our 10-year ambitions. We said our strategy is fundamentally in the right direction, but there may be areas that we can sharpen. So we did surveys, we did workshops with people in the organization to understand where do they see opportunities to improve from a strategic perspective, but what else should we look at to make sure that we set up the organization for the future. And it's very important to do that because that gives you a buy-in from the organization to change things, right, because then they say, yes, we saw it and we need to do something about it. And I think it's also important, as a leader, that you tackle those things. Because if there are things in the organization that swirl where everyone says, why are we not doing this, why are we not fixing this and you're not addressing that. It will take energy in the organization that you're not [ penalizing ] to where it should be. I also think in an organization, clarity is very important. In the organization sometimes I compare it to people rowing on the boat. Let's say, there are 8 people rowing on the boat. If all of these people row in 8 different directions, what will actually happen is that the boat does not move forward, it actually just turns and swirls. But if everyone is clear of the direction that we are heading to, then the boat is going to gain speed and reach the goal. The second thing beyond strategy and 10-year ambitions, it's about the R&D excellence piece, which I'll talk about. The third one was around partnering. The fourth one was about performance management system. So we looked at how can we foster a higher performance culture in our organization through our system. And the fifth one was also to understand what are the new areas in science, new therapeutic areas, where there are breakthroughs where we can play a big growth. So those are the things we looked at. Now the Pharma R&D excellence work started with a fact base and with a root cause analysis and then with the solutions. Now we don't have time to take you through everything. Alone, the fact base was 140 pages. But I can guarantee you that we took a very clear look what's going well and where can we improve. And we had the opportunity in July then to present the outcome to the Board. And last week, we presented it to the organization. And now we're moving into implementation. And as I said in the beginning, implementation is everything. And I can guarantee you, I have a very strong focus and also track record in implementation. So now let me take you through it. There are definitely strengths that we have in our organization. For example, commitment to scientific excellence, discovery of novel modalities like bispecifics and ADCs. Ability to expand into new therapeutic areas, we've done it multiple times. I think the fact that we had autonomous units also led to the fact that we've launched 20 NMEs in the last 8 years, but there are also things that we can do to improve, right, because autonomous doesn't mean that everyone should have their own systems and processes. That just slows you down. That's not -- has nothing to do with freedom if you look at harmonizing these things. I think we have great talent and mindset in our organization, especially also around computation biology, disease models and protein engineering. And we have an asset which is a strong loyalty to this company. When people come to Roche, they want to -- for them, it's not just a company. For them, it's really giving their heart and making sure that this company is successful. And we have a number of leadership in generative AI and learning approaches in our teams. Now if you look at the numbers. We are very committed to innovation. We are one of the highest spenders, probably the highest spender in research and development in the industry. Now these are only the numbers for Pharma. So Pharma spends around CHF 12 billion per year at this stage. Diagnostic spends another CHF 2 billion. So we're at CHF 14 billion for the group. What you see what we have done over the last couple of years is we have reallocated money from other operating cost lines into R&D. And overall, we've kept our margins stable. Now the focus is going to be how do we get the most out of the money that we have put in. And that's what I'm going to talk about. On the right-hand side, what you see is Roche compared to all of our peers, how many NMEs we've launched. And on the blue bars, you see a calculation. This is not a Roche calculation. It's, I think, from Evaluate Pharma. The products that were launched from 2017 to 2022, how much of the total on-market portfolio NPV this contributes. And with that, you see that Roche has one of the youngest portfolio. So we have about 30% -- 35% of our NPV is covered by a young part of our portfolio. And you see the situation in many others, I would say many others would be happy to be in the situation that we are in if you look at that list. Clearly, there are always people that are better. So there are 2 companies that are ahead of us in this analysis. But in absolute numbers, we're actually pretty much the same as the top 3. And on the right-hand side, you see also the EBITDA margins and average R&D spend as a percent of revenue just for completion's sake. Now when we looked at R&D, we really did that end-to-end and we looked at it based on this formula. Success rate, the higher the success rate, the more you're going to get out. So the success rate in the different phases. The more volume, so the more assets you have in the pipeline is the factor and the value. So if you only research in low-value assets, obviously the number on the left-hand side will be lower. But if you have more high-value assets, that will go up. On the denominator, you have cost and you have cycle time. So these are things that work against you, which you have to minimize as much as you can. And we have looked at all of these things in very great detail. I'm just going to show you some highlights. I don't think it's going -- so the first one is success rate on the formula. And I'm not showing you something that you don't know because this is what I've been reading whenever you have your reports. So obviously, you know about it. So if you look at the Phase III success rate that's on the right-hand side, the Phase III success rate in the industry is around 75%, 76%. And recently, we had a downturn and we had a number of trials that were not positive. But it was also not 0, just to be clear. Interestingly, if you look at the same time period, you look at the end to end, so from the very beginning to the very end to the approval, you see that we're in the top quartile. So what does that tell you? That tells you we're pushing too much risk from Phase II into Phase III. Some of these projects, we should probably have stopped earlier in Phase II. And what we see is exactly when we do a deeper analysis is that actually those projects that have been derisked more in Phase II have a very high success rate in Phase III. So there, we're talking about 80%-plus success rate, whereas those that have not been so derisked have lower success rate. Now you may ask why in h*** did we do that? I think one answer was, for example, in immunotherapy with TECENTRIQ, there was a push to go direct into Phase III, skipping Phase II, in order to catch up, in order to play an important role in TECENTRIQ. But clearly, we have stronger gates in making sure that only assets make it into the Phase III that clear the bar which we're setting and this bar includes multiple factors. One is do we have the data? How convincing is the data? What's the probability of success? And also what's the value because you may take different risk/value ratios depending on that. Again, looking on to the left, I just wanted to show this as well. Actually, if you look at the end-to-end, we are pretty much top quartile. Then the volume piece. And again, we have much more detail than these, right? But on the volume side, we actually have the second largest pipeline in the industry. Where we clearly want to be even better is in the late-stage in the Phase III. And that's a big focus of ours. We have 11. We are, I would say, among the best. There are 2 that are a bit higher, but we have 11 assets in Phase III. But we want to make sure that we get more into Phase III without increasing the risk. So that will include looking at external opportunities, but also what are the things that we can do to move faster through different phases. When it comes to value, and one measure of value is the potential revenue that we generate. It's one way. Another way to look at value is which medicines have or NME have the potential first-in-class. And here, again, you can see that we are #2. Again, not our data, external data. But what you can also see is in this others group, we have 45. So what we also said we'll put in more gates where we have more independent reviews because no one likes to kill their own project, where we have more independent reviews and then challenge and then move these projects out and move new, more innovative projects in. The other thing that we have seen is actually if certain projects go beyond a certain period of time. It's looking at it from a statistic analysis perspective. The chance that these projects become successful actually go down to 0. Again, somewhere you can put in the gates where you can make sure that these projects are then reviewed and not dragged on. Doesn't work? So this is the area where I would say we are below average, right? I think the other 3 we were actually above average with some areas of improvement. When it comes to cost, we're above average. And this clearly is also a reflection of our very decentralized model. That doesn't mean that you can't have 3 units. But if you have own systems and processes in each of these units, this actually will bring down the cost -- increase the cost in the system. So there are clear ways in harmonizing system processes across the organization where we can bring down cost. But there are other measures that we are taking also to make sure that we bring down the cost per NME. When it comes to time lines, that's the fifth dimension. We are actually pretty much on par with the industry. We're actually faster than most in the Phase III. We're a bit slower in Phase II, so we're looking to that. But we also looked at the white space between the different phases where we also have opportunities to accelerate. The systems and processes will help because if you have to take data out of one system in one phase, put it into another system, that's just waste of time. Another thing that will help actually also by the -- with the cost of the trials, but also with the speed is that if you compare our trials with any other trials, we actually have more end points than any other company. So we can simplify the trial, which will actually bring down the cost of the trial, but also what it does is actually you're much faster than recruitment because people like to recruit easy trials, right? So there are easy ways -- easy fixes that we have already in store and that we're now rolling out. So again, just looking at the whole calculation, you see that we have an opportunity in all of those areas. If you look at success rates, yes, we had a dip in Phase III, but look at it end-to-end, we're in the top quarter. Volume, we are the second in volume. In first-in-class, we're the second in the industry. Costs, we're below. In cycle time, we're in line. So there is opportunity to improve, but there are also good things that we want to keep. And the way we will work differently also in the future, and we have already implemented that is that we look at the portfolio completely end-to-end. So in the past, we had the early stage portfolio committee. Now Levi and his team will have a much stronger say in the early-stage portfolio committee and we have late-stage portfolio committee. And on each of these committees, people made decisions based on individual projects. With this team, we will look at the portfolio health overall and that the mix between low risk/high reward, the high risk/low rewards in these different categories is in a good shape that we have areas where we take a lot of risk, but that we also have areas where there's a higher level of certainty. So this mix is going to be important. And again, this is the team. So it's basically the CEC that's focused on R&D, including Cristin, who is heading Global Product Strategy. What's also important is that we continue to fund or bring in new innovation outside -- from the outside. And this will be also clinical stage NMEs, and that's a big focus of us. That's something that I mentioned as one of the 5 key focus areas. Also, we need to evolve our R&D engines. On the one hand, through computational biology, Aviv and John Marioni, we are working on our own generative AI, ChatGPT for Roche. And I think there is a lot of opportunity if we do this right, and I think we have some of the world's experts that will help us do this. And in the future, not only will they do this for Genentech, they will do it across the group. And the same thing goes for organoids where we can also test medicines faster. This is also not going to be done only for pRED, but we want to leverage that across the organization. Finally, let me just say on the cultural aspect, these are operating principles and how we want to live our culture. Putting patients first. We are a very ethical company, and patients need the medicine fast. So we need the sense of urgency in the organization. When you follow the science, it's important that we will continue to strengthen our focus on debates, disagreements to make sure we make the best decisions. When we act as one team, it's important that we lead collectively in that there are no egos. Decision should not be made on who has the loudest voice, but it should be around meritocracy. Embracing differences goes in the same direction. It's about idea meritocracy. It's about empowerment in decision-making but not laissez-faire. Laissez-faire would mean everyone can do whatever they want. But if you have certain guardrails like we have to harmonize systems and processes, that's an opportunity for us. Also when we simplify radically, we have an opportunity to reduce the amount of processes and reduce the burden on the organization, but not to 0. Sometimes people say all processes are bad, that's not true. The original idea of processes is to be faster, but you have to be quite selective in that. And importantly, make impact now, we need to deliver value fast and that's the mentality that we want in the organization. And so we have had these 4 -- 6 areas where we are going to focus on to strengthen our R&D engine. On the one hand, you see in areas where we want to improve, we give clear targets. And we incentivize those targets and we also change our incentivization system. We're strengthening our portfolio framework. We have more gates. We have more discussions, so we attrit faster. And we manage the portfolio as a whole. We look at external innovation. We really optimize our R&D engines and take out all of the costs that we see here. And at the end, it's all about talent and culture, which also means we need to bring in the best people in the industry to help us do this. Now finally, let me just highlight the 3 IR events. But this is not yet working, yes. Neuroscience update, October 30. You see a number of very important data that will be presented, OCREVUS; fenebrutinib; trontinemab, the Brain Shuttle antibody against Alzheimer's; gamma secretase modulator is also on Alzheimer's. The Roche Digitalization Day and then also the Diagnostics Day where you see a lot of these products that I've highlighted before. So I hope that you can join us there. And with that, I hand over to Teresa.

