Ladies and gentlemen, welcome to Roche's Half Year Results Webinar 2022. My name is Henrik, and I'm the technical operator for today's call. Kindly note, the webinar is going to be recorded. [Operator Instructions] One last remark. If you would like to follow the presented slides on your end as well, please feel free to go to roche.com/investors to download the presentation. At this time, it's my pleasure to introduce you to Severin Schwan, CEO of Roche Group. Mr. Schwan, the stage is yours.

Thank you very much, and welcome to our half year briefing. Before I get into the financials, let me summarize again the leadership announcements we made this morning. As you have seen, Christoph Franz has decided to step down as Chairman of the Board. And I'm nominated to succeed him as a Chairman at the next general assembly in March 2023. At the same time, Thomas Schinecker will take over as CEO, my current role. And yes, I'm really pleased how things come together. My colleagues at the corporate executive team and myself, we very much look forward to working with Thomas. We are going to prepare for a smooth transition over the coming months. But for now, let's get back to business. Let's get right into the half year results. It's been a solid set of results. You have seen a 5% growth on a group level. A 3% growth for Pharma, which is pretty remarkable. And you will see this in later slides as we still have a significant erosion due to the due to entry of biosimilars. So if you look at the underlying growth, if you correct for Ronapreve and the biosimilars, actually, the new portfolio is growing in the double digits. So this is really a very positive and dynamic evolution on the Pharma side. Diagnostics continues with a strong growth momentum. We have seen additional momentum in the first quarter for the COVID testing. That has decreased in the second quarter and is supposed to decrease further for the remainder of this year. But what is good to note is a really good solid growth of the underlying our routine business, excluding COVID test. So for the future, I think we are well positioned here. Now on the product portfolio, I'll come back to…

Thanks, Severin. Hi, everybody. Great to have a chance to share our results. We're really excited about what's going on in the Pharma division this year. And so fasten your seatbelts. Here we go. 3% overall growth, which I think we're encouraged. It's solid growth, showing that we're really starting to emerge from the biosimilar period. As I'll discuss in a moment, the COVID impact, the pluses and minuses roughly cancel out. So I think you can think about this as a 3%, including the effect of biosimilars. And yes, I think that's back squarely in the black, and we intend to stay there. I'm going to -- I'll talk about it. You can see Europe is down a bit. Japan's way up. That's mostly related to sales of Ronapreve, where we were down in Europe. I think without Ronapreve, we would have been plus 4% in Europe. And in Japan, we would have been at 8% instead of 34%. So that -- but they roughly wash out, and the 3% is pretty close to the actual without pandemic. From a P&L standpoint, so the royalties and other operating income was affected by the large settlement that Chugai had on Ultomiris. And so that was a driver in there. In terms of cost of sales, COGS and period costs were up about 9%. Volume was up 8%. So it was up a little more because we were slightly lower capacity utilization. We were basically at 100% for the last 2 years. And so we came down a little bit, which, frankly, I'm glad to see because I don't like running at 100% capacity. M&D just reflects a little bit of extra spending on launches. And the R&D is down a bit because of really high spending -- extra high spending…

Thank you very much, Bill. And I'm also excited to take you through the Diagnostics results. So let me start with the first slide. Now with almost CHF 10 billion sales, we had a growth rate of 11% at constant exchange rates. And this was driven by a good growth momentum in the base business, but I'll get into that more in a separate slide. In addition, we had CHF 2.1 billion in COVID testing sales. Now just one thing I would like to point out on this slide is around Diabetes Care. I've pointed it out in Q1 but also every quarter last year. We did have a onetime positive effect last year in Q1 due to a settlement agreement. Without this, actually, Diabetes Care would be flat, which shows that it's holding up quite well. Now just going quickly into the different regions. You see North America growing 34%. This is, on the one hand, due to rapid antigen test. But also, the base business is doing really well in North America. In EMEA, you see a decline. And this is driven only due to COVID-19 sales that are going down. Actually, the base business is growing at 6%. In Latin America, you also have COVID-19 testing sales going down. And here, the base business would actually be growing at 16%. And in Asia Pacific, it's a mix between rapid antigen sales and also the base business growth. Now I promised that I would give you a bit more in detail on the different parts of our business. So first, let me start with Core Lab. There are basically 2 effects that are impacting growth in Core Lab: one, the China lockdowns. And you see that basically on the right-hand side, when you look at Immunodiagnostics. So a…

