Morning, everybody. My name is Michael Leuchten, and covering analysts on GSK here at Jefferies. It's my pleasure to have Tony Wood with us this morning for the Fireside Chat, Chief Science Officer. Thank you for joining, Tony. It's a pleasure. What I wanted to start with is one of the questions we keep getting on GSK is sort of where we're headed with the R&D strategy. I think the question is both organic and inorganic. Yeah, so let me just begin answering that to give you a sense of the journey we've been on for the last three years. Then I will embroider the organic versus inorganic components of it. You know, what remains still our central focus is the science of the immune system and the application of technologies. That should be a surprise to no one. You will all have been, I hope, on the journey of creating a very research unit-focused R&D organization over the last three years. Within that, of course, we then have, in addition, the internal versus external component. BD, under Chris Sheldon, reports directly to me. That was a deliberate design, recognizing that what we wanted to do was to bring together both external and internal opportunity at an appropriate point in portfolio decision-making. For the last three years, of course, that point has been the transition into marketed assets, which is why you've seen a lot of focus from Luke and myself on what he's called bolt-on deals, the likes of IDRx, the recent acquisition of the GIST opportunity, for example. What we focused, therefore, at the sort of highest level of capital allocation, the peri phase three launch world, was ensuring that the comparisons that we were making between internal and external assets against the broader research unit strategy, which I can give you in a little bit more detail after I stop, promise I'm not going to answer this question for 20 minutes, is then a proper comparison of internal versus external opportunities. Better capital allocation, and also for me as Head of R&D, a very clear lens on what was coming out of the internal organization, and therefore what I wanted to transform as we seek to deploy technology more deeply within the internal R&D footprint. That is kind of the highest level bit, Michael. Happy to stop there. I can embroider each of the research units from a sort of therapeutic focus in a little bit more detail if that would be useful. Yeah, no, that'd be fantastic. Maybe if we pick on oncology in that context, can you just run us through, because you've got some very interesting assets? You mentioned a KIT, I guess, B7-H3, but they were added from outside the organization. If you apply that lens to oncology, what's coming out of the internal organization? What are the KPIs that you've set? What happens if they don't deliver to those KPIs? Yeah, I might answer that in a slightly different sense, because there's a particular approach that I've taken in oncology that I can then compare with the other research units as well. Very pleased about the focus we've got now in oncology. Obviously, momelotinib, a great example of the BD that I mentioned earlier, doing extremely well. You said yourself that we have both the IDRx and the GIST molecule building out our focus there. Of course, a huge opportunity sitting in front of us for the B7-H3 especially, but also B7-H4 ADCs. In oncology in general, what I wanted to do was pivot the organization's focus into assets where efficacy was apparent in monotherapy in phase one. We've learned as an industry, I think, across IO that the challenges that you face when you pick assets that require combination and then large-scale studies to see effect just really make that approach particularly impractical. Building all of that, of course, on where I stand with Jemperli, particularly in GU and GI cancers, so that we've got a bit of a foundation to build on. Oncology, you'll all appreciate as well, is probably the paramount area in our industry where one can harvest near-term assets from the outside world. What I wanted to do in building the internal organization in oncology was not to compete with what is inevitably a much more scaled external world, but to rather put in place the technology foundations that allow me to make better decisions about which assets I acquire from outside. We've built a significant data capability, be it through collaborations with the likes of Tempus, and now building an internal capability, for example, in organoids. That was bringing Tony Ng in from UCL Kings as a leading academic in the field, and the acquisition of Cellphenomics at the end of last year to build that capability. I very much think about oncology as it is using technology to enable better taste in terms of external acquisitions and partnerships, and then using technology and partnerships to place those assets. Just two quick things on the ADCs. The Hengrui collaboration, of course, gives us a very contemporary ADC portfolio with technology that is already proven. Those of you who follow the field will understand that de novo invention of ADCs is not as straightforward as any of us would like. The pace and scale that we get from the Hansoh partnership on the ADCs, coupled with our own focus internally, as I've just described, gives us an opportunity to compete at scale and with technology in the areas that I choose to go after. That was the logic behind oncology. Okay, fantastic. That's very clear. I guess maybe leaving it to you, the same lens applied to the other areas that you're active in. We