Thanks, Thomas. Contrary to the pictures they showed, I do actually have all my teeth. So that will be exciting for all of you to work on that going forward. So it's great to see all of you back in person again in London. And it's my pleasure today to share with you a quick update on Pharma, so just a little bit of re-level setting where we were at half year. Then I'm going to go into a deeper dive on the on-market portfolio and then we'll tease a little bit what's going to happen in the mid- and late-stage portfolio, much of which my R&D colleagues will be covering later this morning and this afternoon. So first of all, I'm sure you're all very familiar with this slide. In the first half of the year, Pharma grew by a very impressive 8%, adding CHF 2.7 billion in sales. We grew very consistently across all regions in the world. We grew by about 14% volume with a 6% price mix impact, and importantly, a 6% currency headwind. And unfortunately, with the strength of the Swiss franc, we would expect that headwind to actually increase as we head into the back half of the year. Those kind of sales numbers don't just happen. They happen because there are people in the field every day who are out there interacting with our customers and bringing our innovative medicines to patients. Five, 6 years ago, you heard us start talking about changes that we were making to our field and to our marketing teams. And we're now at the point where we've had a sufficient enough -- amount of experience with those teams in place, but I wanted to give you a little sense of how is it going. So first, let's just level set ourselves on why did we make these changes to begin with. And I think some of the rationale is pretty clear. Clearly, the composition of the Roche portfolio was changing. We were going from sort of a one therapeutic area company that was dominated primarily by 3 major assets with Avastin, Herceptin and Rituxan. We knew we were going to have a much more diversified portfolio across many different therapeutic areas. We frankly knew, however, that what physicians needed from us was changing quite dramatically and I think you can see that from this graph. Physicians no longer want the traditional large tell-and-sell models coming into their offices. And that was before COVID. And now that we've been through a global pandemic, many, many institutions just simply don't want a lot of people running around interrupting patient care and potentially spreading disease. So I think what we are really seeing is that the changes that we put into motion a number of years ago are actually really custom fit for the portfolio that we're building and actually for the needs of physicians today. We are very much leaning into data as a disruptor. You heard Thomas talk about the work that's happening in sort of generative AI on the research side. We're also bringing that into the commercial side, thinking about how we can arm our field personnel better. But we're really putting things into motion that allow us to have a much different and much more productive and efficient conversation with our customers. So what exactly is it that we did? So like many people in the industry, not that long ago, we had a very large sales force that was dedicated primarily into one product or one therapeutic area. A very traditional, again, sort of tell-and-sell model. What we moved to is more of a primary point of contact. So while we still have all of the specialist roles that still exist in the company, right now a customer has one point of contact that they can go to when they have questions. That point of contact has a fully empowered team around them. They take their knowledge of the health care system, of the drug, of the patient journey and they can actually solve a problem for a customer much more quickly. So you probably say that looks great on a PowerPoint slide, but what does that actually mean in real life? Well, here's a great example of what I mean in real life. Recently, in Spain, we had an infant who was diagnosed with SMA. Despite the fact that we had EMA approval and Spain had actually agreed to reimburse Evrysdi, the regional formulary had not yet been updated and so that child was denied. The treating physician called their primary point of contact. That primary point of contact within the day put a revised dossier in place with a full understanding of what that patient needed, what the regional office was going to likely need to see to reverse the decision. Twenty-four hours later, that decision was reversed and that child is now on therapy. That situation is playing out around the world every single day. Not only does it mean that our drugs reach more patients who need them when they need them critically, it means that we build a different level of trust and a different level of partnership with our customers. And that is going to be transformational for us as we think about launching all of the innovation that we have ahead of us. So how do I know that this has been successful? I know that it's been successful because I see the sales numbers. VABYSMO is our first new molecular entity that we're launching globally in our new model, and I think we can see the pickup with VABYSMO was really underscoring the success of this model. Thomas talked about the increased flexibility that we've gotten from a P&L perspective with this new model. Interestingly, it also provides us the ability to flex resources across our sales forces. And so we're no longer a switch that you flip, your field force is either on or off. We're a dial. When we need more over here, we can turn it up. We can reallocate it where we need it. And it allows us to be much more effective and efficient in our M&D spend. And finally, we're getting tremendous amounts of positive customer feedback. And so this is just one example of how the journey that we've been on over the last 5 years is actually really paying off for our customers. And to Thomas' earlier point, it's allowing us to implement in a way that is very new and different and will allow us to bring our portfolio to life. Thomas also talked about our commitment to sustainability. And clearly, you can have the best field force in the world, if patients don't have access to your products you have a real barrier. And so clearly, is very important to us that we are continually looking at our corporate access goals and thinking about how we can continue to make progress in some very critical areas. Inclusive clinical trials. It is no longer an option for your clinical trial not to reflect the demographics of the patients who actually suffer from the disease that you wish to treat. That is just something we are going to have to do better at as an industry. Regulatory filing and reimbursement. Once we know we have a medicine that is safe and efficacious and brings benefits to patients, how quickly can we get it through the regulatory process and with the amount of data that will mean it gets reimbursed in countries around the world. Affordability, we'll talk about that in just a minute. But then this other piece of capacity enablement and partnerships. What anybody who follows the health care industry knows is that there is no single part of our sector that can do this alone. It requires pharmaceutical companies, nongovernmental agencies. It requires governments, it requires physicians. It requires patients to come together to create holistic solutions, often building different kinds of capabilities that exist in our health care systems today. And that really is something that we are committed to doing. Health care is fundamentally delivered locally for people, by people. And it is through the kind of innovative things that we have put in place over the last several years that we are able to build those local, expanded capabilities that hopefully create environments where truly innovative medicines can make it to patients. And of course, any conversation around access would not be complete with the conversation on pricing. And this is probably one of the areas that I'm the proudest of as an employee who's been here almost 20 years. I have seen Roche and Genentech consistently make incredibly responsible decisions around pricing. In the U.S., in particular, we have launched many new standard of care changing medicines at prices that are lower than the standards of care that they will be replacing. I think OCREVUS is a tremendous example of this. And because of the distinct pricing discipline that we've had in place over time, our historical net price increases have been below inflation. And so now that we see the IRA beginning to take effect, we expect no impact from those inflationary penalties because we have consistently operated in a very disciplined fashion. So this is one area that I'm particularly proud of and particularly proud of how we have really sought to make sure that the price of our products recommends -- combines the clinical benefit that we bring, the ability for patients and systems to afford it, but also our ability to fund the pipeline of the future, which is really what we're here to do. So now I'm going to get into a little bit more detail about our on-market portfolio. So as Thomas mentioned, I started in March and you may be surprised that when I meet with all of you and all of your investors, the #1 question that you typically ask me is what I'm going to focus on. And this is what I am going to focus on. We talk about ourselves as a leader in immunology. We talk about ourselves as a leader in ophthalmology, historic leader in oncology. I prefer to think of us as a leader in innovation. We are a company that is dedicated to bringing the best possible innovation to patients around the world. And my job is to ensure that we do that as quickly as possible and that we drive as much growth as possible. What we're going to focus on for the rest of the day is in the short, medium and long term how we are going to deliver that growth and where it is going to come from. And I'm going to start -- you weren't lying, Thomas. This thing is actually a little touchy. I'm going to start going through our launch portfolio because I think you all know these products. We've talked about how they've performed in the market. But what I really want to drive home with you today is what opportunities for growth remain with the products that we already have in hand. You have heard Thomas talk today about the fact that we have one of the youngest and broadest portfolios in the industry. These products have a significant amount more growth to give. This is what you saw at half year. You can see that with our top-selling products, VABYSMO, OCREVUS , HEMLIBRA, Evrysdi PHESGO, TECENTRIQ, POLIVY. We added about $2.6 billion in sales just with our strategic products. You heard Thomas say that AH&R are starting to wash out. And we're really starting to begin to hit steady state, which means we can really focus on and continue to drive those strategic products, which are bringing the most value to our portfolio. So clearly, any time you talk about growth, we're going to start with VABYSMO. VABYSMO is on track to deliver $2 billion in sales this year. These are updated market share numbers hot off the press, just got them on Friday, 18% now in the U.S. and AMD, 11% in DME. We continue to see about 1/3 of those patients being naive. We would expect to see even more shifting in the near future. You see greater than 20% market share in the U.K. and Switzerland within just 1 year of launch. We are getting tremendous pickup with VABYSMO. The outlook continues to be very bright. We filed for a third indication in RVO. We expect to get that indication at the end of the year. That will make Roche Genentech the fastest company to deliver all 3 indications in retina. We expect public reimbursement for all EU5 by the end of the year. Currently, globally, we have 75 countries who have approved VABYSMO. We only have reimbursement in '25. So what that tells you is that those curves are going to start growing as reimbursement comes online. You'll hear a lot more from Chris about this in the back half of the afternoon. But I think it's also important to remember that we are not just a VABYSMO company. We are an ophthalmology company. We have the relaunch of SUSVIMO, which we'll talk a little about more in a minute. We have satralizumab in thyroid eye disease as well as a number of other things in the earlier pipeline. We are committed to this space, and our retinal specialists appreciate that we are looking to continue to evolve with them to treat diseases that today we still have significant unmet need. And then again, you'll hear a lot more from Chris about that a little bit later. So why do we believe Vabysmo will be the standard of care? I think it always comes back to those 3 Ds: the dual mechanism of action, the amount of anatomical drying that we see and the clear benefit in drying. And by the way, we have not seen that drying benefit in any of the many trials that have been run with high-dose VEGF, including our own. These are very unique anatomical benefits to VABYSMO, and they are something that retinal specialists look for and appreciate when they treat their patients. In addition to the clinical trial data, which you've clearly seen; the real-world evidence, which continues to mount, new post-hoc analysis continue to demonstrate from an anatomical standpoint over and over that VABYSMO is just performing better than 2-milligram aflibercept. So when you were a kid and you went to grammar in school, your teacher probably told you that words are really important -- the words you pick are really important. The 2 most important words in retina right now are and or because this tells you a lot about why VABYSMO ought to be the standard of care. When we designed our trial, we attempted to design it in exactly the way that retinal specialists treat. We had a series of visual acuity and anatomical criteria and a patient was not allowed to extend treatment if one of those criteria were not met. So it was an or. If you had A and B, but you didn't have this third one, you weren't going to be able to extend. We wanted to make sure that we were gigatons the best possible outcome, and that is very much how retinal specialist treat. And you can see when you do that, that 78% of our patients were able to extend out to Q12 week. When you apply that same very stringent criteria to high-dose aflibercept, you do not see that same result. Only about 58% can extend. If you lose the far less stringent and criteria where you're already starting at that extended dose and you have to have a decrease in both visual acuity and anatomical benefit in order to be required to step down, you see that 96% of VABYSMO patients can actually stay on extended treatment. These data are really telling, and they're really telling because we see it reinforced every single day with physicians as they're treating patients in their clinics. The clinical trial data for the first time ever in ophthalmology is bearing out exactly how we saw it in clinical trials and real-world evidence, both whether that's within real-world trials or whether it's what we're seeing in clinics. And this is why we believe that VABYSMO will be the standard of care. Now SUSVIMO has probably flown a little bit under the radar over the last year. I am happy to say that we are very close to resolving the challenges with VABYSMO. We would expect a commercial relaunch in 2024, ex U.S. by 2025. As you'll remember, this is that little implant that goes into your eye under the eyelid. It allows you to get a refill once or twice a year. This is very compelling for patients who actually don't live particularly close to a retinal specialist or for health care systems where access to retinal specialists is limited just because you don't have many. There is a high level of patient preference for SUSVIMO, and we are very eager to actually bring it back to market. We expect clinical trials to restart later this year, as we've said. And I think it's also important to just call out in February of this year we did actually see positive DMD and DR data, which were presented in angiogenesis. So SUSVIMO, again, we still believe is going to be a very important option for patients and it will be back on the market and back in clinical development quite soon. Let's move on to neurology and OCREVUS. So many of you may not be that aware that when a new molecule actually launches in MS, it's quite typical that for the first 4 or 5 years they grow and then they start to plateau. OCREVUS is 6 years on the market and still growing at double digits. It continues to redefine the standard of care. It continues to have data that no one else can touch, and we continue to be the leading high-efficacy therapy out there. The outlook for OCREVUS continues to be incredibly strong. We have the Phase III OCARINA data, which we'll talk about in a minute. We have some really compelling 10-year safety and efficacy data that will be presented at ECTRIMS. MS is a chronic disease. People will be treated for a very long time. What I think is great about these data is that we always have several years of clinical trial that's ahead of where patients are in the market. So when you think about a physician who's making a decision to put a patient on and keep a patient on, they can always look ahead within the clinical trial data and make sure that the safety and efficacy are going to remain compelling for that patient. We also intend to submit our pregnancy and lactation data to health authorities next year. We will be the only anti-CD20 that has this data. Think about your average MS patient. They're relatively young women, family planning is important to be able to have this data. And the label is going to be extremely helpful. And then, of course, we have the high-dose OCREVUS trials, which are now fully recruited and we would expect to be reading out that data shortly. I'm sure Paulo will talk a little bit more about this, this afternoon. But again, this is just another opportunity to make an even further dent in delaying progression. I also want to point out that fenebrutinib, which is our oral BT,K, now has about 2,500 patients who have been on therapy. Important to note that orals represent about 25 -- about 40% of the global market in MS. The Phase II data for fenebrutinib was quite compelling. This again has the opportunity to be another significant growth driver for us going forward. And when we talk about where is that growth going to come from, from OCREVUS, clearly there is still room for us to grow in IV. But the subcut and the positive Phase III data that we saw with OCARINA and the ability to now bring a subcut formulation to patients is going to be extremely important. OCREVUS subcut has the potential to enable at-home administration by a health care professional. And we go from between 4 and 6 hours for an infusion down to 10 minutes twice a year. Important to note that MS patients will typically go to the neurologist twice a year anyway. So we're not adding visits. You go for your biannual check up, you get your OCREVUS, 10 minutes you go home. So we're very, very excited about bringing this opportunity to patients. High-efficacy therapy is what every MS patient deserves. We believe this is another really important step forward for OCREVUS. When you see on the slide that we estimate that we have significant opportunity remaining in academic hospitals, referral hospitals and particularly community neurologists. Regardless of where patients are getting treated, there is still a significant number of patients who are not getting high-efficacy therapy. And if they are going to go on a high-efficacy therapy, the data would suggest that they should be going on OCREVUS. That is where we will be focusing our efforts. Just want to go forward, I'm going backwards. Can we go forward? It's not moving? Can we go forward one? It's not going? There we go, Evrysdi. Evrysdi is well on its way to being a market leader in SMA. We have well over 11,000 patients treated. Again, important to note here that the vast majority of patients who have SMA are adults who have never been on any disease-modifying treatment for their disease. And so this is a tremendous untapped market for Evrysdi. We're starting to see pickup in these adult untreated patients, particularly in the U.S. We still have a number of countries that are still coming online with reimbursement. We continue to expect market share gains for Evrysdi, and we continue to look for ways to drive growth as reimbursement comes online. The further potential to grow our SMA franchise, and Paulo will talk about this in a little bit. It's actually a combination with an antilatent myostatin. So again, having babies who are on SMA and also partnered with the Chugai molecule, GYM329, to potentially regrow muscle could be an incredibly powerful and potent solution for these patients. So more to come on that. Switching over to oncology. So TECENTRIQ is now the first PD-1/PD-L1 with a subcutaneous formulation approved by the MHRA in the U.K. just a couple of weeks ago. I'm sure that you have all heard that we have a slight delay with our FDA filing. This is a process issue. It's nothing to do with the safety or efficacy package. We will resubmit later this year. We expect approval in 2024. It should have no impact on any of our other filings globally. But I think what we see with TECENTRIQ subcu, very similar to what we saw from PHESGO, is the opportunity to actually bring a significant process savings to health care systems that are otherwise quite strapped. And so we expect that this is going to be a very popular option for many health care systems. You can see it go from a 30-minute infusion down to a 7-minute subcu, it's a 77% reduction. Now that we have more things like PHESGO in the market, physicians and hospital systems are far more used to doing -- or to seeing options like this. And so we would expect in those markets to see relatively rapid pickup. This is going to be extremely exciting. TECENTRIQ in sort of other news, we're reaching peak share in those first-in-class indication, small cell, HCC, where we are the undisputed market leader. When it comes to adjuvant non-small cell, we continue to have the first-mover advantage in adjuvant. And we continue to benefit from the fact that our data is understandable and straightforward, and we actually have quite a clean label. We are continuing to see utilization grow there, particularly as new countries come online and reimbursement is adopted. The outlook for TECENTRIQ over the rest of the year. We expect the CHMP opinion on TECENTRIQ by Q4. We expect to see the head and neck adjuvant data this year as well. I'm sure Levi will talk in a little bit about SKYSCRAPER-01, that's an event-driven trial. It looks like it will now read out in Q1. We've initiated the first-line HCC trial with TECENTRIQ and tiragolumab. And then the adjuvant HCC Phase III IMbrave met its primary endpoint in RFS and then that OS was immature. So still lots of good opportunity available for TECENTRIQ. And we are -- as I said, you'll get a much deeper overview and update on TIGIT from Levi in a moment. Okay. Moving on to breast cancer. So I want to orient you a minute to this slide. The bar in the middle is actually all of the different kinds of breast cancer. And you'll see that HER2-positive [ risks ] actually represents a fairly small portion of patients living with breast cancer. And so while Genentech and Roche have had a legacy of leadership in HER2-positive breast cancer, for us, the question isn't can we cure women with HER2-positive breast cancer. The question is can we cure women with breast cancer. And 70% to 80% of women out there who suffer from breast cancer actually do not have a HER2-positive tumor. And so what we will continue to serve is the standard of care in many different respects with the portfolio that originated from our company, we are also actively moving on to try and see how we can help the larger percentage of women out there who currently don't have acceptable treatment options. Giredestrant is, as all of you know, one of my favorite molecules. I'm very interested to see how the adjuvant data here plays out. It has the potential to redefine the backbone of care for hormone receptor-positive breast cancer, which is by far the largest segment of the market. And involalisib, our PI3 kinase similarly has the opportunity to really revolutionize the treatment of breast cancer. So again, Levi will talk to you more about both of these in a little bit. But by no means are we walking away from breast cancer. In fact, our aspirations in breast cancer have only increased. And they have increased because we want to help more women who actually currently don't have options that might help them live longer. But we are in no way, shape or form walking away from HER2-positive. PHESGO continues to drive increased penetration where we continue on our trajectory of converting people, particularly in early launch countries. We continue to hear really great things about PHESGO from all of our patients who are on it. And not only are we seeing people convert to PHESGO utilization, we're also seeing in places where PHESGO utilization is growing that we are actually growing the early breast cancer market as well. And so PHESGO, I think, is just a really great example of we're bringing a very needed innovation into the health care system with a standard of care-changing medicine can really be quite a development. And we're very, very pleased with how PHESGO has grown and will continue to go in the innovation that it represents for patients. Charlie is going to talk all about our malignant hematology portfolio. This is an incredibly vast market. You can see, again, with the bar chart, these are all of the different kinds of hematologic tumor cases by subtype. And you can see with our current end market portfolio that we actually cover either with approved indications or with trials quite a number of those. And then as you get into our pipeline molecules, we're expanding even further and into places where we haven't been to date. I'm not going to talk too much about the clinical development plans because Charlie is going to cover a lot of that in the not-too-distant future. But we will take a little bit of a deeper dive into Columvi and Lunsumio since those have both just recently gotten their first approval. But let's start with POLIVY. POLIVY is truly revolutionizing the standard of care in first-line DLBCL. This is the first combination in 20 years that has shown an improvement over Rituxan, and we love raising the bar on ourselves. That's just a really great thing to be able to do. We are now approved in over 80 countries. We have incredibly strong uptake. And I think what's actually most exciting about POLIVY is what we're hearing from the environment, is that physicians are no longer excited to enroll their patients in trials where R-CHOP is the control arm. And we are getting an increasing amount of requests for POLARIX, particularly for Phase I/II trials that are starting to kick off now. And I think what this tells you is that our competitors, the physicians are all starting to lean towards POLARIX as the regimen that ought to be the standard of care. And consistently, what we hear is it gives my patients a better chance for survival. It gives them a better chance for cure, why would I not use POLIVY. And I think this is just a really great example of taking on a really tough challenge, bringing a new innovation to the market and really being able to very appreciably help patients. And this is certainly what we intend to do -- continue to do in CLL, DLBCL and in follicular lymphoma. You can see the evolution of standard of care from chemo to Rituxan plus chemo, GAZYVA plus chemo and now the new MOA is really coming in. This is obviously an area where Roche has tremendous legacy. I want to talk to you a little bit about POLIVY and Lunsumio, given that they both have just received their approvals in third-line plus DLBCL and third-line plus follicular lymphoma, respectively. Talking specifically about Columvi. This drug is very early days in launch, but what the immediate feedback is from the field is that those deep, durable responses that we see with Columvi, combined with a very simple dosing regimen and a fixed duration of treatment, are an incredibly compelling package. We are very interested to see how this drug actually continues to expand in third-line DLBCL. And then as we continue to move it up the treatment algorithm in DLBCL and into other hematologic diseases, we think -- I think personally, this is a drug that you really should be watching. This one seems quite undervalued relative to the competition, considering the profile that it brings. Similarly, Lunsumio is being very well received by community physicians. Community physicians, particularly in the U.S., represent a very significant portion of treaters. They don't particularly appreciate hospitalization. Something like Lunsumio in third-line follicular that keeps their patients out of the hospital is a very compelling value proposition. And again, we would expect that Lunsumio and Columvi will be able to find a very complementary place in hematology and we'll be able to really serve very different patient populations quite well. No conversation in hematology is ever complete without a discussion about HEMLIBRA. HEMLIBRA has redefined the standard of care in hemophilia A, continues to be the market leader. It continues to grow very significantly even as new therapies start to come into the market. We have more than 21,000 patients have been treated globally. 2/3 of those patients are already on every month or every other week treatment. And I think what we hear from patients is that they are just extremely happy with their HEMLIBRA treatment. One of the things that I think has kind of come up in the environment is what's happening with some of the new extended factor treatments that are out there today. One thing to keep in mind is that by guidelines, hemophilia patients are encouraged to have a factor on hand in case of a breakthrough bleed. So it's just something that they encourage anyone who has hemophilia to do. It would not be unreasonable to suggest that patients who are replacing their backup factor might be replacing their backup factor with a new factor. So I would encourage you to take some of the news that's kind of coming out about switches, potentially with a smidge of a grain of salt because I think we don't have the data to suggest that, that's true. The guidelines would suggest a very rational explanation for what's happening. And I think there is no reason to believe that patients would switch from HEMLIBRA to a factor product at this point. We continue to believe that we've set the standard of care patients who are on prophylactic treatment will remain on prophylactic treatment. And if you're going to do that, you're going to remain on HEMLIBRA. We continue to focus on opportunities in non-inhibitors. I think when we talk about growth, this is an area where our own very ambitious internal models had us peaking at a certain place and we're blowing past that peak in many of our accounts in many countries in the non-inhibitor population. As physicians get comfortable, as patients get comfortable, you just see more and more increased use. We're driving growth in key accounts in many of our key countries, and we continue to increase international access. We are not stopping there, however, in hemophilia A. You know that through Spark, we have a gene therapy program, which is about to kick off. And then the NXT-007 bispecific program, which is Chugai's follow-on molecule continues, to generate data. And finally, ALECENSA. I covered ALECENSA when I was in the U.S. so I have a very soft spot in my heart for ALECENSA. This is a great drug, and it has just released some -- it has just announced some extremely compelling data in adjuvant. That data will be presented at an upcoming congress. But I think what is really heartening here is that we've seen ALK-positive patients have a real chance to live a very different kind of life with ALECENSA. And with this adjuvant data, we believe we're going to take another step change in what's going to be possible for these patients. So please look out for that data, which we do hope will be presented shortly. The data were really quite stunning, and I'm sure you're going to like what you see. So this is our typical news flow slide. So I'm only going to hit the things that have changed since half year. We talked about the TECENTRIQ subcut delay in the U.S. We talked about the positive ALINA trial. I want to call your attention down to the bottom here, these last 2 bullets. ELEVIDYS received its U.S. accelerated approval in DMD for boys who are 4 to 5. This allows us to file in countries that reference the FDA. We have gotten our first approval in UAE and hope to be able to start bringing the promise of ELEVIDYS to boys in that country soon. And then the involalisib trial has actually been pulled forward from 2024 into 2023. So we'll have some new data by the end of the year, and I think Charlie or Levi will cover that in a little bit more detail. And finally, we have some significant growth opportunities ahead. Everything that I just covered are really drugs that are currently already on the market. We have a significant amount in our mid- and long-term pipeline that have the opportunity to be standard of care-changing molecules that can continue to drive very significant growth. I want to talk about only one of these before I pass it off to Levi and that is our newest entrant to the pipeline, which is zilebesiran. So you heard us talk at half year about the partnership that we signed with Alnylam on zilebesiran. We believe that this very novel siRNA targeting agent has the opportunity, again, to be quite transformative for patients living with uncontrolled hypertension. There are tens of millions of patients around the world who cannot control their hypertension with standard of -- the current standard of care. We know that these patients are at risk for worse cardiovascular outcomes. We also know that the cost of controlling those cardiovascular outcomes is one of the major burdens on health care systems today. The opportunity that we may be able to control that in as little as 2 doses of zilebesiran a year is truly transformative potentially for these patients. You saw that the Phase II KARDIA-1/2 program is already ongoing. You saw the KARDIA-1 results were press released last week by Alnylam. We would expect those to be presented at an upcoming scientific conference and KARDIA-2 is expected early next year. But this molecule really has the chance to completely revolutionize how we think about uncontrolled hypertension and potentially to really improve cardiovascular outcomes for patients. So on the innovation side, it fits very squarely with what we, as Roche, are all about. But it also fits very squarely with our strengths in commercialization, our track record of entering new therapeutic areas, our expertise in getting the right patients into trials, the ability to work with our diagnostics colleagues. This partnership actually is right in our sweet spot of where we believe we can really make a difference for patients and do it in a way that is more effective and efficient than would have historically been done in the past. And before I hand it off, just again, to put a double underline on where this growth in the future is coming from. This is just a summary of our key late-stage assets that have CHF 1 billion in sales potential or more. The drugs that are on the top half of this list, we have 8 late-stage NMEs that have the potential to generate more than $2 billion worth of sales. We have another 4 that will clearly be blockbuster status. And if you look at our in-market portfolio, we have 2 marketed products with upcoming line extensions, so just a line extension that could add an additional CHF 2 billion. And we have 4 in-market products that could have a line extension that it adds another CHF 1 billion in revenue. There is a lot of opportunity for growth in our pipeline. We have to deliver it, we have to get it approved, we have to get it reimbursed and we have to get it to patients. But the reality is we continue to have a significant opportunity to change the course of treatment for diseases that we are in today and diseases that we will be in tomorrow. And I think we are very excited about the opportunity to think very differently about how we're bringing our medicines to market and the opportunity that we have to drive growth across the entire portfolio. And with that, I will hand it over to Dr. Garraway.