Yes. Thanks, Thomas. My pleasure to wrap it up a little bit and to lead you through a couple of effects that we have had in the first half of 2022. And let me get right into it. You see the highlights here. I will address them all anyway in my slide. So I move forward quickly, and let me give you the overview here. Before I dig really deeply into that, and that's another point I will address in the presentations as well, and you will see it on different slides, is I want to make a couple of comments about the H2 performance because we're sure, and let's grab the bull by the horns, that we will get questions on the guidance. So let me give you a little bit of background for the second half. One element certainly is, when you look at our guidance and when you look how we dealt with COVID, that we assume it's just one scenario, and Severin said it, it's just one scenario, but it looks like that we lose COVID sales in H2 compared to H2 2021. As said, it's just an underlying scenario but the current assumption, which influences the guidance. Another one is a fact. Last year in the second half, we had a base effect on the tax side. We had a significant release of a tax provision, CHF 601 million, which certainly was a major boost to core EPS in the second half of 2021, which is not reoccurring. Another point is I will talk about the Ultomiris in-license agreement a couple of times from Chugai. And you know it had a positive impact of CHF 765 million. And there might be a question, okay, fine, at least this effect should be mirrored in the guidance.…

Thanks, Alan. And I think with that, we'll come to the Q&A session the first question would come from Jo Walton from Credit Suisse. Jo, please.

Jo, can you hear us?

Jo?

Apologies. Yes, I can hear you.

Great, Jo. You have a question?

Yes, I do. I think could you ask us when -- could we get some sense of when you're going to be able to present the TIGIT non-small cell lung cancer PFS data? I think a number of us thought this could be at ESMO. Would it be possible to see this ahead of a possible OS data, which I assume we'll see in 2023? And following up from what Bill said about 100% capacity and the ability to move back a little bit from that, can I ask what would happen if gantenerumab were to be successful? That's got quite a heavy manufacturing burden, I assume. And a final question, if I could just ask Alan. There was a write-off of CHF 336 million of Flatiron in the half. Is there anything that you can give us as an update as to maybe what's gone wrong with Flatiron? And is this an opportunity to update us as to what the success of Flatiron and Foundation held for doing for your business?

Very good. Over to you, Bill.

Great. Thanks for the questions, Jo. In terms of the SKYSCRAPER 1 data in non-small cell lung cancer with tiragolumab, that I think the team is still considering what the best timing is for release of the data, as you can appreciate, because it's an ongoing study with an OS component. And it's also a competitively sensitive field. I think there's just kind of multiple complexities there. So we don't have a fixed date at this point for a presentation on that. The question about gantenerumab success and our capacity. Yes, actually, we were gearing up for gantenerumab production capacity as well as other important medicines in our portfolio. When COVID hit, we were just recommissioning the so-called CCP2 plant in Vacaville, California, which is one of the biggest manufacturing plants in the world. And it was -- basically, we were doing engineering runs with Actemra when COVID hit, which was proved fortuitous because that's the plant where we produced almost all the Actemra and all the Ronapreve that we needed for the pandemic use. But basically, now we expect the demand on those 2 products to be much lower. And so that -- pretty much that whole plant would be available. And that's quite sufficient for gantenerumab for quite a while. So it's not -- gantenerumab would require quite a bit of protein, but the process is relatively efficient. So we feel pretty good about our capacity situation. And let's see. Was it Alan?

Alan, yes. Do you want to comment on [indiscernible]?

Yes, happy. I think it's well addressed because it's more an accounting topic, I would argue, because basically, Flatiron, I think we're pretty happy with the sales development this year. The point is when you compare it to the original business plan that we have had, I think we have been too optimistic. Let's face it. I think that's the point here. But I think, really, we went now through a couple of discussions to reset this and where the business is going. And we would argue, the momentum that we are seeing this year is quite satisfying. So really, it's really going back when we acquired the business and all the expectations we have had. This is not really the point that we doubt in our insights approach. I think we think insight is a great topic to be in. And really working on real-world data is a great topic. So all that strategy is completely intact. The only thing we had to adjust is really the net book value that we had for this asset on our balance sheet because as I said, admittedly, everything is coming a little bit later than originally thought.

Right. Thank you, Alan. And I just confirm from my side, our commitment to leveraging real-world data and advanced analytics, there's a huge opportunity in itself in the long term. And I believe from a strategic point of view, it's very synergistic with our core businesses, both in Pharma and in Diagnostics. Very good. Can we have the next question, please?