can talk about respiratory, we can talk about HIV, we can talk about infectious disease, we can do MASH where you added an asset. I guess one good question might be, which area needs most attention through that lens? Like, where do you think there needs to be more to kind of get into the same logical sequence as you just outlined for me? Yeah, so let me talk about the logical sequence for each of them quickly. There's nothing that I do that doesn't need more. So, everything you'd expect me to say as a Head of R&D, there's always things that I want to change. Hopefully, oncology is clear now. Let's do RIRU next. So, we're fantastically positioned there with regards to lung disease. Couldn't be happier with the label we got for Nucala in the middle of the year for COPD in IL-5. You'll all appreciate an enormous opportunity. 300 million individuals with lung disease. And COPD, a disease that if you are unfortunate enough to be hospitalized with, has a prognosis that is worse than many cancers on a five-year setting. So, what we have there is the move in technology. We talked about ADCs for oncology, for lung disease. Think about it as a shift into long-acting monoclonals. We've led the world in what we might call symptomatic relief of lung disease through bronchodilators. Next up was establishing exacerbations as a key issue of improving and ultimately obtaining clinical remission in lung disease. Depemokimab now really well positioned with two significant phase III objectives started in the long-acting IL-5 class for COPD and IL-33 and TSLP just behind that for both asthma and COPD. You'll see us making decisions about moving those into pivotal studies around about 2027, 2028 when I've learned a little bit more about each of the assets. That, coupled with where we are with camlipixant and the opportunity for the PDE3/4 inhibitor we have in the Hengrui collaboration, which sort of peppers the gaps, if you like, in the GOLD guidelines, positions us really well. As a consequence of that, and our anchor point in T2 inflammation, you start to think about what comes next, which is fibrosis. Across the three tissues that I care most about for fibrosis, remember, fibrosis is a component of almost half of all deaths. And the tissues that I care most about there are liver, kidney, and lung. Liver is the place where really the opportunity to resolve fibrosis and the worst form of it, cirrhosis in the context of liver, is now starting to look mechanistically reasonable on the back of the FGF21 class, hence the move into EFI. Respiratory is really about finishing the job on exacerbations and using it as a bridge into fibroinflammation across the tissues that I described. HIV is all about long-acting treatment. Q4M will be up in 2028, Q6M between 2028 and 2030. PrEP will follow alongside that. Lastly, quickly in infectious diseases, the bridge between fibroinflammation and infection is bepirovirsen, which we'll see a lot more data for in hepatitis B towards the end of the year. For vaccines, very much focused on life cycle innovation, boosting associated with both SHINGRIX and AREXVY, while we build the technical platforms to a level of quality and surety that I'm now confident we have for mRNA and MAPS before I pull the trigger on large-scale phase three studies in either of those areas, given the cost of that proposition. That's a quick walk across all of the individual areas. Oh, no, that's very clear. Thank you. I guess before we maybe drill into some of them in more detail, there is sort of this perception out there that GSK doesn't have a strong pipeline, which I think is unfair. As you outlined, there is actually a lot of thinking and logic behind what it is you're doing. How do you comp your pipeline? How do you look at whether it's competitive? Is it number of assets? Is it revenue potential? Are you agnostic to it? How do you? Oh, certainly not agnostic. I mean, one of the, look, at the very highest level, capital allocation, I use the same approach that many of you who manage portfolios do yourselves. I'm looking for the intersection of medical need, commercial opportunity. One of the big things I did in partnership with Luke when I took over was cull the portfolio of everything that did not have the potential to at least be a major blockbuster. That, of course, at the level that one wants to make those changes, is associated with medical need and epidemiology of disease, as well as the confidence in the science and the operational characteristics of the studies that need to be developed in order to deliver. That is very much sitting behind everything that I've just described in terms of the relative areas of focus within the portfolio. Understood. Okay. And then maybe sort of as we think about catalysts for 2026, you outlined DEPI approval, and then obviously that becomes more of a commercial question. We had Blenrep approved, so we're looking forward to that launch curve. From your perspective, what would you suggest we focus on? Is it sort of Jemperli? Is it B7-H3 now gone into phase three? Where do you think the next proof point is? Yeah, look, quite a few. I just want to underscore the point about the quality of the portfolio, because as you can imagine, I feel that quite a lot. As well as that, operational characteristics, so 13 successful phase III studies last year. That is a record for GSK. Five filing launches this year, of which we are four down and one to go. I would say our ability to execute, I feel, is now well proven. They