Thank you, Teresa, and hi, everybody. It's always a pleasure to speak at Pharma Day. It's great to be back after a year. I also have the distinction of speaking right before lunch and we're running a little behind schedule, very un-Swiss of us, but I'll try to keep us on track. So I run, as mentioned earlier, the product development organization at Roche and actually my colleagues who will speak after lunch are also part of the product leaders in the product development organization. And I've been running the organization now for about 4 years, and it's actually interesting to reflect over the past several years, roughly the same interval, the field change in therapeutics -- several therapeutic areas has been quite dramatic, in some cases, unprecedented actually, in some cases. And those broader changes together with our aspiration and expectation of our own productivity were a key driver of the end-to-end review that Thomas described on R&D excellence. And as Thomas mentioned, on the one hand, that review has affirmed several strengths. And actually, my job and my colleagues' job will be to tell you about several pipeline strengths that we see, some of which, as Thomas mentioned, maybe are perhaps underappreciated by some in the investor community. So that's one thing. But of course, there are also some areas of improvement that we identified, whether it's setting and enforcing the bar around our investments, what we dial up, what we stop, our -- particularly approaching how we improve our Phase III success rates, expanding external innovation across the R&D continuum and then, of course, opportunities to become more cost effective in certain areas. But all of that is geared toward creating an enterprise in an environment where we can recognize and capture and deliver transformative innovation wherever it might be found, whether that's in disease areas where we are very active, or as Teresa just mentioned, in cases where we haven't been as active, such as our collaboration with Alnylam in the cardiovascular space. So I'm going to begin by starting with an area that we have been very familiar with and have been leaders in for a long time, which is in oncology. I'll talk about our -- the solid tumor side of the pipeline. And then Charlie after lunch will talk about the hematologic malignancy side of the pipeline. But one approach that we have taken over the years in oncology is to identify pillars, we call strategic pillars, where there remains substantial unmet need and where innovations, for example, from our pipeline might be able to bring meaningful impact. And so there are 5 such pillars on this slide. Starting on the left side here, precision medicine or precision oncology. They're delivering targeted therapies to genetically or molecularly defined tumor subtypes. There continues to be substantial unmet need there, including even in some of the best-known examples of genetically defined tumor subtypes. Early disease, of course, moving our innovation into the curative setting and boosting durable control of cure rates. Novel immunotherapies that go beyond, for example, immune checkpoint blockade. Rational combinations. Here is where our broad pipeline can be a particular strength because it gives the opportunity to build differentiated regimens, not just differentiated individual medicine. And then finally, new modalities. We've seen a number of examples of how new modalities can really disrupt areas and bring exciting advances. And so that certainly continues to be an area of emphasis for us. So at a high level, this is sort of the cut of our solid tumor portion of our pipeline. What you can see here is that there are 8 assets that are in Phase III studies or pivotal studies of various types of indications or line extensions. And then when you go a little earlier in development, we have upwards of 30 additional assets that are being tested in various ways. And so when you look at this all together, it really underscores what has been true for a long time, which is that we continue to have one of the largest and most diverse oncology pipelines in the industry. And as Teresa and also Thomas have sort of shown, the examples that I'm going to hone in on are particularly solid tumor examples that provide potential midterm opportunities. And so with those examples, it will also be clear, hopefully, how we're trying to identify opportunities in these strategic pillars where we can add substantial value. So the first example will be tiragolumab. And so tiragolumab, as you -- most of you know, is our anti-TIGIT monoclonal antibody. TIGIT, of course, is an immune checkpoint like PD-1, like CTLA-4. And for a while now, there's been a robust preclinical rationale, scientific rationale, for targeting TIGIT in conjunction with other immunotherapy. And because of that and because of some of the emerging clinical data, we have invested in quite a large clinical development program to look at combining tiragolumab with TECENTRIQ in several indications. And so the clinical picture that we have now seen, the aggregate picture really convinces us that there is activity with tiragolumab. Next slide, please. So there are a couple of lines of evidence that I'll just highlight here. One of them is the SKYSCRAPER-01 study, which is our study combining tiragolumab with TECENTRIQ in PD-L1 high non-small cell lung cancer. We have pointed out for a while to this community that we saw numeric increases in both progression-free survival and overall survival in that setting. Inadvertently, the second interim of that data was disclosed a short while ago, but that confirmed that from an overall survival standpoint, it looks like tiragolumab could be adding more than 6 months of benefit to -- on top of Tecentriq in that setting. And so we, of course, are continuing to follow those patients as planned. There will be a final analysis once we -- it's event-driven, so once we achieve the event. And -- but certainly, the study remains blinded, of course, to physicians and patients. But together, that is one clear line of evidence that we're seeing activity with tiragolumab. The other line of evidence that we recently described an ASCO was -- can we go back one slide, please? Yes. The other line of evidence was presented in the setting of hepatocellular carcinoma. Now as you know, in hepatocellular carcinoma, the standard of care -- a standard of care is the combination of Tecentriq and Avastin. But we did a -- we presented Phase I and Phase II data showing the addition of tiragolumab, so basically triplet on top of that doublet. And what we saw was an increase in both overall response rates upwards of 43% with the triplet combination and then also an increase in progression-free survival. Now of course, these Phase I/II data are smaller patient numbers, et cetera. So it was helpful also to overlay or to benchmark, I should say, those data on what we've seen in other context. So for example, the IMbrave150 data from the pivotal study that established Tecentriq and Avastin standard of care. And when you do that, what you see sort of this dark gray line between the red and blue curves, shows that the potential impact of adding tiragolumab still holds. And so when you look at all this data together with hepatocellular carcinoma and also other context that provided a justification for starting a pivotal study in our minds in the setting of events, hepatocellular carcinoma. And so that is just getting underway. But altogether, these data certainly support the notion that tiragolumab could emerge as an important novel immunotherapy in our solid tumor arsenal. Okay. So the second example I'll describe is giredestrant. Teresa has already set that up very nicely by pointing out that estrogen receptor positive breast cancer. First of all, it's about 70% of all breast cancers, and it's an area that we have not previously been in. But what's also important from a biological standpoint, is that when you are -- when the breast cancer is estrogen receptor positive, estrogen receptor is an oncogene driver. And of course, it's been the target of many therapies in breast cancer, but we believe that our SERD giredestrant has best-in-class properties. So this is why when Teresa talks about it has the potential to become a new backbone of therapy, the best-in-class characteristics that we think we are seeing support that. And so some of those are preclinical, and we presented them in the past IR events like this, so I won't go through them in any detail, but suffice it to say, it's about increased potency amongst all the SERDs that have been in development. It's about increased combinability, including a full therapeutic dose with other full therapeutic dose assets that are relevant in breast cancer, and it's about the safety profile that looks very reasonable. So because of those preclinical data as well as the emerging clinical data, we have invested now across several important areas, context in breast cancer, which include the persevERA study, which is our frontline metastatic study of the combination of giredestrant plus palbociclib compared to aromatase inhibitor plus palbociclib. And then -- in the middle here, we also have our adjuvant study. So this is the lidERA Phase III trial where we're looking at head-to-head comparing giredestrant to physician's choice, which is usually aromatase inhibitor. And so this is a study where, number one, we're well ahead of competitor surge but also it's an opportunity really to look at the initiation of giredestrant from the very beginning in the adjuvant setting. Now more recently, we've also added 2 additional components to this program, one of which is a single arm study that's part of lidERA, which is looking at how giredestrant does in combination with abemaciclib in the adjuvant setting. And the second is a study that's called pioneer, which is looking at the combination of giredestrant plus a CDK inhibitor of choice compared to fulvestrant plus a CDK inhibitor of choice. So really, these data will add to the full spectrum of context in which you might want to use an endocrine-directed therapy backbone. Okay. So the next slide, shows sort of a composite of clinical data that justifies this investment sort of across the range of ER-positive breast cancer. And there are 2 takeaways that I'll just give you at a high level of this clinical data. The first is that when estrogen receptor signaling is positive or active, giredestrant is active. So that's the first takeaway. And the second takeaway is when giredestrant has been compared head-to-head against potential standard of care ER-directed therapy, it has looked superior. So -- and now I'll go through some of the elements here that unpack that a little bit. So here on the left is some data that's -- that we presented more recently and maybe you haven't seen before. But basically, it's sort of a -- this is looking at pretreatment specimens from various tumor settings and is looking at the gene expression signature of estrogen receptor responsive genes, it basically gives you a readout, a composite readout of estrogen receptor activity in those settings. And so you see -- shown on this graph, violin plots, looking at a Z-score of ER activity. And what you can see is in the early breast cancer setting, or in the pre first-line metastatic setting, you have a distribution of the violin plots, which is sort of right around 0, Z-score of 0 basically tells you that you're basically right at the mean of what you would expect if estrogen receptor was active. So that all makes sense. Now -- but when you look at the sort of teal violin plot, where you see it's a shift downward. This is the setting of -- this is the post treatment setting, metastatic setting, where -- but it's a case where there are -- the sequence of the estrogen receptor gene is wild type. And so as you would expect, when patients who have been previously treated with ER-directed therapy in earlier lines now progress to later metastatic disease, they are often -- number one, they're clinically resistant to ER-directed therapy. But they also -- what this shows is that basically the estrogen receptor signaling has gone down. So basically, the tumors are indifferent. They don't care about the estrogen receptor anymore. However, when you look at the subset of patients who actually have activating mutations, in ESR1, the gene for estrogen receptor. Now the violin plot is sort of back with what you expect. So basically, you've got 3 indications where you would expect -- number one, ER signaling is active, so you would expect giredestrant to be active, but you have one where you would not expect to be active. But the bottom line is now when you look at the rest of this clinical data, that you're basically seeing giredestrant active in the settings where you'd expect. So on the far right, is a subset of the acelERA study, which overall did not meet its primary endpoint, largely because it was -- it had a substantial percentage of patients that were wild type in later stage. So we now understand why they didn't respond. However, when you look at the subset that had ESR1 mutations, you see not only that giredestrant is active, you see that it is actually looking better than the physician's choice of endocrine therapy, which the majority of the time was fulvestrant. So that's now an encouraging line of evidence on the later stage. And then the middle panel here, is our data from the neoadjuvant setting, which is looking -- it was a neoadjuvant window of opportunity study, which looked at the addition -- either giredestrant head-to-head comparison with anastrazole at the time of surgery or giredestrant plus palbociclib together compared to anastrazole plus palbociclib. And in both cases, what you see is the ability of giredestrant to substantially suppress Ki67 levels compared to the -- in the head-to-head setting. So basically, in that setting, you're seeing evidence that giredestrant can beat an aromatase inhibitor in early-stage breast cancer. So again, these data together provide evidence not only that giredestrant is active, but it's leaning in the direction of showing superiority compared to standard of care endocrine therapy in a variety of settings. So this is sort of the basis for our confidence in the possibility that giredestrant could be a new backbone therapy in the future. Okay. So the third example that I'll show is inavolisib. Inavolisib is our PI3 kinase inhibitor. And so here again, there is a marketed PI3 kinase inhibitor. But when you look at the preclinical data package of inavolisib, it also shows best-in-class potential. When you look at the potency compared to alpelisib which is the competitor, 58-fold greater potency. When you look at the selectivity against other PI3 kinase isoforms, which, of course, can be a basis for toxicity, greater select between 6 and 32-fold greater selectivity than alpelisib for the other PI3-kinase isoforms. Now in the past, previous Pharma days, we've shown the early clinical data that indicated that full dose inavolisib could be given in combination with full dose CDK inhibitor and full dose fulvestrant. So that's not new, but it's important to remember. But what is new is that the initial pivotal trial of inavolisib, which is our INAVO120 study actually is going to readout sooner than we thought. Actually, we expect a readout in Q4 of this year. So that's -- that, of course, will be of great interest. And then we've also expanded our program, for example, to add a study of inavolisib in conjunction with Phesgo, which Teresa talked a lot about in the subset of HER2-positive breast cancer that also happens to be positive for PIK3CA mutation, which is a reasonable subset of HER2-positive breast cancer. So a lot is happening, inflection point coming for the inavolisib program quite soon. So while we're on the general topic of breast cancer, I'll briefly give you a teaser of a recent addition to our pipeline from -- through the external innovation avenue, which is ZN-1041. This came from an in-licensing from Zion therapeutics. This is a small molecule HER2 tyrosine kinase inhibitor that again has best-in-class characteristics from a brain penetrant standpoint. So you can see on the table here on the left that compared to the other HER2 TKIs ZN-1041 has quite robust brain penetrants as assessed based on CSF studies versus plasma studies. And one potential reason for that is that ZN-1041 is not a substrate of the P-glycoprotein transporter that is known to produce efflux of these kinds of compounds out of the CNS. And so this is sort of the preclinical package. And now when you look clinically at what we're seeing, the waterfall plots on the right here are looking at ZN-1041 in combination with trastuzumab and capecitabine. And you can see that whether or not you're looking at overall tumor size, you get encouraging data, but in particular of interest to us is the effect in the brain. The CNS tumor size shrinking, sometimes complete responses seen in the brain. And so these data would suggest that the ZN-1041 could be important not just to treat brain mets, but importantly, to prevent brain mets. And many of you know that HER2-positive breast cancer, this is one of the feared complications, particularly late recurrent in the brain, unfortunately, all too prevalent in the setting of late management of oncology patients so this could be an important therapeutic opportunity in this setting. Okay. So the fourth main example that I wanted to describe is divarasib. This is our KRAS G12C inhibitor. Again, a situation where there are 2 marketed competitors, and yet we're all aware that there is plenty of unmet need even in this particular mutant setting. And so it's intriguing to us that, again, best-in-class properties preclinically. The data here kind of small, but what we're showing is potency of divarasib compared to the 2 other well-known KRAS G12C competitors. So that's preclinical data. And on the right, we're showing data that was recently published in the New England Journal of Medicine of the clinical picture that we've seen with divarasib so far. And I'll just highlight that in the second-line non-small cell lung cancer setting, of course, these are the subset of non-small cell lung cancer that have KRAS G12C mutations. What you're seeing at the 400-milligram dose is a 56% overall response rate in the colorectal cancer therapy setting, with KRAS G12C mutations, you've got a 36% overall response rate at the 400-milligram dose. Again, there's always caveats of trial comparisons. But nonetheless, these data kind of, at a high level, stack up very well with what you would see from competitors in analogous kind of clinical settings. Similarly, the progression-free survival of greater than 13 months is very encouraging. So overall, these are data that justify additional studies that we now have ongoing of divarasib for example, in combination with pembrolizumab and also in various combinations in the colorectal cancer setting. Okay. So having shown those specific examples, I'll now just kind of briefly discuss some emerging areas where we're also becoming more active. One of those emerging areas is DNA Damage Response. And of course, as you all know, PARP inhibitors have sort of been the exemplar success of targeting the DNA Damage Response pathway or one component of it. But we have been sort of exploring the next generation of DDR therapeutic opportunities. And so there are 2 examples of how we've done that. The first is a collaboration or I should say, in-licensing of an ATR inhibitor from Repare Therapeutics called camonsertib. So ATR is a key component of the DDR pathway. And there was certainly a very strong preclinical data package, but importantly, the ongoing clinical data is showing monotherapy activity of camonsertib in several tumor contexts. And so we're now in the process of stepping up combinations of camonsertib with PARP inhibitors to have potentially that as a novel regimen, a combination that could be taken into the setting where we would expect efficacy, the tumor types where we would expect efficacy. So that's one example. We actually discussed that last year. But a newer example also in the DNA Damage Response space is our recent collaboration to bring in KSQ-4279. So this is a USP1 inhibitor. USP1 controls ubiquitination and therefore, activation of several proteins, particularly several DDR proteins, but these are distinct DDR proteins than those that are targeted by some of these other mechanisms. And so here again, this is an intriguing first-in-class opportunity in the DDR space. And so this -- we are now involved in -- we have Phase I trials ongoing across a range of solid tumors. So together, this is an example of building an arsenal to take shots on goal in the DDR space in a way that could be interesting going forward. And the last vignette, I'll mention before closing is the neoantigen vaccine space. And so many of you are aware that the logic, the scientific rationale for expanding the immune -- particularly the T cell activity around tumor-associated neoantigens has been a very sound one for many years. But it really wasn't until over the past year or so that we started to see clinical data emerge that justify the scientific enthusiasm. And of course, one example of that clinical data has been the data from Merck and Moderna in the adjuvant melanoma setting. But we also have data in a setting of surgically resected pancreatic cancer, a much smaller study, but nonetheless, very intriguing because what we saw in that study, which is shown in the panel here on the right, is that in patients where there was a T-cell response to the personalized neoantigens that were enacted via the vaccine, what was seen as a much greater relapse-free survival than patients who did not have such a response. So based on these results, this is a collaboration with BioNTech to be able to personalize neoantigen vaccine. So based on these results, we are now pursuing a randomized study in pancreatic cancer, and we also expect to initiate randomized studies in other indications that we'll be able to disclose a bit later on. But overall, this is another very intriguing area that could be -- that could hold great promise in the early disease setting in the future. So my last slide before I close just comes back to that fifth pillar, which is about advancing new modalities in oncology. And here, the -- this has been an area of interest for us for a while. And as you've heard briefly from Teresa and you will hear a lot more from Charlie, in the area of bispecific antibodies, we've brought inroads in blood cancers and also non-oncology areas, there are some trispecific opportunities that we're now exploring, which will be fun to keep an eye on. There are degraders of the androgen receptor. We just brought an AR degrader into our pipeline last year. And then one other modality, I'll just call out is our colleagues at Chugai have now really -- have really made major inroads in the cyclic peptide, particularly in their ability to create membrane permeable -- sorry, I should say, membrane-penetrant in orally bioavailable cyclic peptides. So this is quite interesting. The first example of this is a molecule called LUNA18, which is a pan-RAS inhibitor. It's now in the clinic. So we look forward to seeing those data emerge which we think could be a proof of concept really if we see what we expect to see with the pan-RAS inhibitor, this could really be a proof of concept for the entire platform, and this is of great interest because there are a number of cyclic peptides behind LUNA18 that are going after a variety of difficult-to-drug targets. And so this is something that we will look forward to updating you on a regular basis but it's just another example of how the breadth of modalities can give new opportunities either to target -- well-known targets in new ways or to go after targets that we really could never go after [indiscernible]. So altogether, we think that this -- this is an example in the solid tumor oncology space of some highlights or some strength that we want to keep on building on. And then after lunch, we'll move into heme malignancy and other areas. So thank you for your attention, and we'll see you after lunch.