Next question would be from Wimal Kapadia.

Wimal Kapadia from Bernstein. So first, can I just ask about the Vabysmo launch, please? So clearly, a very, very strong launch. And I appreciate you cannot be specific. But when we think about the product long term, should we really be thinking about this as a Lucentis-type revenue product of CHF 4 billion? Or should we be thinking of it as an Eylea-type product, closer to CHF 10 billion? So any context there. And then just tied to that, is it fair to assume 2023 will be a blockbuster year for the product? And then my second question is just on Polivy in first-line DLBCL. The first part is, can we really expect to see any real momentum for the product in the o U.S. markets without OS data? And then just tied to that, I think, Bill, you suggested at least CHF 1 billion. And previously, you talked about CHF 2 billion. So just curious what assumptions you now make for this first-line DLBCL setting. What share is feasible? And which patients do you think will use the product?

So Bill, now I'm really wondering how much you lean out of the window. Over to you.

Well, see. Where's our crystal ball? I think maybe -- I don't know if I can give you a number. But what I can tell you is maybe how we're thinking about it and what are the reasons to believe or the reasons to the questioning, whether it gets all the way up into the high single-digits billions. I think the reasons to believe are kind of obvious from the clinical data, right? So yes. I mean -- and you could say, well, that's great for patients who don't get extended dosing on another product. But if you've got a product like this, why would you even start with something else, because the other thing that we know from the clinical trials is the effect on drying of the eyes on central subfield -- subfold thickness, the anatomical impact. It's very evident. And in fact, we're hearing all kinds of anecdotes about that from the switching that's going on in the market now, how patients who really haven't had thorough drying are -- their doctors are seeing this after one injection. And they're calling us. They're calling the investigators. They're calling and saying, wow, you didn't tell me it was this good. And so we're that -- the 4-month dosing interval for more than 60% of patients and, frankly, the fact that the biggest problem in this field is all about dosing, and the fact that patients don't get the doses, they don't come in, and they lose their vision. So I think for those reasons, we ought to be aiming for the sky. The reasons to temper that have to do with the fact that there's going to be biosimilars to both Lucentis and Eylea that are going to be launching around the world. And so it has a little bit to do with price sensitivity and also the trade-off between price and capacity because in -- for example, in the NHS, I mentioned almost unprecedented that we got approval from NICE the day we got regulatory approval in the U.K. I think that's the first time that's ever happened for us. And it's almost the first time it's happened in the world. And that's really about -- in the U.K., they just don't have the capacity to do the injections. And so again, patients lose their sight. So look, I'm bullish. And I'm thinking big. But I guess it's the analyst's job to figure out what the number is.

I'm going to take that CHF 10 billion. That's fine.

That's great. That's great. And then Polivy, you asked about without OS data, will ex U.S. adoption occur? I think so. I mean it's been a long time since R-CHOP launched. What was that, almost 20 years ago? But this is curative intent. I mean we're talking about first-line DLBCL. These are middle-aged people who have a -- basically, a disease that can be cured. Or if it relapses, it's usually going to be terminal. Now we have better second- and third-line therapies that can postpone that. But basically, you have a chance to cure a person in the prime of life, and they're done forever. Or if they have a relapse, then they're going to have this disease, and they're probably going to die from the disease. And so as I showed you from the Kaplan-Meier curve, you're talking 6, 7, 8 patients out of 100. As that curve matures and then the OS curve will come in, we should see similar things. And I think the fact that people take this very seriously, it was underscored by the fact that the EMA moved so fast to approve it. Japan is going to approve it in August. We'll have approval by the FDA later this year. So -- and just to clarify, I said more than CHF 1 billion. But I think the total pool of the potential, including the use in later lines as well as the POLARIX is probably closer to CHF 2 billion than CHF 1 billion. So we'll see.