are all in areas which carry significant opportunity. To your point, Michael, they are now part of the commercial execution looking out towards 2031. Behind that are another 15 scaled opportunities that also contribute to that 31 position. There are numerous assets. What I am going to do is just take them roughly in chronological order rather than choose my favorites, as it were. First up, of course, in respiratory is camla... Actually, first up, as I mentioned earlier, is bepirovirsen in chronic hepatitis B. We'll have the data from the B well study early next year. Chronic hepatitis B, a huge opportunity in terms of epidemiology. Again, several hundred million individuals in a massively underdiagnosed and undertreated area. We stand a chance to, very real chance, for a molecule with the first real functional cure. If we had more time, you'd be asking me what that looks like. I'll tell you, it's sort of 15% will be deemed clinically significant, I hope, based on the data that we have from B well and the stratification in that population that we should see something north of that. In the middle of next year, we have camlapixant. COM1 is last subject, last visit this year. COM2 reads out in the middle of next year. I've deliberately adjusted those studies so that I get higher coughing frequency individuals into COM2, which is increasing the probability of success of outcome. We've taken care of a lot of the, let's call them technical issues that Merck faced with gefap ixant as we look through that. I mentioned earlier that the emerging COPD portfolio and Depemokimab in long-acting format, we've now got Endura-1 and Endura-2 and Vigilant. Waited until Nucala results because I wanted to design the characteristics of those studies in such a way that we learned from the Matinee study for the moderate to severe COPD patients. Vigilant is an entirely new study looking at progression characteristics in a previously untreated population in the mild to moderates. Again, it's all about moving long-acting into ultimately clinical remission, if you like. COPD portfolio, the big one, of course, sitting behind all of that is the ADCs. I promise I'll stop after this, and you can ask another question. B7--H3 being the biggest opportunity there. We're very much in molecule characteristics, building and understanding clinically of what an excellent molecule, the B7H3 molecule is. We have a number of expedited development assignments associated with that, which underscore the quality of the data. You'll see more from Hansoh and from us next year, particularly in extensive stage small cell, where we've now started pivotal studies, but as well as that signal finding that will clarify the position with regards to our focus for B7-H3. That is largely in colorectal, where I already have a foundation of activity in the DMMR setting, thanks to Jemperli, and building ultimately into prostate. I've been quietly building a portfolio for prostate through partnerships. That will also be a significant focus. Lung, it's impossible not to be in the area and not be part of non-small cell lung, but I'm expecting what we'll do there is be a close temporal second to Merck. There's still significant opportunity in that setting. Much more on B7-H3. I won't go on to H4. It's a similar sort of story. Okay, fantastic. Maybe in the 10 or so minutes we've got, let's try to do maybe HIV and MASH and see how we go. Just on HIV, there'll be a capital markets day next year. Obviously, one of the concerns out there is that your competitor might be winning in the field, be that in PrEP or in treatment. You mentioned the every four months, but there's a decision point next year where GSK will decide what asset to take forward every six monthly in combination with 184. Can you just talk about the data points that we're still waiting for? What's going to trigger the decision that then allows you to go ahead? Yeah, and look, I mean, before I get into Q6M, because I do want to just underscore the reality of the situation in treatment. For those of you who don't follow the field, 90% of the HIV market is in treatment. We're all worried about what happens to the market when dolutegravir comes off patent. Dolutegravir is a fantastic backbone to all oral therapies. Something like 70% of people living with HIV have a dolutegravir-based regimen. The answer to that, generic dolutegravir, which is going to be very hard to compete with in an oral setting, is to move into the long-acting injectable world. What we've shown is that that not only brings a number of things that HIV, people living with HIV tell us they want, but it also brings huge opportunities and benefits in compliance in the difficult-to-treat population. That is particularly in the U.S., where the majority of market that we're talking about exists. We are the only ones with long-acting HIV treatment regimens. We already have the Q2 monthly regimen on the market. We've been in the lead in that area for nearly a decade now. Although Gilead are making great progress with lenacapavir in PrEP, and certainly one has to acknowledge that they do not have an integrase-based treatment regimen yet. Without integrase, a treatment, I think, looks much, much harder. It's a unique mechanism in terms of its robustness and its efficacy. We will, as I said earlier, we expect to be on the market with the Q4M PrEP in 2028. We have now a very well-established formulation. We had a minor delay with Janssen, and that was more about the drug product