Maybe just as a quick update, we're a bit over time, 10 minutes. So I would recommend we push out lunch a bit and go until 1:00 and we then -- at 1:00 for the other late-stage pipeline updates. Thanks. [Break]

Well, welcome back. I hope you all enjoyed what was a terrific lunch. If you don't know me, I'm Charlie Fuchs and I'm the Global Lead for Hematology and Oncology Product Development. And actually, just earlier, somebody mentioned to me that Charlie just be aware, after a large lunch the energy level tends to diminish precipitously. So you have your work cut out for you. So I hope that I can maintain really what was an exceptional level of energy this morning. Levi really did a terrific job covering the solid tumor side of our portfolio. What I'd like to share with you is some highlights from both the malignant hematology portfolio as well as our work in non-malignant indications. So as you know, Roche and Genentech have a rich legacy in transforming the therapeutic landscape of B cell malignancies since really the initiation of Rituxan, which transformed that component of oncology. But we continue to establish new standards of care, including Hemlibra and hemophilia A. Polivy R-CHP, as Teresa mentioned, in the first-line setting of diffuse large B-cell lymphoma and Lunsumio and Columvi now in follicular and large cell lymphoma, respectively. As we're discussing today, we continue our investments in our established indications of lymphoma, CLL and hemophilia A while expanding into new important areas, namely AML and myelodysplastic syndromes, multiple myeloma, mantle cell lymphoma as well as other nonmalignant hematology indications. As you can see, paroxysmal nocturnal hemoglobinuria, sickle cell disease and atypical hemolytic uremic syndrome. We're leveraging our broad portfolio of molecules to develop novel combinations as well as look at different modalities, most notably the next generation of bispecifics that I'll talk about as well as next-generation gene therapy and allogeneic CAR-T cell. No less importantly, we're committed to improving the patient experience using fixed-duration therapies, simpler dosing schedules, off-the-shelf availability and in the means of preventing hospitalization outpatient administration of our therapies. At Roche, we're privileged to develop one of the broadest and richest portfolios in malignant and nonmalignant hematology and beyond our pioneering work in antibody drug conjugates and bispecific antibodies we're developing molecules across the full spectrum of clinical investigation, as you can see, including novel therapies from malignant conditions, next-generation approaches for hemophilia as well as a number of unmet needs that I'll describe. Well, I won't cover all of the agents in those previous slides during my presentation, and I'll focus principally on our work in B-cell malignancies in PNH and hemophilia A. But of course, during the Q&A, I'm happy to discuss any of the programs that I didn't cover in my 15 minutes. Malignant and nonmalignant hematology represent an integral part of the building blocks of the future growth of Roche and beyond the agents that are listed here, there's a number of, I think, a number of other important programs, namely the bispecifics in myeloma, fusion proteins and non-Hodgkin's lymphoma as well as a number of off-the-shelf programs in CAR-T therapies, and I'll cover some of them during my time. I'll start, of course, with our investment in malignant hematology in my TAM focus on the advances we're now taking to further improve the therapeutic landscape in B-cell malignancies. So Teresa mentioned that beyond the historic development of Rituxan for the first-line treatment of diffuse large B-cell lymphoma. The POLARIX regimen demonstrated the importance of putting the CD79b antibody drug conjugate Polivy into the standard first-line therapy in the form of Polivy R-CHP which showed a clinically meaningful superiority over R-CHP. The regimen, as you heard, is now approved in 80 countries earlier this year. We had a successful FDA approval following our ODAC and moreover, we have Category 1 designation by NCCN as well as other numerous Asian and European guideline committees. In addition, we continue to leverage the POLARIX database and as you can see, this is an analysis we reported recently at ASCO looking at the growing population of elderly patients with diffuse large B-cell lymphoma. And as you see from the data, individual -- among the patients 70 years or older. We saw a statistically significant 36% improvement in progression-free survival as well as a trend towards a 26% improvement in overall survival, clinically meaningful with a safety profile in the elderly patients that largely matched that for the younger patients enrolled within POLARIX. Turning now to our CD20/CD3 bispecifics with the benefit of both Lunsumio and Columvi in our portfolio, we have -- we're uniquely positioned to provide CD20/CD3 therapies tailored to the disease, the patient, the physician and the health care system. For patients with more indolent follicular lymphoma, Lunsumio does not require hospitalization, therefore, supporting routine use in the outpatient setting for a disease where ease of administration is really an expectation. At the same time, Columvi provides potential best-in-class therapy with off-the-shelf efficacy that is comparable to CAR-T for patients with diffuse large B-cell lymphoma. Leveraging the full impact of these molecules in B-cell malignancies, we have a bold portfolio, which is included, as you can see, the combination of Lunsumio on lenalidomide in the second-line setting of follicular lymphoma and the CELESTIMO trial. The combination of Lunsumio and Polivy in second-line transplant ineligible diffuse large B-cell lymphoma and SUNMO and for Columvi and among others, the combination of Columvi plus GemOx in the second-line setting of diffuse large B-cell lymphoma in the STARGLO trial. And I'll discuss some of the other studies you can see on this list, which really rounds out what I think is an ability to move these molecules in combinations and ultimately important to earlier lines of therapy. As you know, Lunsumio was the first CD20/CD3 bispecific approved for non-Hodgkin's lymphoma and the only fixed duration bispecific in the treatment of relapsed/refractory follicular lymphoma. We recently updated our third-line data showing high response rates and very durable responses using a fixed duration of therapy in the outpatient setting. Additionally, Lunsumio provides a very manageable adverse event profile, principally limited to low-grade cytokine release syndrome. Importantly, at the 24-month landmark, 100% of patients who achieved complete response at end of treatment remain alive. And I think that 100% overall survival for that 40% who achieved the CR is an important landmark for this therapy in the third-line setting. Recently, also, we compared the Lunsumio response to the response of patients with their last prior line of therapy. The median duration of complete response for Lunsumio has not yet been reached, whereas the median duration of complete response in the last prior therapy was 16 months. As you can see in this graphic, the median progression-free survival for Lunsumio was twice that of their immediate past prior therapy, 24 months versus 12 months. As you're aware in oncology, we typically see a steady decline in response to therapy with increasing lines of therapy. And the fact that Lunsumio produces a clearly superior outcome compared to the immediate previous therapy, speak significantly to the impact of this molecule in B-cell malignancies. So in some Lunsumio represents a new standard of care, first-in-class, off-the-shelf fixed duration and an outpatient therapy for patients with follicular lymphoma. Columvi was recently approved, as you know, in the EU, in the U.S. And at the recent ASCO meeting, we also updated our third-line pivotal data. The data continues to show clinically meaningful outcomes in heavily pretreated refractory patients with diffuse large B-cell lymphoma, including those who had progressed or recurred after CAR-T therapy with a complete response rate of 40% and a median duration of complete response of roughly 27 months. And that median duration of complete response of note compares favorably to the other bispecific now approved pre-diffuse large B-cell lymphoma, more of in contrast to that other bispecific. Columvi offers a fixed duration and therefore, a welcome treatment-free interval for patients who then complete that fixed duration of Columvi. At the same time, Columvi offers physicians a much similar dosing schedule and ultimately an easier time to deliver this therapy in a routine setting. Further building on the success of the Pola-R-CHP regimen in the first-line setting of diffuse large B-cell as well as the durability and robust response of Columvi in refractory settings, we're now launching our next major initiative in the curative setting of diffuse large B-cell lymphoma, namely the regimen of Columvi plus Polivy and R-CHP versus Polivy R-CHP. We've completed a Phase I study of this combination, which shows a 100% overall response rate, a high complete response rate and excellent tolerability with really very limited CRS, in fact, limited to just a few cases of grade 1 toxicity and we'll be updating that Phase I data at the upcoming ASH meeting, which includes additional progression-free survival data, which we believe supports the utility of this combination of the bispecific plus Pola-R-CHP in the curative first-line setting of diffuse large B-cell lymphoma. Further building on this data for Columvi is we are advancing a dose escalation trial of the combination of Columvi with our CD19-4-1BB Ligand bispecific in relapsed/refractory non-Hodgkin's lymphoma. The addition of this signal 2 co-stimulator appears to reverse T cell exhaustion. As you can see in the right panel of this slide, we demonstrate a dose response with a reversal of terminally differentiated PD-1-positive CD8 T cells. And in preclinical models, this combination leads to a more durable response. So far in our Phase I study of heavily pretreated refractory DLBCL, the combination is demonstrating activity with manageable safety and we continue this effort. So at the moment, we're continuing on with this chemotherapy-free combination that leverages a signal 1 and a signal 2 co-stimulatory combination typically associated with CAR-T therapy, but in this format, providing an off-the-shelf approach for providing a chemotherapy-free combination of these co-stimulatory approaches. Beyond non-Hodgkin's lymphoma autologous CAR-T therapy targeting BCMA has shown significant activity in multiple myeloma but, of course, has its unique and a serious number of challenges, namely a lack of access for patients given the nature of CAR-T therapy, the need for bridging therapies, the need for apheresis, long manufacturing times and high manufacturing costs. Therefore, an allogeneic off-the-shelf CAR-T could address these unmet needs by eliminating the need for apheresis providing on-demand therapy and offering better quality T cells from healthy donors. In partnership with Poseida Therapeutics we're advancing an allogeneic CAR-T targeting BCMA for the treatment of relapsed/refractory multiple myeloma. The technology uses donor-derived T stem cell memory cells as well as a highly specific piggyBac nonviral transposon-based integration that introduces a humanized BC -- anti-humanized BCMA CAR. Moreover, leveraging Poseida's Cas-CLOVER technology, site-specific gene editing, knocking out the endogenous TCR to reduce or eliminate graft versus host as well as knocking a beta-2 microglobulin to eliminate or reduce host graft versus graft reactivity. So far in preclinical models, these CARs are actually very active against refractory multiple myeloma. And in Phase I in dose escalation, we're seeing activity and very manageable safety profiles. These data have been submitted for the ASH, the upcoming ASH meeting, we look forward to sharing those data. Beyond this, we recently had approval for our IND for a CD19/CD20 allogeneic CAR for the treatment of refractory relapsed refractory diffuse large B-cell lymphoma, and that Phase I study is anticipated to start later or early next year. Finally, I want to offer some limited updates on our portfolio in nonmalignant hematology. There's a growing recognition of the medical needs of people living with complement-mediated disorders, including PNH and overcoming the challenges of currently available treatment options. Crovalimab, as you're familiar, is an anti-C5 antibody using a recycling technology that enables rapid and sustained complement inhibition and further allows for monthly subcutaneous dosing. Last year, we reported out the results for our single arm initial positive COMMODORE 3 study, and those data have been submitted to China for fast -- for a breakthrough designation and priority review, and we're looking forward to approval in 2024. And more recently, we reported the positive results of the Phase III COMMODORE 1 and 2 trials, which both met their co-primary endpoints and those data have now been submitted to both FDA and EMA with expected approval, we think, in 2024. Beyond PNH, we have ongoing studies in atypical hemolytic uremic syndrome as well as both acute and chronic sickle cell disease, a very large unmet need to which we think crovalimab could significantly affect the natural history of patients with sickle cell. And then finally, building on our leadership in hemophilia A is SPK-8011 a recombinant AAV delivered factor VIII gene therapy containing a codon-optimized factor VIII gene, which is under the control of a liver-specific promoter. In contrast to other gene therapies for hemophilia A SPK-8011 demonstrates stable and durable factor VIII expression for years, in fact, now with follow-up, up to 5 years. Moreover, with the majority of patients with factor VIII expression in the mild hemophilia range. Moreover, in contrast to the other gene therapies in development, we do not see an initial overshoot or over expression of factor VIII, which can lead to thrombosis and has required anticoagulation. Ultimately, SPK-8011 has provided clinically meaningful reductions in the data that's been delivered so far with an 82% reduction in annualized bleeding rates for those patients who require prophylaxis and a 99% reduction in annualized bleeding rates for those required on-demand therapy. The Phase III for this program is planned for later this year, and we look forward to the continued advancement of this important gene therapy for patients with hemophilia A. So that concludes my presentation. I'm happy to answer any of your questions during the Q&A session. And now I'll turn to my colleague and friend, Paulo Fontoura, to talk about indications outside of oncology.