Perhaps I can also add here now that I get into this phase of looking back of my CEO tenure. There have been 2 -- I mean those who know me for a long time, they know that I typically are not -- I'm not super optimistic, and I'm careful because I know that many things can go wrong, and I've seen things going wrong a lot of time. So I tend to be on the careful side. But there were 2 instances where I was more optimistic than the investor community when I saw the first consensus coming in and where we had debates with investors. And that was Perjeta. This was after the APHINITY trial, where I was really certain. And I also spoke out in a very typical optimistic way about the potential of this medicine. And the second one was Ocrevus. I remember the initial estimates being relatively low. And we all were convinced that this is a really fundamental change in the way MS is going to be treated. And I put a bias point to this category. So in my discussions with ophthalmologists, of the initial feedback I get from the community, I think this is a fantastic medicine, and it has really a great potential. And to see this now confirmed already so early with the launch in the U.S. and to see the uptick in some other markets where it's starting to get it out to patients, again, this is one where I'm -- and this is exceptionally more optimistic than the investor community used to be. So anyway, on that note, Bruno, the next question.

So Wimal, were we able to answer your questions?

Yes. Great.

Okay. Next one would be -- come from Sachin Jain, Bank of America. Sachin, please.

I'm going to kick off with a big picture one first, if I may. So a lot of investors we speak to perceive risks around the growth profile if TIGIT and gante both fail given the pressures we know, COVID decline [indiscernible] [ Kadcyla ], Lucentis, Esbriet generics. [ A lot of the presentation of -- so it talked ] about the strength of Pharma launches and Diagnostics-based business. So the very simple question for me is how does all of that balance out for you, in your mind, in the next 2 or 3 years, '23 specifically? Do you have an ambition to grow given the substantial COVID decline? And how should we think about the next 2 to 3 years without any digital gante launches? So that's the big picture. Then 2 specifics. On gante, I think, Severin, with your [ cautious hat ], you talked about this as 50-50 before. Any shift in that thought process as we get closer [ to data ]? And then the final question also, I guess, for Severin specifically, is on TIGIT. I think investors, through road shows, post results, have received mixed messaging as to where you sit on TIGIT in terms of the [ missed ] PFS, but clinically relevant OS possible. So Severin, from your seat, where do you sit on commercial potential? And how much of a disappointment was it not hitting PFS?

Okay. So on the first question, it's a bit of a difficult one because it will very much depend on COVID. And if we lose COVID-related sales altogether, I mean, you can do the math, then it will be a steep curve for next year. Even though we see fantastic growth with these new medicines, even though there is the Vabysmo opportunity and others, I think it will be tough. If we have some tail business, if potentially COVID goes into next year, who knows? Then I think growth is possible next year. That's how I would frame it, without making a guidance prematurely. Now if you look beyond that, as soon as we have washed out this COVID effect, then I clearly get very optimistic because not only [ get me ] over the COVID washout, but also the impact of biosimilars will come down. And I see no reason why we should see a declining momentum for our franchise. On the contrary, there's a lot getting online as we speak. So 2023, we always said this year might be a transition year, the last year might be a transition year, and then COVID kicked in, and then suddenly, we had to start all over again. And now, of course, the success of today is getting the hurdle for tomorrow. If we wouldn't have had COVID, if there was no COVID over the last 2 years, I can guarantee you, we will grow next year, right? And I remember I was once saying in '23, we will grow, and we would grow if there was no COVID effect. Now with COVID, which we could benefit from and hopefully also made some contributions for health care and societies around the world, it's a bit more of a challenge in relative…

Next questions would come from Richard Vosser.

So first question just on the adjuvant trial for Tecentriq. The head and neck seems like -- obviously delayed, but it seems like there was probably an interim analysis this year. And also, renal cancer, as you said, hasn't hit. So just thinking about your confidence in the remaining Tecentriq adjuvant trials and particularly head and neck, given it's been pushed out a little bit. Second question just on Diagnostics' margins. Margins in first half, very similar to second half. Clearly, COVID revenue is going down. It will reduce the margin, as you say. But should we be thinking of that 20% margin this year and then a decline next year? Perhaps some shape of the curve there. And I'll probably stop there.

Thank you. Bill?