and the complexity of making a drug product that supports long-acting. Once you've got that right, there's a significant technical moat that provides opportunity there as well. Secondarily, about the regulatory framework, which requires three consecutive doses in order to establish the long-acting steady-state proposition. That, of course, is going to be the same for everyone. We're very well placed for the move to Q4M. As I said, keep in mind that's 28, matching with LOE for dolutegravir. Next year, we will be making decisions on regimen selection amongst three integrase opportunities. Chief amongst those, if you'll allow me to be a bit nerdy on the science, given my background at the moment, from my standpoint, is VH184, where we have the already established background in pharmacokinetics and an emerging resistance profile, which surprisingly is even better than dolutegravir. Very confident about that. We're also making great progress, as it happens, with the other opportunity heading into the six-month schedule, which would be an N6LS containing regimen. This is a unique broadly acting monoclonal. It's the only example of the case where you can have a single monoclonal to get coverage. We've published this year a number of very encouraging clinical data studies looking at both efficacy and tolerability against the background of a CAB combination. Trying to squeeze as much in as I can, Michael, in such a short period of time. Yeah, okay. Perfect. Maybe lastly, I think MASH, the EFI deal is maybe a good one to circle back to decision-making of what to bring into the organization. You mentioned sort of the overlap with fibrosis, but obviously there were multiple assets out there. You were a first mover, and then others followed quite quickly. Can you just talk about why EFI, why not the others? Like what decision factors? Yeah, simply put, we're very much in the fibrotic world. My focus is more in the severe end. Although we're running an F2, F3 study for EFI, we just started that pivotal study this year. That's about the most convenient regulatory path to a label, bearing in mind what I said earlier in terms of what's important to me for the clinical portfolio. Really, our focus is in the F3, F4 arena. In that particular arena, you have significantly ill individuals. The attendant mechanistic benefits that come with the FGF21 class in terms of lipid reduction and oxidative stress and insulin improvement are all important clinical features for people who are at that later stage of the disease. What's also important for them is reduced dosing frequency. EFI is unique amongst all of the FGF21s. It has the longest half-life, and it is the one that is therefore most suitable for a Q4 monthly presentation. There are a couple of, let's call them prosaic considerations. First of all, all of these molecules have some form of fusion, which is about extending their half-life. What I wanted was a molecule whose manufacturing process was rooted in standard mammalian systems. That's what EFI has, so it's more scalable. In addition, whenever you have a fusion of that type, you always worry about the production of autoantibodies across the class. Just to be clear, they are present, but not neutralizing yet. The TNF world tells you that you sometimes have to wait before neutralizing antibodies appear. EFI has, again, by far the best profile with regards to neutralizing antibodies. Put all of that together, and we felt we'd found the best molecule. We found it at a point in time at which I could tailor the pivotal study as well. Always a good thing in terms of my BD objectives. I want to be able to inherit something that I have the opportunity to have exactly the phase three study design that I'm looking for. We found EFI at exactly the right point in time. I think the economic or financial components of the deal so far, I would think we'd argue that we did a pretty good deal on that front too, thanks to Chris Sheldon. That's my EFI story. Excellent. Okay, and then maybe lastly, in the last minute, if you were to do more BDs in areas of interest, would that be accommodated within your R&D budget? If you picked up another, I do not know, five assets that come in, can that be tolerated within your R&D line, or is there a budget conflict? Look, I mean, you always want, as a Head of R&D, to have a budget in which you feel you're spending every penny on things that are valuable, timely, and technically advanced contributions to your portfolio. You asked me earlier about how is the internal R&D organization going. That is the component of our progress that I hope to be able to show you more of in the coming years. Inevitably, if you start early, it's going to take a little bit longer before it becomes apparent at the level that the folks in this room are interested in. That means that anything that I bring into my portfolio, be it an internal asset or an external asset, has to compete with what is already an optimized position in terms of allocation of capital. I continue, of course, to focus strongly on making R&D more effective so that I can do more with what I've got. There's a big component of tech playing into all of that. I'm always in a position when I bring something new in, whether it's from the outside world or the inside world, that I'm asking it not only to pay for itself, but to pay for the opportunity that it displaces in the portfolio. Fantastic. I think that's all we have time for. Tony Wood, thank you so much. All right, thanks, Michael.