Thank you, Charlie. And hi, everyone. Always a pleasure to be here with you all. I see if this works. Yes. So it's my job in the next 2.5 hours to tell you about all these molecules and I've just been told I only have 25 minutes. So you'll pardon me if this is a little bit of a whirlwind tour. As you can see here, there's multiple opportunities in the next the mid- and the long term that we feel very excited about. And just starting on the neuroscience pipeline. This is, as Thomas mentioned this morning, very, very proud to say, we're the #1 company in neuroscience, and you see it reflected here on this industry-leading portfolio that has both a lot of differentiated targets and modalities, but also really targets major medical needs, both in immunology and neuro degeneration and rare diseases. And of course, we have some launch molecules here on the left-hand side, as you can see, 4 drugs on the market now, and we're looking forward to further key readouts in the next 6 months. And I'll start maybe with the first one that people are asking about, which is our gene therapy in Duchenne muscular dystrophy. This is a partnership with our colleagues at Sarepta and as you know, this is a gene therapy, which was approved in the FDA in the U.S. for accelerated approval just a few months ago. And here on the left-hand side, you see the data on which this approval was based. This was a composite of 3 Phase Ib trials. Trials 101, 102 and 103, which together showed a mean average improvement of 2.3 points on the NSAA as a primary endpoint compared to a natural history of control, which you see here, had a decline of 0.1 points. Now of course, this resulted in a limited label in 4 to 5 year old boys, and we're eagerly awaiting the results of the EMBARK study, the 301 study which is part of a really ambitious and very, very broad program in Duchenne muscular dystrophy here on the right-hand side. There's 3 Phase III trials in both 4 to 7 years old, which is a core program and then younger patients from 0 to 3 and older nonambulatory patients from 8 to 18 years old. The EMBARK trial that you see the design here for is a 2-year duration trial. The first year gives you a primary endpoint, placebo versus [indiscernible] on the NSAA in boys aged 4 to 7 years old then there's a crossover. If you want the patients on placebo will be given the gene therapy a subsequent endpoint with [indiscernible] 2 years, and there's an open label study that gears up to 5 years. As I mentioned, the primary endpoint is the NSAA as well as we have digital endpoints such as stride velocity 95, which is used in all of these trials and has recently been endorsed by the EMA as acceptable endpoint for approval in the year. Now of course, we're very excited about this trial and the readout comes in just a few weeks. And we based on this trial, the subsequent submission for approval in the EMA and as Teresa mentioned, based on the U.S. approval we've already submitted in several other countries and have recently received approval in the UAE. Now switching over to multiple sclerosis and fenebrutinib. As you know, this is our BTK inhibitor molecule. Dual mode of action targets both innate immune system, macrophages and myeloid cells as well as B cells. And we believe this is a best-in-class molecule, and I'll show you why in a minute. We recently had Phase II data from our FENopta trial. This is a trial in relapsing patients that you can see the results here on the right-hand side, recently presented at the EAN. And in that trial, we measured imaging markers of inflammation in the CNS. T1 gap lesions as well as T2 lesions, as you can see here, this is a very positive trial with over 90% of the reduction in both T1 and T2 lesions, very convincing evidence of benefit and at the same time, a very safe profile. Now of course, we get asked this every time. And last year, we got asked again, where is your evidence of brain penetration and I can only say, if you're curious, we're going to be presenting our first data at ECTRIMS in just a few weeks. So if you're interested, look out for that one. Now I mentioned this is, in our opinion and move forward, maybe this one. In our opinion, this is the best BTK out there and the data is here. So first of all, when you speak about BTK inhibition, you really want to talk about specificity, potency reversibility and safety. Now on the left-hand side, what you see here is really the comparison to the 4 most advanced BTK drugs in development right now in Phase III. And as you can see here, fenebrutinib is the only reversible noncovalent binder. Now of course, it's very important, not just from a safety standpoint, but also from a reversibility standpoint. You see here in terms of potency, both for assays of B cells as well as myeloid cells. It's a very potent drug. It's the most selective one and I've showed you last year, over 150-fold selectivity versus the next one. But importantly as well, it does not have this red warhead here, which is really what the covalent binders have that binds to the BTK pocket. And essentially, that is what we think is related to the mode of action but also potentially to the safety risks with covalency. We have a very broad Phase III program. You see here on the right-hand side, trials in both relapsing as well as PPMS. Our relapsing trial is against AUBAGIO and our PPMS trial versus OCREVUS. So we think that also brings in a certain differentiation because it's really a trial against the best standard of PPMS and what we think is a comparable oral drug in relapsing disease. Now Teresa mentioned, we have already over 2,500 patients dosed with fenebrutinib, not just in MS, but across a range of immune diseases as well. And we have a very robust safety profile, no highs, lows cases, [ trains ] and liver enzyme elevations that are monitorable. Moving over to Enspryng. Enspryng is our best-in-class anti-IL-6 receptor antibodies, satralizuman. It has the same recycling sweeping technology that ensures with once-a-month dosing, you get extremely robust targeting of the IL-6 receptor. And we're selling it across a range of new indications. One I want to call out is generalized myasthenia gravis. As you know, this is a non-immune condition that affects about 1 million people worldwide. A lot of things that happened in recent years with approval of the C55 and now the FcRn class. But we feel that even with that, there's still a substantial percentage of patients that don't get optimal control of their disease and at Enspryng, we feel can provide that additional control for patients who move on from standard of renal care to the first biologic. We have a trial which is reading out mid next year, where you can see here on the right-hand side really targets the best-in-disease profile. So we're looking for a very high efficacy as well as the same convenience and safety that we've already seen with this drug. Continuing on the topic of these next-generation antibodies are really targeted very consistently, in this case, the anti-latent myostatin antibody GYM329. Myostatin is the key negative muscle growth regulator, and you know this is one way to improve muscle regrowth in patients with muscle dystrophy. Of course, there's been trials in Duchenne's muscular dystrophy, which has failed. We're now studying our myostatin inhibitor as a combination in SMA with risdiplam. And we feel this is really relevant because when you're talking about rare genetic disorders, you need to correct both the genetic defect with risdiplam and then add the anti-myostatin on top. And we feel this is a particularly good combo because risdiplam being an oral drug goes to all the muscles in the body and therefore, you can really see more robust muscle regrowth. We have animal data, as you see here on the right-hand side, that shows increasing both muscle mass and muscle strength in animals. We actually have combination data in the SMA models as well that shows similar data. And therefore, we think this will actually be a really good next-generation therapy for these children with SMA. They have now a variety of options of correct a genetic defect, but still need muscle regrowth to increase functional ability. We're also looking at a number of other indications for this molecule. One of them is FSHD. FSHD is a muscle dystrophy. It's a rare disease as well, affects mainly older young adults. It's a chronic debilitating disorder as well, but it moves slower. So it's not as fast decaying, and there's no approved therapies right now. Now we feel in this particular case, the correction of the genetic defect perhaps is less relevant, and therefore, we can actually see more muscle regrowth and more functional impact. So we're looking at this in one trial, which you see here is a Phase II proof-of-concept study where we're going to look at functional benefit after a 1-year treatment of GYM329 and in addition to FSHD, we're also looking at this molecule in a number of other related muscle and metabolic disorders. And finally, on the neuroscience pipeline, I want to end up with Alzheimer's. I remember standing here about a year ago and telling you about the risks and dangers of an anti-amyloid targeting and Gantenerumab. It seems like our managing of risks for you wasn't as robust perhaps as we kept telling you that this is a very high risk program. And sure enough, we did not meet our endpoints last year. Now we did learn a lot from again, including how to target the right population, how I have to write endpoints, right diagnosis, but also importantly, that we need more and faster robust reduction in amyloid in the brain. To do that is essential to get success. And that's really what we're trying to do with trontinemab, I'll show you later. Now Alzheimer's remains a core disease area for us. You see here on the right-hand side, our ongoing program targeting both amyloid as well as tau. So this is still very much for us a medical need. Of course, the recent approval of medicines in this area have been an advancement for patients but also illustrate how much more is needed to get really robust clinical efficacy. So what we're trying to do with trontinemab is base ourselves on our really deep knowledge of Alzheimer's disease development and bring in the next-generation targeting best-in-disease profile. Trontinemab is a brain show Gantenerumab module. So basically, it still has the very high specificity and potency of targeting of fribble area of amyloid species. But the Brainshuttle module increases the penetration in the CNS and we've had data already in animals about 50-fold higher and in humans in the CSF, about ninefold higher from the concentration of drug. Now that's important because that will likely result in faster and deeper amyloid clear. So on the right-hand side, I'm showing you here the design of the trial, which is ongoing now in patients with Alzheimer's. It's a dose escalation trial, and we hope to show some of these data at TDAT later this year. And really, what we're looking for is, on one hand, faster, deeper amyloid reduction in these patients and at the same time to do so safely. And of course, if you can achieve this best industry's profile, we feel that we can build on this emerging confidence in anti-amyloid targeting as a disease modification in Alzheimer's. Now switching to our immunology, infectious disease and the cardiovascular pipeline. As you can see here, it's still a very broad pipeline. We've recently acquired the rights to co-develop the best run with our partners. And of course, the fact is that to say that this is an area that has seen a lot of advancements recently. We're very excited about this, and I'll highlight a few of our programs, one of which we did not speak about last year is the astegolimab which is a first-in-class anti-ST2 antibody in COPD. COPD is the third leading cause of death worldwide. It's a huge medical need. There's no real effective treatments out there yet. Exacerbations happened like you do in asthma, but unlike in asthma exacerbation in COPD leads to very high degrees of visibility, hospitalization that lasts for weeks to months. So preventing relapses, if you want, is a really key endpoint here. The ST2/IL-33 biological axis on the left-hand side is actually a really important axis for COPD and asthma. And as you can see, the ST2 receptor is present not just in, let's call Th2 cells, like helper cell, but also in type 1 inflammatory cells like macrophages, neutrophils and mast cells. And therefore, we believe that targeting ST2 targets these 2 aspects of the inflammatory response in the airway. Now we have Phase II data. This is an IIS, you see here on the right-hand side, the top COPD trial, where we saw around a 22% reduction in annualized exacerbation rate, which, again, is a really important endpoint and at the same time, a reduction in symptoms measured by St. George respiratory questionnaire the SGRQ, which you see here on the top right-hand side. The drug was safe and very well tolerated. And based on that, we went forward into Phase II for us. And recently, we've converted the whole program into a Phase III program. So now we have 2 parallel ongoing Phase III studies ALIENTO AND ARNASA looking at 2 regimens of astegolimab versus placebo, and we're going to look at the annualized relapse rate after 1 year and at the same time looking at function as well. Now importantly, in this population, you've seen some recent advances with other drugs that target more the Th2 pathway. Now what seems to happen as in asthma is that the population in COPD appears to have either an eosinophil high Th2 profile or eosinophil low [ MoRTH 1 ] profile. And recent drugs repurposed asthma drugs really targeted 20% of patients at or EOS high. What ST2 does is we believe it can target the whole range of patients. And actually, we saw some inklings in our Phase II data that in fact, we have higher efficacy in EOS low patients. At the same time, we have both current and former smokers. And of course, there's more and more former smokers, fortunately. This is a big population, but we don't want to leave behind the current smokers as well. So in our Phase III trial, we're actually targeting the whole range of this population and we are powered to see differences here. So hopefully, if we get this drug right, we're going to have a drug that can address the medical needs into a very broadest label. Moving over to our efforts in nephrology. So for GAZYVA, this is our glyco-engineered anti-CD20 antibody that you know it's been approved in many other diseases, we actually are basing this whole Phase III program, this Phase II mobility data. This is a trial in lupus nephritis where we saw complete renal response rate of upwards of 40% very convincing data. And based on that, we moved on to a Phase III trial that looks both at lupus nephritis as well as systemic lupus. We have trials in membranous nephropathy and in the idiopathic nephrotic syndrome as well. All of these read in about the 2025, 2026 time frame. And as Teresa mentioned, together, this is actually a pretty sizable market. Now at the same time, we've recently moved forward into Phase III with our antisense Factor B for IgA nephropathy. This is the most common form of glomerular disease in adult still a major medical need. We know that the complement system plays a key role. You see here on the left-hand side, the complement system that essentially activates and perpetuate the inflammation in the kidney and what we're trying to do is stop that amplification pathway by targeting factor B robustly with this antisense molecule. Our Phase II data that you see here in the middle panel, shows about over 40% reduction in proteinuria over 6 months. This is an approvable endpoint, very robust results and at the same time, and maintenance of the glomerular filtration rate for these patients. So again, really, really encouraging data that led us to start the Phase III trial you see here on the right-hand side, that has just started, and this is going to compare patients over 2 years to see again preservation of renal function. Finally, Teresa mentioned already that we're very excited about zilebesiran and you can't really overstate the medical need in uncontrolled hypertension. Of about 80% of the patients don't have good control over 24 hours. And we know about 50% of patients are at high risk of cardiovascular outcomes or stroke. The reason why it's so hard to control this, despite having so many drugs is that even for a very well-known biology such as the RIS axis that you see here on the left-hand side, there's escape right? So they escape both in the liver and in the kidney, combining drugs has limitations, both from a compliance as well as from a safety standpoint. So what we decided to do with our partners at Alnylam is to go upstream and to go upstream to target the beginning of this pathway very robustly in the liver. And to do that with an RNAi that essentially deplete the angiotensinogen, which is the beginning of this pathway. This has the potential not only to be more efficacious than, let's say, other ways of targeting this pathway but at the same time also because of the sheer robustness of the reduction that you see to do that very consistently over 24 hours and throughout time. And you see here the Phase I data on the right-hand side, which was recently published in the New England Journal with the single dose, you see about 90% suppression of serum angiotensinogen over 24 weeks, very robust data. And at the same time, on the extreme right, you see the data on blood pressure control both at week 8 and week 24. And you see, again, on average, at least a 10 millimeter mercury reduction in these patients, atonic suppression of both diastolic and systolic blood pressure, which is 24 hours duration. Again, this drug was very safe and well tolerated. We've moved forward or our partners did into Phase II trials. There are 3 Phase II trials ongoing now KARDIA-1, 2 and 3. KARDIA-1 top line results were actually just announced last week, study that all its primary and secondary endpoints, again, very safe, very well tolerated, very robust suppression of blood pressure, both diastolic and systolic and between at the highest dose with dosing once every 6 months. Now that is really a game-changing medicine in terms of controlling hypertension. We have 2 other Phase II trials ongoing. KARDIA-2 looks at patients that have uncontrolled hypertension on top of 1 standard of care medicine. KARDIA-3 is going to look at add-on to 2 standard of care medicines in uncontrolled hypertension with high cardiovascular risk. And the results should come next year. Now based on these results, our purpose is to go or into a Phase III cardiovascular outcomes trial. You see it here on the right-hand side, a very robust, very ambitious clinical development plan, looking at these patients first, but there's obviously potential for extension to other cardiovascular indications as well. And with that, I'll turn it over to my colleague, Chris Brittain, and look forward to your questions. Thank you.