Yes. Sure. Let's see. In terms of the adjuvant studies -- so I think this sort of scales with the effects in the metastatic setting. So for example, one of the readouts we'll have in the second half is adjuvant liver cancer. And I think that the Tecentriq plus Avastin data in metastatic liver cancer was quite unambiguous. Really big impacts on PFS, OS. Very clear results. And so that makes us pretty bullish in adjuvant. In other indications, like RCC, where the data has been a little more mixed, then maybe it's not as surprising that the adjuvant readout would be, yes, less strong. And I think in head and neck cancer, I mean, basically there, we have an IDMC. They've looked at the data and said the trial should continue. So if they had looked at it and said it was futile, then the trial wouldn't continue. And so we'll have to wait and see what we learn in 2023. So I think we've got -- yes, we've got some things ahead of us still with Tecentriq. And I would say, Severin, if -- you challenge me on this if I get it wrong. But like, for example, when you said we're not planning for gantenerumab, we're not counting on it, but we're definitely gearing up for it. [ It's stuff you do ] when you have a sort of 50-50 thing. From an operational standpoint, we absolutely -- we have teams. We're working on all the things you work on in terms of health economics and access approaches in different countries and obviously, manufacturing preparations and things. So -- but I think the point you made is right on, that we're not reliant on gantenerumab. In fact, it's not even that significant even in a positive case in the next 18 to 24 months because we would just be launching out at the end of that period. So -- and then I would just say on tiragolumab, as Severin said, the OS, that's a big part of it. That's why we put most of the alpha on OS. And I think there's more than 10 cancer immunotherapy studies so far that missed on PFS and hit on OS, which is why we use co-primary endpoints and why we put most of the alpha on the OS, because at the end of the day, that's the one that counts. So I think -- I don't know if the message has been received in a mixed way. But I think, yes, we have reasonable confidence that we'll have a positive outcome on OS. And we won't know for sure until we get it. But until then, we're -- yes, we're still optimistic about tiragolumab.

Thanks, Bill. And thanks for clarifying the important points on gante, very important. There was another question on the Diagnostics' side. Thomas?

Exactly. The question was on the Diagnostics' margins. The -- I think the way you are thinking about it is also the way I would think about it. So that will stay above 20% for this year. And in fact, when we were managing the P&L for Diagnostics, we were always managing excluding COVID sales, right? Because we know the COVID sales can go up, it can go down, and so we try to manage costs diligently based on the base business actually. And then if you actually look back in, I think, 2019, we had margins of 13.9% in Diagnostics. So going to 20% excluding COVID would be a huge jump. But what I can say is that we've made huge progress in really managing our P&L, and we have significantly improved our margins just excluding COVID. So not quite 20%, somewhere in between where we were and the 20%. And also, I mean, if I look at how our P&L, how much money we're investing in R&D, I mean, this is the reason why we're not yet at 20%. But we have so many important programs currently in our pipeline that are really exciting, like mass spectrometry, just to name one of them, and they will be not in the too distant future. And so we really want to see that through to make sure that these products have the impact that they can have.

Can we have the next question, please?

Yes. Maybe a question we'll take from the chat. And it goes to you, Thomas, a follow-on on Diagnostics. Do you have visibility that the additional cobas machines placed through the pandemic are still active and now used for other purposes? This question comes from Luisa Hector from Berenberg.

Yes. So we actually have all of our systems -- or let's say, 90% of our systems are connected to [ our -- these systems ], so we do see the usage of those machines. And we see that they are being used. And what we also see is a lot of consolidation on these -- kind of these trends. And why do we see that consolidation? It's the high level of automation, the high throughput and the menu that we have. So we do see that this is the primary choice. And by the way, another way I see that is that looking at how many systems we placed in the last couple of years, you may think that the demand of these systems is going down, but that's not the case. We're still producing more of those systems than we did prepandemic times. And now on top with the cobas 5800 launch, we have even more opportunity in the segments where actually it's more suitable for, let's say, countries in Latin America or Asia with the 5800 or smaller European countries in terms of throughput. So I'm actually very positive if I look at how the instrument placement is continuing.

Next question would come from Keyur Parekh, Goldman Sachs.

Severin, you kind of mentioned as you kind of looked back at your tenure as CEO, I wanted to ask you kind of a slightly more forward-looking statement. As the Chairman -- as the future Chairman of Roche and as you look at the next decade, what do you think are going to be some of the critical changes to the structure of the global kind of industry? Do you think there will be fewer pharma companies than there are today? Do you think there will be more pharma companies? And what role do you think kind of Roche will play in that new structure? Separately for Bill. Bill, as we look at kind of the bispecific antibodies -- and you kind of were very keen to point out that the indications we are seeing in the near term are very much kind of the -- at top of the iceberg and there's a lot more to go. I'm just wondering kind of as you look at the broader competitive landscape for agents like this, what gives you the confidence that Roche have a differentiated asset here? And then lastly, for both of you, Roche spends more on R&D than any other company in the industry. I'm curious to hear your thoughts on what the trajectory of returns on R&D investments that Roche have seen over the course of the last 10 years and what justifies you continuing to spend at 500, 600 basis points ahead of industry average.