Great. Thanks very much,. So I think I'm the only thing standing between you and the Q&A later on. So I'm not entirely sure why all the Roche folks keep texting me, just take a bit more time, just take a bit more time, no worries. But I should intend to stick time. I want to really start off by saying that it's a great pleasure for me. This is my first pharma day. So it's a great pleasure for me to kind of represent the hundreds of scientists and ophthalmologists and partners across the world. We've enabled to kind of a robust pipeline and ophthalmology. And what I really want to leave you all with is the kind of the message that in ophthalmology, we're really looking to be -- we've got momentum. We've got great partnerships, and we've got the potential for real leadership in retina and in ophthalmology. And starting with momentum. You've already heard from Teresa about the opportunity that we're seeing is Vabysmo, but at the same time, we've got a number of other assets in late stage. Susvimo really, really exciting. This is going to be returning to clinic, as you said, at the end of this year and returning to U.S. commercialization next year. And on top of that, we're thrilled that for the first time, we're actually going into disease areas outside of the retina with satralizumab in thyroid disease, which we'll be talking about shortly as well. So just before I kind of talk through this , I just want to take a little bit of a step back and just to remind you, over the last few years, we talked -- been talking a lot about our desire in ophthalmology to transition from a U.S.-only retina company, run it vision over the centers it's kind of a global retina company. And I think as you saw with Vabysmo launch, sales of over kind of broadbuster status, sales of over 1.5 million vials globally. I think we've really -- approvals in over 75 countries. I think I hope you all agree that we certainly achieved that global status in terms of retina. However, the unmet need remains, and I think I'll just probably highlight that by saying that all our Phase III studies in diabetic macroedema neovascular AMD in terms of epoetin durability. 50% of patients after their treatment, still don't have vision good enough to drive in the treated eye. So there's an enormous amount of unmet need remains. And that's why we're going to continue to double down in retina. We've talked about Susvimo, the relaunch but also I should be talking in a few minutes about they are a new molecule, the anti-IL-6 monoclonal antibody in uveitic macular edema. At the same time, we're developing a broad range of research and scientific capabilities such as in-house imaging analysis, fluid and genetic biomarkers and cell therapy research which is enabling us to kind of really take over this leadership of the retina space and also excitingly, expand, which is our longer-term goal, expand into the role of leaders in ophthalmology, not just in retina. Again, we've heard a little bit about momentum coming back to momentum. And we've seen -- we've had a lot of commercial momentum with Vabysmo. And equally, we continue to have an enormous manner of development and research momentum with Vabysmo as well. We filed for approval in retinal vein inclusion in the U.S. and globally. And on the research and the data side, we continue to demonstrate the benefits of the Ang-2 component of the Vabysmo as well. A good example is recent data that we shared, showing a reduced risk of epiretinal membrane formation in patients on [indiscernible] to those on aflibercept. For those of you who don't know the retinal membrane a complication of diabetic retinopathy. And if it progresses patients can lose vision and they needed to be surgically treated. Outside of that, obviously, just to reiterate because I love the streamer. We -- it is coming back at the end of this year in clinical trials, U.S. commercialization in 2024. And equally exciting, we have our first -- our second molecule, which is going into the port delivery implant with an expected first patient in a zifibancimig previously known as the VEGF-Ang 2 DutaFab at the end of this year. Now on the right-hand side of the graph of the slide, I just really wanted to highlight the approach that we're taking within ophthalmology, which is not just to treat vision loss, but to go earlier and later in disease stages. And the good example of the later disease stages is our partnership with OpRegen, which is the -- which I'll talk about later in terms of a potential opportunity to restore vision and restore anatomy. So we talked a little bit about unmet need in retina and ophthalmology. And so I just want to talk about how we're going to go about addressing that. Now in terms of the launch of Vabysmo, our standard of care has been anti-VEGF monotherapy. It's been very successful. It's been really great for patients in halved rates and blindness in many countries. However, there's so much more to see as I think of explain. And therefore, we've been exploring partnerships across the world and across with our scientists and practicing physicians, different ways to approach biomarkers with increase humor, imaging, clinical data, genotyping and machine learning algorithms, which really enable us to have much more personalized treatment options and opportunities here because the diseases are very specific to individual patients. We know that atrophy, inflammation, fibrosis, angiogenesis and ischemia all play roles at different levels in these different retinal conditions. And the opportunity there is we can have a more individualized treatment approach to patients, we can provide more value to not only the patient but also the health care system. So I've talked a little bit about Vabysmo and Susvimo. I think I'm going to be sharing some more information about the Phase III and essentially [indiscernible] moving out of the retina for the first time. We also have -- I want to highlight here that we have 2 Phase II programs reading out in 2024. The more data to come next year. This is for the Antioch program in DME and also the antisense oligonucleotide program in our geographic atrophy. But importantly, we also have a robust Phase I pipeline as well. The example here being as I've already commented on, [indiscernible] and the port delivery and plant. So I'm just going to talk through some of the individual programs to get some updates. Again, I'm going to start with Susvimo as Teresa commented, we've had some really exciting data, which I'm going to just talk briefly through. On the left-hand side, you've got the Pagoda study. This was data shared earlier this year in patients with diabetic macroedema. Patients here were given 6 monthly refills of the port delivery and pump in comparison to monthly intravitreal ranibizumab and the visual acuity was equivalent between the 2. So a real incredible reduction in treatment but than that. And on the right, these patients were -- this is a Pavilion study in patients with diabetic retinopathy. So the exciting thing here is these patients were given refills every 9 months. And what we saw is a 71% difference in 2-step improvement in the diabetic retinopathy severity score, which is obviously a measure of severity of retinopathy. Now taken in combination, these 2 programs really give the opportunity for physicians to offer treatments to the whole range of severity of diabetic eye disease all the way from kind of some opportunities in terms of almost presymptomatic all the way through to the symptomatic patients with vision loss. And importantly, again, just worth highlighting that as we do more and more of these studies, the safety profile due to the improvements in surgical training and in surgical technique is consistently trending better and better. So we're really excited about Susvimo. And we'll be looking forward to filing in DME and DR in 2024. Now we talked a little bit about inflammation earlier on. And it's really exciting that our first molecule specifically targeting inflammation has made it into our Phase III portfolio. This is an anti-IL-6 monoclonal antibody. This is designed specifically for the eye. In terms of a modified Fc region, which enables rapid systemic clearance, and it binds all known forms of IL-6 signaling. The condition that we're going into first is uveitic macular edema. This provided us with an opportunity to accelerate the launch of the product while we wait for the DME results, which come out next year. So the reasons that we accelerated are based on the unmet need, the limited treatment options in UME and also the data. So on the right-hand side, you can see the Phase I DOVETAIL study in about approximately 30 patients in the about 30% of patients had 15 letter or more gain in visual acuity by week 12. And that's why we moved into these 2 identical Phase III studies, Meerkat and Sandcat for those of you who can't spell, Meerkat is the 1 with the K, not Sandcat. And the advantage of this study program is that we have a primary endpoint at week 16, although the studies go to week 48. The approval primary end for the week 16, which gives us a great opportunity to get these data in 2025. So this is moving very quickly this product. Our second IL-6 program is satralizumab, which you've heard a little bit about from Paulo. This is our first foray outside of retina in ophthalmology. So we're all thrilled that this program is ongoing. We're expecting first patient in this year, the Phase IIIs of launch. For those of you who don't know too much about thyroid disease, this is a kind of a very complex orbital inflammatory disease. Symptoms include redness, dry eye, double vision for that protrusion of the eye as well, which results in the double vision and also in severe cases, you can get irreversible vision loss. Now we know IL-6 plays an important role in thyroid eye disease, biomarkers of IL-6 activity have been found in the orbital fat and in addition, we've seen anti artics programs with in small case series demonstrating efficacy. And that's why we've moved again identical Phase III program, SatraGo-1 and SatraGo-2 with primary endpoints at week 24. This is another program, which has the potential to read out in 2025. Now moving on to one of my kind of one of the reasons which I came to the company originally was to try and to find treatments for geographic atrophy. So this is an exciting program. This is, again, as Paulo has described, the antisense oligonucleotide is the same molecule for patients with geographic atrophy. We know that [indiscernible] plays a really important role in patients with geographic atrophy and we've recently seen 2 approved programs in the U.S. And the opportunity here is that -- it's given subcutaneously, which has, therefore, the potential to treat both eyes simultaneously. It's a small base, it's a small volume, so potentially for self-administration potentially at home. And if the safety is good, has, therefore, the potential per my earlier mention, to treat earlier in the disease. So this has the potential if it works to treat kind of [indiscernible] MD before patients become truly symptomatic. So this data is reading out in 2024. OpRegen is another partnership going back to partnerships with Lineage Cell Therapeutics. This is the allogeneic cell therapy transplant. The aim here is to replace the RPE cells under the retina, which are dying or have already atrophy. And then a small number of Phase I patients, we saw an improvement in visual acuity, but more importantly, resolution of some of the anatomical changes which we see in patients with geographic atrophy. So this is an ongoing study in Phase II currently. Now just going back to the -- what I mentioned at the beginning of my presentation around partnerships, research in OmiX and AI and imaging. What we've done is we brought this all together, and we've been able to identify these biomarkers called hyperreflective foci in our images and we kind of keep all our images in-house of all of our studies. But not only have we used these techniques in-house to identify these hyperreflective foci images, but when we look at the data, we've actually found that Vabysmo in these biomarkers are markers of inflammation. We found that on the graph on the right-hand side, you can see that Vabysmo significantly reduces these biomarkers of inflammation in comparison to the anti-VEGF monotherapy. And this is a really important example of why kind of partnering with our physician, our [indiscernible] and other organizations around the world and patient groups is so important. And it's with this kind of partnerships that I truly believe that we are well on the path from becoming just the U.S. only retina company to the global retina company that we are today. And in the coming years becoming a leading global retina company and then a leading global ophthalmology company. So with that, I'm very happy to pass it back to Thomas for some closing remarks.

Thank you very much and I'll keep it quite short. I just want to show you one slide that I showed in the beginning. And I think we've taken you through the slide today in a different part of the presentations. Again, on the left-hand side, you see the biosimilar gap that we are expecting, including the decline in COVID. But I think that Teresa very convincingly talked you through the on-market portfolio. We've launched 20 new medicines over the last 8 years. And I also show you one slide that shows that we have actually one of the youngest portfolios in the industry and this portfolio will carry us forward for the next years. You've heard from a number of us about the Phase III pipeline, elevated tiragolumab, giredestrant, crovalimab and fenebrutinib. These are the only 5 assets that are covered by more than 50% of the analysts, which goes into this consensus out of these 18 analysts. On the right-hand and these are all the assets that are not covered at all, not included in consensus. We talked about 6 CCA, we talked about KRAS, we talked about cancer vaccine. We talked about the [indiscernible] in geographic atrophy, the anti-IL-6 antibody, which is also already in Phase III. So you see there are a number of things, Gazyva in lupus nephritis. There are a number of things in the pipeline that will then help us grow further in the outer years of this decade. On top of that, of course, we are very active also in the partnering space, and we look at assets in all phases of development, including late-stage development. But that's not something that you can snip and plan for. It's something that you have to do with due diligence. You have to make sure that these things make sense from a financial perspective, they make sense from a scientific perspective. And so we keep you posted on all the things that we're doing to shift the pipeline and keep adding new things into the pipeline to keep the growth going into the later part of the decade, but also beyond. With that, I will ask the team up, and we'll go into Q&A. Thank you.

Two questions, please. The first one is very much for Thomas, and then I don't know whether the second one is for Thomas or for Levi. At the very beginning, Thomas, you presented data which showed or you said that you felt that the failure in Phase III that we've all seen more recently could be attributed to the fact that molecules didn't spend enough time in Phase II and moved forward too quickly. And then you also showed a data that said you spent more than your peers in Phase III. One other way of interpreting the same data is that you've got too much money and essentially, there's no competitive tension in your pipeline that forces you to make difficult decisions about which is the best of the best molecule that you have. And so I guess the question is in your pipeline review, did you ever consider spending less? And Levi or to Thomas. What I missed in the opening comments was what's going to change in the pipeline? So after the review that you've obviously spent so much time on, are we going to see in Q3 a big reduction of compounds that drop out of the pipeline or everything carries ongoing, but you spend new money differently. I'd just love to know about what's next from the pipeline review?

I can start and then maybe Levi you can do the second part. First of all, the money that is spent is not spending to Phase III. This is really the money that spends from very beginning of research to launch an NME. That's the total spend, right? So that was not Phase III linked. So one of the things that I said is that we shifted money into R&D and now we will work much more on the effective use of the money because I would agree that there is a certain need for tension to make sure that you also do it faster those assets that are maybe not as interesting. And that's something that we have seen, something that we're working on. So you will see that we'll attract more molecules. But what we want to do is we want to invest into areas that are more promising and first-in-class and really innovative medicines. But it's true that it would be our assessment that just putting more money in is not good enough. You actually have to put a certain pressure in the system that you have constantly also -- some that are moving out, and that's been something that didn't happen as much as it should have.

And maybe I'll add that the specific topic of Phase III success rates actually -- it wasn't that we wait until the end of the R&D excellence initiative to discover that. That's actually something we've been attentive to for quite some time. So in fact, we have -- we've already had some sort of forerunner prioritization in our pipeline stopping programs. So I would say, we won't suddenly see in one quarter a dramatic reduction. But certainly, we will expect even more of that in terms of having a stringent look around things where -- particularly moving to Phase III where there is, in fact, sufficient evidence to make that decision. And then I guess the other point is that certainly, as part of this shift, the stewardship of our investment. So going forward, saying no to more investments, it's addition to sort of saying -- to stopping programs that are maybe fairly far along, but even saying no to things that we might have said yes to, because, frankly, we don't need to take that kind of -- those kinds of calculated risks in some of these spaces. It's not the same dynamics as we may have been dealing with in the heavy days of immunotherapy, for example. So I think there are things in flight, but certainly that you might expect to see more of, yes, because I can also say, the shift is over, right? So now we're putting more pressure into the system so that we get exactly to that situation.

You might want to pass on to Luisa.

Luisa Hector from Berenberg. I also have 2 questions, please. So really following on the R&D productivity from Matthew's question. So there was a lot of data on pipeline volume and your spend but I'm wondering if you're actually calculating returns internally, perhaps there's still kind of projected returns and what kind of trends you're seeing there perhaps similar to your probability of success in Phase III, but hopefully, some signals that, that could improve? So that's the first question. Second one is on the product. So COPD, I'm interested in astegolimab, I just wondered why you think targeting the receptor is better than targeting the IL-33 and whether you might have some endpoints in the former smokers despite the fact you're looking at the product?

Yes. So let me answer your first question. And I think one of the slides that I showed is that if you look at our -- the NPD calculated out our on-market pipeline, our on-market portfolio, I think $74 million was the -- $74 billion was the part for products launched in the last 5 years. So if you calculate how much money we actually spent in those 5 years in R&D, you see that we've gained more back than what we spent. So we have a good return of invest. Going forward, we obviously also looked at the value per NME. And that's something definitely where we also want to bring in a shift to increase the value for NME. It's one of those hurdles that we're setting much earlier in the pipeline and development. And that will also mean when we look at external opportunities, that's something that would play a major in assessing those opportunities.

So I'll take the other one. So regarding the IL-33 versus ST2 question, it's a great question. And the short answer is that no one knows yet, right? Because there's no Phase III data that's been generated in large populations that actually allows us to make some analysis and to really figure out which one is better. Now we are -- in other biologies, I'm not saying it's the same thing as you believe, in other biologies, the receptor tends to be, let's say, the critical node. If you target that, you stop pretty much everything else. And you [indiscernible] usually more abundant and harder to target efficiently. Again, speculations because we really don't have any data. Your question around former smokers, like we don't have a very large Phase II program with astegolimab. What I showed you is basically about 85 patients worth of data. It's very hard to do subgroup analysis in those patients. Naturally -- we did not select for smokers or former smokers in that Phase II trial. Naturally, I imagine we have about the same cut, so about 70% former smokers, 30% current smokers. We're not powered to do analysis to look at that. And we do not have any reason to believe, however, that the data looks different so far. So what we're doing in Phase III is that we're powering the trial essentially to make sure that we can get an answer on the biggest of the populations, former smokers, EOS as well, but we have enough data in there to be able to look at current smokers as well. I hope that answers the question you have.

Peter?

Peter Welford with Jefferies. I'm going to do 2 questions as well, I'm afraid. So firstly, if we just stick with the sort of R&D productivity topic, I'm curious when we think about immuno-oncology and you mentioned obviously, Tecentriq. But obviously, in oncology and in fact, you've done it yourself in Phase I to Phase III jumps of the norm. So I appreciate what you're saying about spending more time, more robust Phase IIs. But I guess, equally, is this an indication that you're going to spend less time on that therapeutic area? Or how should we interpret this? Because I guess a lot of the jumps you've talked about have been in an area where that is basically the normal. If anything, we're going more there with the way we've seen more predictive Phase I studies and data we can get from those early trials. Secondly then, just when we consider -- I guess you highlighted that Roche has very deep R&D with a lot of different areas in the early stages in particular, almost basic science at a level that a lot of other companies don't do. I guess, are you still going to preserve that extent of investment, including, I guess, imagine in a lot of areas that the pharma company may not yet have commercial or late-stage development expertise. So are you in any way narrowing down the focus of that early sort of basic research footprint? Or are they still given free rein to basically go where, if you like, interest takes them?

Yes. Okay. So first part is on Tecentriq, you're right. I mean, this was basically the norm that you would jump over Phase II, right? But it explains also part of why we have lots of [indiscernible] rates and Tecentriq definitely play a role in that. But beyond that, we had other cases as well where we didn't complete the risk in Phase II and then ran into issues in Phase III. So -- and what we know based on the data is that actually if we have robust Phase II data, we have very high success rates,I think almost close to 90s when it comes then to Phase III. So you have to, and we will continue to take a certain risk as we go into the future. But you have to look at it from a portfolio perspective, how many risks do you take, you can just take only risk, right? So I think it's -- that's why we have this portfolio governance topic where we try to manage the risk as a whole for the portfolio. The second question was, again.

Research, early research.

The research -- the early research piece. I think it's one of the strengths that we have these organizations that really do the research to develop new targets. And we also bring in from external targets very early, right? It's quite a lower cost. So I think it's actually one strength. Now what it does mean is that value, for example, shouldn't play a role, right? So value should also play a role when you do research in the very beginning. That's the first thing I would say. The second thing is that -- what it also shouldn't mean is that you don't leverage, for example, platform technologies or you don't leverage systems processes. So I think we have a lot of opportunity to take cost out of the system by simplification and that has nothing to do with autonomy or no autonomy. I mean, if I have everyone in the -- I mean, simplified, not simplified, everyone uses Microsoft or everyone uses Google, that's a simple thing that doesn't mean that I can't work in an autonomous way. But maybe you want to add something on that?

One more point on the first question that you asked, which is you might remember in the 2017, 2018, 2019 time frame, I remember myself giving talks about -- in the world of immunotherapy and just how fast moving the field was, was the model of Phase I and Phase II and Phase III going to shift because there was so much deviation from that around the indications that we've been looking at. But I would say that a lesson from the past 5 years is that if you don't have single-agent activity, you have to go fairly deep into Phase II before you know if you have anything, immunotherapy. And so that doesn't imply that we're going to do fundamentally less, but I think it does change the approach. It turns out the Phase I and Phase II, Phase III model is not actually broken. It is, in fact, the business model. And so that's, I think, a learning for the past several years.

And Levi, I would just add that the world of cancer immunotherapy has changed from the initial time when we didn't know if anything worked. When PD-1 and PD-L1 drugs were first being studied, there were studied in refractory settings where all you needed to show that it was better than nothing and that there were selected responses. And so it availed opportunities that many companies took advantage of at that time to move from Phase I to Phase III. We're in a different world now, right? We have accepted an effective checkpoint inhibitors. So studies now are going to require a more detailed data beyond Phase I. So I think this is going to be across the industry that you're not going to be able to take those risks as easily because the bar is much higher.

And I think regulators are telling us that.

Sachin Jain from Bank of America. Two questions. Firstly, your last slide focused on underappreciated opportunities. I wonder if you could focus on underappreciated risk. So that slide shows that you've got $9 billion of HER2 sales the consensus has roughly flat. So what do you think in HER2 risk is to that versus what consensus represents? And then on your commercial slide, 2 of the assets where you flagged greater than $2 billion potential, maybe you could just frame a little bit. So TIGIT, how much of that is SKY-01 versus a broader opportunity and which indications do you think are the biggest? And for DMD, if you could touch on how big the initial data set in a 4- to 7-year-old is versus the total population, or understanding was that initial data set was a minority of that total adjustable population?

So why don't we take them in a row. So do you want to comment, Pablo, on the percentages of patients in the DMD population?

Sure. I mean, it is a good question, in a sense that the 301 trial is designed to study the population we feel can benefit the most the soonest, right? Not that it doesn't benefit anyone else, right? So it is the highest PTS population also because it's a population where we have the most data from 101, 102 and 103, that's the design and by design, we're doing it this way. In terms of the prevalence, you're right. I mean, there are more people beyond those age range, right, usually younger and older, although the life expectancy usually tapers off after 20 years old. So -- but it is a minority of that. Now, of course, we hope to extend that, right? And the 301 and 302 data are coming in subsequently after that. So it's really kind of a staggered approach also to manage the risk, frankly.

And it is true that in order to reach the full potential of eligibility would need more than just the 4 to 5 points, right...

4 to 7.

4 to 7, sorry.

301 is an extension of the current label, it's not 4 to 5, it's 4 to 7.

In terms of TIGIT, so the first one to read out, obviously, SKY-01 represents a good but certainly not the largest opportunity for TIGIT plus Tecentriq. SKY-06 is probably the largest opportunity and that's why SKY-01, whatever happens isn't a referendum on TIGIT plus Tecentriq. It's just simply the first trial to read out, and there are others that will also have large commercial viability. But I think certainly, SKY-06 is the big kahuna, if you will. And then when we talk about breadth, this is complicated when you kind of look at all the moving parts. So as we have -- as we've said, conservatively, we expect the HER2 franchise to continue to grow through 2006. We don't currently see bio -- I'm sorry, 2026, coffee at lunch probably would have been a good idea. We don't see biosimilars coming in for KADCYLA. So this forecast really depends on competitive readouts. And there are puts and takes that could happen in there that would determine exactly what happens with KADCYLA. As you guys know, more than 70% of KADCYLA sales are currently in eBC, we don't expect competition there until after 2005. It will take time for those patients -- 2025, we're not going back. We're not going back in the past. We don't expect competition to erode that quickly. So I think what we're not expecting with HER2 breast is to really see a cliff and then for Perjeta and Phesgo, 60% of that is in -- 60% of EDC patients are currently treated in adjuvant only, and we don't expect that setting as we at risk given that these trials have high DFS, those rates are at 90% and above. And those trials are very long. We think the [indiscernible] trials. So it will take some time for competition to really be able to penetrate into that area. And right now, 40% of early breast cancer patients are treated in that neoadjuvant setting and that neoadjuvant portion of the business really represents a relatively small percentage of business for Phesgo. So net-net, I think HER2 is a great molecule and it's going to help patients. And it's going to come in into the HER2 positive space. But I think what we are seeing is more of a long slow decline for our HER2 franchise versus we're just going to hit a cliff.

After '26.

After '26 not 2006, that would be...

Maybe one question here to add from the -- from Andrew Baum and since we are touching on the HER2 franchise. So given HER2-TKI acquisition, do you plan to develop in combination with approved HER2 agents, including in HER2 or if this is just a prelude to further business development and a next-generation HER2 ADC?

Yes, we haven't fully disclosed our development plan. But as you can imagine, given the value of this molecule, both as an active systemic therapy and its ability to prevent brain metastasis, we see the future of it as being an integral part in addition to other established standards of care, where not only can you improve the outcome in terms of recurrence or prevention or progression, but ultimately eliminating or we're significantly reducing brain metastasis, realizing that at the time of progression, ultimately 50% of HER2-positive patients will develop brain metastasis. It's a devastating effect. So we see this as something that we believe can be part of a series of standard therapies for HER2 disease.

Okay. I think we get over, Richard.

Richard Vosser here from JPMorgan. Two questions, please, as well. One, just on the demographics of the SELECT trial -- sorry, the SKY-01 trial. So I know the demographic -- I think I'll start again. Richard Vosser here from JPMorgan. Looking at the baseline characteristics for the SKY-01 trial. I think there's only about 5% or less of patients in the U.S. So I wondered your thoughts on that for the approval of the product in the U.S. and the commercialization potential of the product, given that enrollment? And then the second question is SELECT related, actually. So on Zubereitung or whatever I'm going to call it, what's your thoughts in terms of the inclusion of [indiscernible] into the outcomes trial? Or would you just go for non-obese secondary prevention patients and how do you think the SELECT trial impacts the commercial potential of that product?

Yes. With regard to your first question, we -- obviously, despite the inadvertent data disclosure, our take on the data from SKY-01 is that assuming it achieves its desired end with a successful outcome, we see it very likely to be a successful submission to the FDA, EMA, among others. Obviously, we have a growing body of evidence for tiragolumab and its ability to add to the efficacy of Tecentriq. And so should SKY-01 read out positive, we don't see any hindrances if that efficacy data and safety data are what we believe it can be.

Nor do we see hindrances to commercial uptake. It's a data we have positively.

Do you want me to take rest of the ones?

Yes.

So it is a great question, and you're right that the treatment landscape around cardiovascular morbidity, mortality, risk factors is changing very, very fast. We don't -- we haven't finalized our Phase III plans, right? And obviously, a lot of that depends on the outcomes of the KARDIA-1 2 and 3 trials because we still need to know how safe it is to combine with multiple antihypertensive drugs before we design the Phase III. Very preliminarily, we're looking first at uncontrolled patients at a high risk basically. That's what we're focusing on now. Of course, as things change, there is an option to do other things as well, not just in -- again, people with -- vis-à-vis being controlled with [indiscernible] or other incretin drug after other indications that are dependent on chronic control, right? So it's just very premature to even speculate, I'd say, right now.

Richard Parkes from BNP Paribas Exane. A couple of product questions. Firstly, given the topic of accelerating programs from Phase I to Phase III, I just wondered if you could talk about the Brain Shuttle Gantenerumab program because arguably, you've now got good validation if you're impacting the biomarker then likely to see efficacy. So I know we'll have to wait and see the data at CTAD, but what else are you going to need to take that program to Phase III? And if you could talk about your willingness to do that. Then the second is also on the hypertension program. Obviously, one of the characteristics of treating hypertension is the ability to up titrate, down titrate dosing in order to kind of improve pressure control or importantly, reduce risk of side effects like hypertension. So what gives you comfort that you're not going to see any kind of safety issues when you go into a broader population of patients with that program?