Okay. Thank you, Keyur, for your questions. And I think actually, the last question and the first one is connected. I do believe that going forward, at the end of the day, what counts is that we maintain our power to innovate. That's really what drives growth in the long term, both in Pharma and in Diagnostics. It's all about translating cutting-edge signs into breakthrough, potentially life-saving, curative medicines in the best case and novel diagnostic solution. That's what it is about. And we will continue to focus on that. And I do believe that -- yes, there's a lot of people who say, "Well, is there an end to innovation and this tailing off," et cetera, and I really don't believe so because the complexity of the human body, the complexity of diseases is enormous. And we have only scratched the surface. There is so much more we have to better understand, and couple that with this human kind of curiosity. Those 2 things, if they come together -- humans, we are curious to make progress, to understand. And having such a complex challenge to solve, like how the human body works or actually doesn't work if you have a disease, that cries for further progress in science. And that will continue, and we want to be part of that. So that will be absolutely vital. I think there is one component which we can leverage now on top of it, which is relatively new to our industry, and it was touched on earlier. That's around digitalization. There's no doubt that health care will get more digital. We can leverage data from the real world. We can leverage new, advanced, analytic tools. And that, again, will help us to innovate and to tackle diseases which we couldn't tackle in the past. Now your question, do we see fewer or more companies? I don't think that in the pharmaceutical industry, it is about scale. I think it is much more about ingenuity, creativity, resolving a specific issue, discoveries. And you see that over and over again, also very small groups, if these are talented people, they can innovate, they can make progress. And therefore, I also think that there will not be a consolidation, as you see in other industries like in the generics industries where it is much more about economies of scale. In our industry, we will see and we will continue to see smaller companies popping up. And I don't think that there will be a big structural change. And I'm talking about the innovative industry, right? So yes, it's about innovation. It was about innovation in the last 125 years. And I think it will be about innovation as we go forward. Bill, anything to add on that?

Yes. I think you said it really well. And I think the other thing, Keyur, to remember is that we didn't just increase R&D spending, we took down everything else. And again, we just think that's the future. As Severin said, we got to -- we're going to innovate our way to the future. We're putting a huge amount of energy into how do we better spend the R&D money. So again, we don't think it's just like, oh, we're going to spend more. We need to make sure that it counts. So we're investing in foundational investments like the Center for Human Tissue -- what is it? Sorry.

Translational...

Thank you. Human Translational Biology (sic) [ Roche Institute for Translational Bioengineering ], which is like the organoid center. It's probably going to be the biggest thing of its kind, which is trying to get to much better models, that will be much more predictive of what's going to happen in the clinic and less reliance on animal models. The work we're putting into computation and analytics, I mean, these are all things to increase PTS, to decrease cost, to reduce cycle times, so it's all of that. You asked about the bispecific antibodies, and I agree, it's a very crowded field. And I think anyone could be confused in hematology. Wondering, well, how is anybody going to make money here if there's so many options and so many alternatives? And I think the reason we believe that we can really lead out here and actually make a viable business of it as well is twofold. One is just, in oncology, it's still [ first pass to post ]. If you're second with some efficacy result, I mean, nobody cares. You get 5% market share. If you're first, you get 70% market share. And we're first with mosun. We're going to be first with glofit. We're first with Polivy. We have Gazyva. And that comes to the second point which is combinations. And surely, combinations are going to be the key. And we think we have a lot more faith, for example, in combinations of 2 or 3 off-the-shelf medicines versus CAR-T, which requires a much more elaborate process. So I agree, it's a bit confusing if you look at the field, but I'm really positive on us emerging with a really strong business.

We'll have the next question.

Yes. Next question comes from Andrew Baum, Citi.

A couple of questions for Bill. Firstly, on tiragolumab and SKYSCRAPER, you disclosed qualitative comments about PFS and OS being numerically greater in the tiragolumab arm. You're in possession of the response rate. Could you say anything on a qualitative basis, similarly, on the response rates in the 2 arms, the depth of the responses and the durability of the response? That would help inform us in anticipation of the maturity of the survival endpoint. And then second, Roche has a legacy in infectious disease. Obviously, you are a dominant player in COVID-related diagnostics. You have more biologic manufacturing capacity than anyone else. And where I'm going with this is there's sizable market for antibody prophylaxis for patients at high risk of COVID that plays directly into your hematology or solid tumor franchises. Evusheld is the only player out there. Why hasn't Roche entered this market given seemingly you have all the tools required and a group of patients that's going to need therapy forever given the poor response to monoclonals?