Starting with your second question first. I mean, what will give us confidence is the Phase II trial that we're running right now, right? So we still don't know. We know in Phase I, we had one arm of the Phase I trial, where we had an arm on top of zilebesiran. We saw additional control of blood pressure, about another 6 millimeters of mercury in these patients with no side effects, including no increase in creatinine, increase in potassium. So we know that at least from those preliminary data, we don't seem to run into the same problems that doubling down on the RAF system typically are associated with. But of course, we're going to take this very sequentially. We're going to look at the Phase II data carefully before we decide on which population we're going to go for. And of course, this is in the context of the benefit risk. So that's why we're also going to start in patients with really high unmet needs. Okay. So it's high cardiac risk and, of course, uncontrolled disease, but it's very early on, okay? So the data is going to be key for us. Your first question around the Brain Shuttle. Of course, you're right that the recent approvals, including the accelerated approvals and kind of validation of amyloid PET imaging as a service biomarker opens doors in terms of regulatory pathways. And we're very interested in looking at that. We were pioneers in doing that in the past. We want to keep doing that. We're dependent on the data from our Phase I trial, right? So I guess what we're hoping to see is speed and depth of amyloid reduction, safety events, including ARIA. And we have to determine what's the best dose for these patients and the best regimens as well. One thing that we see right now is that the evolution of the treatment paradigm in Alzheimer's is also going pretty fast, right, where you might decide that it's better for these patients to target hard and fast and get them below amyloid positivity and then maintain or not. So there's a lot of moving pieces. For us, it's really dependent on the Phase I data and you'll see some of that at [indiscernible].

Along those lines. We would imply that the conclusion of R&D excellence is that we would never accelerate from Phase I to Phase II. There are clearly context where it makes sense, and Paul is alluding to things that we might see with the branch level that would cause us to think about that.

I would just recommend we do 10 more minutes for our Q&A. And next one would be Michael.

It's Mike Leuchten from UBS. Question on your added strength. You made the point that you are significantly ahead of competition, early setting, but that also means you made decisions before you had a lot of data that now is available. And if I compare your trial to what the competition is doing, it seemed to be enriching in that patient population with a similar number of patients in the trial. So just your thoughts around powering that enrichment strategy versus not enriching the patient in the early setting? And then a quick one for Teresa, just confirming, you're still not seeing Hemlibra switches. Is that correct if I read that correct?

It is incredibly early days, but no.

Yes. I think with regards to powering Phase I or early-phase studies, this is something we've been looking at. And I think whether it be within the MORPHEUS platform or the others, we want to make sure that we have -- when we do randomize trials in that setting, that we actually have a control arm that is sufficiently powered to be -- to give us a stable estimate of the behavior control. I think Levi showed very nicely the compelling data for tiragolumab in hepatocellular carcinoma. And albeit in that design, the control was relatively modest with 11 patients. We have the benefit then of statistically bringing in the data from IMbrave150 to confirm that benefit for the combination that led to the Phase III. But I think we want to leverage all available data sets, particularly in our early data assessments and decisions, and to some extent, I think we were revisiting just how large, for instance, these arms should be to give us sufficient confidence and maximize the probability of technical success.

Comment on giredestrant specifically. I mean, the question was on giredestrant.

Yes. So maybe just a couple of additional points on giredestrant. The first one is that the neoadjuvant, so in general in ER-positive breast cancer and ER-directed therapy, the results of neoadjuvant studies have very commonly portended success in the adjuvant setting. So that's one line -- so even though we don't already have a metastatic data set, that was a result that was important. The other thing is that part of our inclusion criteria, it's a high respiration population. And part of the -- even the inclusion criteria is the same kind of [ KICC7 ] levels where we tended to see the most dramatic benefit of giradestrant . So there are -- the specific powering question is one that, as Charlie mentioned, we've put a lot of effort into including even after starting the trial to think about in cases whether we might need to modify the size to maximize the likelihood of success. So that's something that we've looked at and of course, we'll continue to do so as long as things are enrolling and makes sense.

Thank you for picking it up. And the coopERA study that Levi's referring to really is probably the best data set for ER sensitive patients where we do show both as monotherapy and with CDK superiority when looking at Ki-67 as a proxy for response.

It's Ben Yeoh from RBC Blue Bay on the Asset Management side. You spoke on the incentives about changing the incentivization system. Can you give us some more details here? Would you actually be incentivizing to end projects early? Or how is that changing, particularly on the R&D side? And then you also spoke about evolving culture and talent. What does that evolution mean in practice? Does it mean anything?

Yes. So first, on the incentive side, rather than incentivizing progression, we're incentivizing more outcomes. That's one. And we can have some leading indicators when we look at outcomes. And the other thing, when you talk about performance management system, really much stronger differentiation between high performance and low performance. When it's about culture, I think that -- yes, I think in our organization, we have a very, very good culture. I think it's one of the benefits of working in our company. I think at times, we can also have stronger debates. And the debates are not bad and conflicts in my views are not bad if it's not about the person, but if it's about the topic, right? And so I think those things I think are things that we definitely can and want to strengthen in the organization. Also the topic about autonomy. I think we have a very decentralized organization, it has a lot of strength, but it has also weaknesses. So I think what you can do is you can balance it to a certain degree. For example, as I mentioned, it's -- one of the cost drivers is that we are very decentralized when it comes to systems. I'm not losing any autonomy if we all use the same systems. If you have to go and enter data from one system to the next system, and you have manual labor, that doesn't add value to anyone and for certain patients. So I think my view of the world is -- you have to understand when is the best to be decentral, when is the best to be central, right? And the answer is never 100% all the time centralized, 100% all the time decentralized. I think you have to be with more situational around that. And I think these are just cultural topics that we have in the organization.

And when you say [indiscernible]?

Yes. So for example, if we have different parts of the organization, that some of the internalization would be to push a molecule from one phase into another phase, from one organization to another organization. I think you can look at that more holistically. If you look at an end-to-end development plan and people should take accountability not only what happens in their organization, but they're along the value chain. And I think you can measure that in a way that you can implement that in the culture.

I'm Emily Field from Barclays. I have 2 product questions. The first is on Polivy, there was a recent New England Journal Medicine article noting potentially better efficacy across different subgroups in the POLARIX setting. Are you expecting any stratification of uptake in that setting? Or is what you're seeing quite broad? And then secondly, on the PI3K [indiscernible] third Phase III I believe in the fourth quarter. I believe [indiscernible] has only improved in second-line plus sort of, can you just give us some guidance on what to expect out of the first-line trial if patients in that setting would be willing to tolerate the added toxicity of that regimen?

Yes. With regard to your first question, sure, we saw that letter in the New England Journal. And just to be clear, the data set was leveraging retrospectively a series of Phase II studies. And things to bear in mind: one, the target of Polivy CD79B is expressed on all subsets that were described, including cell-of-origin. And so just mechanistically, it'd be hard to understand why the addition of Polivy would work better in one subset for another. The other is within POLARIX, we really don't see any statistically significant differences within subsets. And I think if you look at cell-of-origin with POLARIX now, it's really an incomplete database. We didn't do central review, probably more than 1/4 don't have cell-of-origin data but we are actually conducting a central assessment. And then the other thing I'll mention is that for those of you who followed the ODAC that we had earlier this year, the FDA did a prime job of investigating every conceivable subset imaginable and really was not convincing that there was a subset that did or didn't respond. And ultimately, I think the proof in the pudding was that there was an 11-2 vote to not only approved, but to give it a broad indication. I mean, Teresa obviously spoke quite well to the uptake of this. And what we're hearing from KOLs is they're embracing it as a routine standard of care for all patients.

Yes. I mean, I think it's causing some conversation, particularly in sort of therapeutic expert circles. But generally speaking, I think we're seeing quite broad uptake in keeping with the label.

Yes, of course. So you're absolutely right that [indiscernible] is in the second or later-line setting. There's a big difference with inavolisib which is in terms of its tolerability, namely, if you look at [indiscernible], 20% or more of patients discontinue because of toxicity. And that's really important because what's clear is prolonged inhibition of the PI3 pathway is what's going to drive the efficacy. What we see with inavolisib is a dramatically greater efficacy and potency, but also far greater tolerability for patients to stay on, we see really very low single-digit numbers in terms of discontinuation. So in fact, if Novo 120, which you're right, is first line, if that study is positive, we think it will become standard of care because ultimately, the best chance to really change the natural history of HR-positive breast cancer is to do it right the first time.

It's Simon Baker from Redburn Atlantic. Two, if I may, please. One on R&D and one product question. Going back to the R&D review. One thing that sensibly survived the pRED, gRED structure. And I think, Thomas, you said at the beginning, that may impose additional costs, but I suppose one of the benefits is it kind of institutionalize a bit of diversity of opinion having 2 organizations. However, in the past, we've had situations where both have been working on the same target at the same time, which perhaps is a little over indulgence. So I just wanted to know have there been any changes in the rules of engagement between pRED and gRED? And then secondly, you gave us a global market share number for OCREVUS. So if you could break that down a little bit between U.S. and other markets because one would assume the share is higher in the U.S. where there is greater infusion capacity and the share is lower and the opportunity bigger elsewhere, any sort of granularity on that would be very helpful?

Yes. So I'll take the first question, pRED, gRED. Some people always throw a [indiscernible] into the mix as well, but the fact is we are a majority shareholder, but we don't have 100%. So I couldn't even combine everything if I wanted to. But with pRED and gRED, I think it's actually a strength that we have also in terms of diversity of approaches, diversity of opinions. And I think with [indiscernible] we have someone who is a really strong leader who is leading this organization for the last 2 to 3 years. Now I think that also the rules of engagement have changed completely. I think if I look at Levi, Hans and [indiscernible], they all work extremely well together, and they always try to find the best solution for the organization overall. So I think what you referred to as from the past, I think we are way beyond that with the leadership that we have in place and also going now deeper into the organization. I think that's the case. In fact, I think if you look at one of the reasons sometimes why people want to stay on the different systems is that they can keep autonomy, but it doesn't really make sense. And everyone's lines that we're taking the next stage of not integrating it, but using more of the synergies across, using the same systems, et cetera, and everyone is fully committed to that. So I think that's the best approach. And by the way, having gRED in the U.S., right, having pRED mostly in Switzerland and Germany, having sites in China, having something in Japan. I think this is actually a good way to tap into the ecosystem of innovation into different parts of the world. And the concern that I had is if you throw everything together, you, first of all, will create chaos for 2 or 3 years. We don't want that. So we want to keep the innovation spirit alive. And we want to work on the things that we can improve and can make it better without disrupting the innovation. And usually that happens when you put things together, it gets bigger, it gets more geographic, it doesn't really work well and actually like the fact that we have these local centers of excellence. And then when it comes about scale, we move into development where we have in the late stage portfolio committee where [indiscernible] and Teresa sits, where the decision is being made which assets go into the Phase III trials. And I think that works well. And then on top, we have this subset of the Corporate Executive Committee, where also Teresa, Levi, Aviv, Hans, [Cristina], James from Partnering where we look at it holistically and portfolio-wise. And then we can bring up things like does it make sense? Or how do we act the team on that? And I think this works really well. And in general, I'm very happy with the team and the teamwork that we have because the team is very open, honest, challenging each other, and everyone wants the best for the company. And I think this is the way how you can achieve great sales.

One other very point on your -- so the one thing that has been interesting to me and even somewhat surprising, during my time at Roche, is how uncommon it is that pRED and gRED end up working on exactly the same targets and actually, in the occasions that they have, they have tended to be incredibly high value targets, where very often even if you only had one organization, you might have still taken multiple shots on goal. One example, of course, now is the bispecific, the CD20, CD3 bispecifics. We now have 2 on the market, but actually, they're kind of non-overlapping. So it turned out the fact that both we're being worked on actually wasn't a liability. So actually, that has although to your point, in principle, it would seem [inefficient] in practice, it's actually highly unusual to see that.

And in terms of OCREVUS, I don't remember the breakdown off the top of my head. I don't want to give you wrong numbers. We can certainly follow up with you. But you are right, the majority of the sales for OCREVUS are in the U.S. that has to do with a variety of different factors, including IV capacity, but we'll follow up with you.

I can add, maybe standing on the market share. In the U.S., we stand with a dynamic market share for some time, I think, at 40%. And in the other markets, I think we'll have to pull off and give you some [indiscernible].

And I don't know, should we do one more question? One final one. Emmanuel?

Emmanuel Papadakis, Deutsche Bank. A couple of follow-ups, please. It was notable that zilebesiran deal took you back into cardiovascular. But to what extent should we regard that as just a very asset-specific opportunity in specialty CV versus a willingness to invest more broadly back into primary care? And then perhaps, Thomas, it's a good opportunity just to remind us your intent and that's due to inorganic sources of replenishing the pipeline. I think we saw some comments in the press recently. Maybe you could talk about that both in relation to primary care and some of your current core therapeutic areas? And then if I could, just to ask a specific question follow-up on trontinemab. Your persistent belief in the amyloid beta hypothesis has been validated. It's a [indiscernible] present to participate in at least the first wave of opportunities to offer patients the therapy and part of the problem seemed to be sticking with gantenerumab for over a decade. How sure are you that Brain Shuttle version of that molecule is the right one taking forward at this juncture and [indiscernible] no to see some backup alternatives very rapidly progressed in development?

Sure. So let me start with tronti. So it is a good question. I mean we do have, let's say, a North Star to aim for because if you look at the trial data from the [indiscernible] gantenerumab and we've modeled these data, we've presented it, actually we've given it to the FDA, we know how much and by how -- and how soon you need to remove amyloid or reach amyloid negativity to achieve a certain clinical outcome. And that's a little bit our guiding star for tronti. So we know how much we want to get. And that's what the Phase I is aimed at, is determining whether gante is a good enough molecule to do that or not. Again, we'll be guided by those data. Now do I think there's going to be like the next generation of analysts, I think it's a really interesting question. We know there are antibodies out there who could have that profile. So I think we'll remain open to that. Brain Shuttle [indiscernible] a twist. If there's a whole new molecule, okay, including a whole new molecular entity. The potency and binding are the same for the [indiscernible]. But in terms of the concentration of target you reach, as I mentioned, in tissue and [indiscernible] is up to 50 fold higher, and it's not limited to just a peri-ventricular areas. It's the whole parenchyma. So when we look at that, I think there's reason to think that we're going to get much more amyloid out and much, much quicker. And again, the data will show.

I'll just add something on gantenerumab and trontinemab. First of all, you should look at the data then that's going to be presented at CTAD. That's the first thing. The second thing is if you look at the gantenerumab data, you clearly see that for those patients where we've got more antibody into the brain, you had attained the benefits. So it's clearly a problem of how much antibody we got into the brain. Now why is that? We can speculate on a number of things. But probably what didn't help is that we uptitrated because the idea was that with the uptitration over 6 months, you reduced the ARIA rate, which didn't happen. So the ARIA rate was pretty much the same as in Phase II, but actually got less antibody into the brain. And so with the Brain Shuttle you're addressing exactly this issue. Again, we can discuss after you see data in [indiscernible]. That's the first thing. The second thing on [indiscernible] look, we have shown that we have established and changed the standard of care in other areas, we were not in Hemlibra, we were not in hemophilia, we were not in spinal muscular atrophy. I think what we really need to look for are these assets that are pretty transformative. And this asset has the potential to be transformative. When it comes to M&A, we look for transformative assets. We know that if you look at early-stage assets, you take less financial risk but you take more scientific risk. At later-stage assets, you take less scientific and more financial risk. And so you have to balance that as well. You can only spend the money once. And at this stage, when you spend the money, the interest rates are significantly higher than they were in the past. You also have to calculate that into the equation that you are able to earn back that money. That being said, we are screening the market, leaving nothing unturned but it has to make sense. And it doesn't -- you can't snip with a finger and say it's more, we're going to announce something because usually, these things are multiple months due diligence really trying to understand every piece of the product, every piece of the company, et cetera. And oftentimes, when you look at the due diligence, you find out, maybe not so good, then you stop it, right? So I think the worst thing would be to buy something that's bad because we want to make sure that the investments are made in a smart way. And so we'll continue looking and we'll announce when we have something and will not speculate until then.

Very good. I think this makes a perfect final sentence. I think we just -- we are at the end of the event and close the Q&A session. I would like to thank again all the speakers and also the IR team, which has been largely present here today in [indiscernible] from the United States. And then we have [indiscernible] where it was in the preparation work and then, of course, I have to mention the back office, Melanie Wolf who was responsible for organizing the event and then also be addressed how and [indiscernible]. So we will have a buffet perception, as mentioned before. So you're welcome to join us. And I think there is more time for you to challenge our speakers on whatever topics you're interested in. Thank you.

Yes. Thank you for coming.