Thank you. Bill?

Sure. I'm sorry to say, I don't think we can really disclose any more on tira at this point, so I'd look forward to being able to share all of that. In terms of your question about prophylaxis and antibodies, this really has to do with the trajectory of the pandemic and the mutations. So we and our partner, Regeneron, we're very far down the path with prophylaxis. We had data out to 8 months from a single dose of Ronapreve. It looked great. By the way, it might look great again, depending on what the next major variant is because remember that Omicron was not descended from Delta. Omicron came from an earlier variant. And so we had great efficacy against Delta, but then when Omicron came, it kind of wiped us out. So I -- the fact -- if you have an antibody that works against the present prevalent strain, then you can get that indication or that application. If we started right now to develop an antibody for Omicron, there's no way by the time it was out that Omicron is still going to be the thing. So that's kind of the problem. I mean there are some protein engineering approaches to try to design around that, but I don't think we're confident enough in that to go whole hog in. It's not to say we're not -- we're still evaluating this with Regeneron, with Chugai. We have -- our scientists are looking at it and working at it, but I don't want to promise anything in the near term.

Thanks, Bill. Next questions would come from Emily Field, Barclays.

Just a follow-up question on the guidance. I appreciate that there was a lot of mechanical moving parts. But I just wanted to confirm that, if there was any components of the business fundamentally that have deteriorated since you gave the full year outlook, are you allowing more room perhaps given the volatility of lockdowns in China? Did the lockdowns in China have any material impact on the Pharma business in 2Q also? And then just on the guidance for AHR, annualizing the first half would suggest an erosion closer to CHF 2 billion. Any reason to suspect acceleration there in the second half?

Right. The short answer on the guidance is no, that underlying business is doing very well. But perhaps, Bill, you can comment also specifically on China.

Yes. So in China, we saw some impact in Q2 on the Pharma business, but it wasn't particularly severe. So I think maybe because -- well, I would say we worked heroically, our people there. I think we had like 100 people who spent 75 days and nights at the plant, sort of sleeping on cots and things to keep the product flowing. And so the patients, even if they were on lockdown, they were still able to get their medicines and things. So the main impacts, pluses and minuses in China, we have growth of new medicines and [ AHR ] biosimilar impact, which is a drag on things, so yes. And then you asked about acceleration, in fact, [ of AHR ] and the outlook. I think we're looking at international. Possibly, China could be accelerating. And so that's why we've maintained the guidance.

But it is -- I mean, it is [ roughly ] CHF 2.5 billion. And probably, that is the higher end of the expectations as we see it today. Good. Can we have the next question?

Yes. Next questions will come from Stephen.

I have 2 questions. First, I apologize, I'd like to go back [ to the OS trend ] in Skyscraper. Bill, first, can you confirm that you have seen the curve? And the question is management's very tempered tone suggests the trend is barely there. There were no optimistic words chosen at all to describe it. Since we have nothing else to go on at this point, would acknowledging your trend be the most accurate metric that we will have until we see the data? And then the second question is on Hemlibra. Does any of the recent competitive data, either from Sanofi or Novo, concern you? And if not, why not?

So Steve, I can't really answer your question specifically. But what I can say is you've probably heard 2 kinds of things from us about tira. One kind of thing you've heard is us explain sort of in detail about co-primary endpoints and allocation of alpha, that we put [ more on ] OS and that kind of thing. And then you've heard from us, say, "Hey, we've not given up on this." And then the other thing you've heard from us is our [ sort of normally ] conservative thing that we say about anything that's uncertain, as we say, "Well, it may not happen. It may not work." So I don't know. I don't know what else to say. But I think the fact that we're still talking about it would -- should suggest to you that -- like, for example, if we say, "Yes, we're still reasonably optimistic, we could have a medicine here," that means we're reasonably optimistic that it's going to be a medicine. And this is Roche talking and translate it accordingly. In terms of Hemlibra and competition, I think the modified factor product, I think, remains to be seen, what that's going to look like from an inhibitor standpoint because, as you know, if you have hemophilia, you don't want to develop inhibitors to Factor VIII. And highly engineered medicines tend to have -- be more antigenic. And so I think probably the field is going to want to see considerable amount of real-world evidence and see what does actually happen in terms of inhibitor formation before that would be sort of like wholesale switching, just like what we saw with Hemlibra, that there was a cautious uptake. If you're going to have a condition for your whole life and you have options…

Next question would come from Mark Purcell, Morgan Stanley.

I have 3. Firstly, on Vabysmo launch. Now it's clear you're very confident in the potential of the product. And the question is how quickly should we expect share to be gained? That gives us a degree of extra confidence in the products as well. So could you help us understand the sort of initial patients that are switching, the reasons for that, whether it's injection burden or whether it's an efficacy consideration? You talked to the J code coming into effect on the 1st of October. Beyond the net pricing effect, should we expect another uplift there to your analyzing sales rate of CHF 500 million at the moment? And are you seeing any sort of dynamic differences between AMD and DME? So just sort of more color on both would be really helpful. The second one is on tiragolumab. I'm not going to ask you about survival curves and things like that. But could you sort of -- given you've got a 1-year head start in terms of understanding for the experience to SKY-01, can you help us understand how you're looking to build on that experience? Are you looking to introduce more combinations to partner Tecentriq and tiragolumab, whether it's to improve the depth of responses or whether it's to target sort of T-cell exhaustion pathways that could generate resistance by CD226, for example? And then the last one is for Thomas. Can you provide us a breakdown of COVID sales in Q2 of rapid antigen versus PCR? The reason for the question is to try and understand where the base of the more stable and the more durable sales component is, whilst we appreciate the fact that your guidance assumes little contribution from COVID diagnostic sales in the second half of the year.

Thank you. Bill?

Yes. Thanks, Mark. So yes, let me describe how I would see the Vabysmo kind of dynamic. So the initial dynamic for the first couple of months was will you get reimbursed? And so that was a period of real hesitancy in the retinal specialists, so you really only have the kind of leading adopters using it. And there, they were really only using it for patients where they're really having breakthrough disease and very problematic on efficacy. So that was sort of the initial. Then we started seeing reimbursement coming in. The word was getting out in kind of April that, yes, this drug's getting reimbursed. In the meantime, we've managed to get -- 80% of the commercial payers are on board as well. And so now -- then we started seeing switching that was not only where we have breakthrough disease, but also, for example, patients that need monthly dosing. I think looking forward, that we should see basically more patients that are requiring, let's say, every other month dosing or every 3-month dosing or also naive patients. Why would you start them on something else? And so I think we can look forward to that dynamic. Finally, the J code will kick in on October 1, which will mean that the prescribers who use relatively less or treat relatively fewer of these patients, maybe they were waiting to see that, that will kick in. As the additional real-world safety evidence comes in, that should also drive momentum and breadth of use. And then finally, we'll have international expansion. As I mentioned, Japan, U.K. have already launched, and we'll have many more countries coming online. So yes, in fact, we should have EMA approval in September, so we'll be really covering a large part of the world at that point. Tira combos, we're not really discussing that. What other things we would do with tira right now, we're definitely contemplating that with the data we have. And we'll look forward to saying more about that in upcoming sessions. And I'll turn it over to Thomas.

Yes. So thanks for the question. So in Q2, we still had quite a lot of rapid antigen sales. So more than 2/3 of the sales were rapid antigen sales. And I agree with your comment that the more stable business is PCR. And the reason is simple. I mean if I just look in Q2 alone, 1 government ordered for CHF 550 million, right? So that was one customer. And if that one customer doesn't reappear, you're missing CHF 550 million. So -- because governments are currently not investing anymore into rapid antigen test, perhaps it will change again in Q4. Let's see how things develop. I mean we were talking about potentially new variants appearing, maybe hospitalization going up. But the reality is that you have very few customers with [ very high ] orders. There is a very high volatility. PCR, these are, I would say, more normal customers that we had prepandemic, labs, hospitals, et cetera, and [ thousands and 10,000 ] of customers, so it's a more stable situation. So I would agree with that assessment. Now also keep in mind that now people don't have to test anymore when they go in quarantine. In some countries, they don't have to quarantine at all. There's no more testing before flights. So there is an impact also on the PCR side from that aspect.

Thank you. I think Bruno, we are really running over time. We have to bring it to a close now. I'd like to thank you very much for your interest. I'm sorry for running over time. I know there are still a couple of questions in the queue, I would suggest that you directly get back to Bruno and his team. And I wish you a good day. Thank you very much. See you next time.