Document Details

# GSK plc GSK plc - Special Call - GSK plc

Thursday, November 30, 2023 8:00 AM

## Event Participants

#### Executives 4

Unknown Executive, Tony Wood, Luke Miels, Nick Stone

#### Analysts 8

Simon Baker, Richard Parkes, Andrew Baum, Peter Welford, Unknown Analyst, Kerry Holford, Graham Parry, Emmanuel Papadakis

#### Unknown Executive Executive

Hello, everyone. Welcome to our Getting ahead of Respiratory Diseases event with GSK management. As usual, the presentation was e-mailed to our distribution list earlier today and is available on gsk.com. Please turn to Slide 2. This is the usual cautionary and forwardlooking statements.

Please turn to Slide 3. Today's speakers are our Chief Scienti!c O"cer, Tony Wood; and Chief Commercial O"cer, Luke Miels. And our presentation will last approximately 30 minutes a day followed by 30 minutes for Q&A. Turning to Slide 4. I will now hand the call over to Tony.

### Tony Wood Executive

Hi, everyone, and welcome to our Respiratory Meet the Man...

# GSK plc GSK plc - Special Call - GSK plc

Thursday, November 30, 2023 8:00 AM

## Event Participants

#### Executives 4

Unknown Executive, Tony Wood, Luke Miels, Nick Stone

#### Analysts 8

Simon Baker, Richard Parkes, Andrew Baum, Peter Welford, Unknown Analyst, Kerry Holford, Graham Parry, Emmanuel Papadakis

#### Unknown Executive Executive

Hello, everyone. Welcome to our Getting ahead of Respiratory Diseases event with GSK management. As usual, the presentation was e-mailed to our distribution list earlier today and is available on gsk.com. Please turn to Slide 2. This is the usual cautionary and forwardlooking statements.

Please turn to Slide 3. Today's speakers are our Chief Scienti!c O"cer, Tony Wood; and Chief Commercial O"cer, Luke Miels. And our presentation will last approximately 30 minutes a day followed by 30 minutes for Q&A. Turning to Slide 4. I will now hand the call over to Tony.

### Tony Wood Executive

Hi, everyone, and welcome to our Respiratory Meet the Management event. As you know, GSK prevents and treats disease in 4 therapy areas. Most recently, we spoke to you about HIV. Today, we'll focus on respiratory, where we have assets across vaccines, specialty and general medicines product areas. Please turn to the next slide.

In this deep dive, we'll cover the following areas: !rstly, our expertise in respiratory disease where our leadership and innovation have advanced treatment from symptom control to disease stabilization with a path towards clinical remission now becoming apparent; secondly, I'll brie#y cover the role of eosinophils in respiratory disease and the importance of IL-5 in treatment as shown by our !rst biologic Nucala, and our ongoing development of depemokimab as a long-acting solution driving new potential in the respiratory market. Our expectations for depemokimab have increased from GBP 1 billion to GBP 2 billion to not more than GBP 3 billion in PK sales. Next, I'll turn to the signi!cant unmet need and market potential in refractory chronic cough and camlipixant di\$erentiation as a best-in-class asset with potential to deliver more than GBP 2.5 billion in PK sales before !nally reviewing key respiratory data readouts from 2024 to 2026 and beyond. Next slide, please. GSK has been a

leader in the prevention and treatment of respiratory disease more than 5 decades.

In asthma and COPD, we started in 1969 with small molecules, primarily to address symptomatic relief through bronchodilation in easy-to-use devices, including most recently with Trelegy, our once daily triple therapy in a single inhaler. In 2015, we launched Nucala for severe asthma, the !rst monoclonal antibody to target IL-5, which added symptom relief by reducing disease exacerbation. This has led to greater disease stabilization and reduces the need for oral steroids. We're also a leader in seasonal respiratory infection, having launched the world's !rst RSV vaccine. We have a heritage in #u with ongoing studies using mRNA for high dose #u and COVID vaccines.

While this is not a focus for today, it contributes to our deep understanding of respiratory disease. Lastly, in addition to reducing exacerbations in severe asthma, we pursue life cycle management and other eosinophil-related diseases, speci!cally EGPA, HES and chronic rhinosinusitis with nasal polyps. For the future, we're looking at the next wave of innovation, including long-acting options and further evaluation of the role of eosinophils in asthma in the pursuit of clinical remission and ultimately, disease modi!cation. Next slide, please. Together, our respiratory medicines and vaccines are driving competitive sales growth, delivering over GBP 11 billion in 2022 with Trelegy and Nucala as major contributors.

Trelegy is the most prescribed single inhaled triple therapy worldwide and is on track to deliver over GBP 2 billion in 2023.

Nucala has already contributed over GBP 1 billion in sales this year and continues to grow in a market with low biological penetration. We'll talk more on this later.

Next slide, please. Even with these major events, respiratory disease remains an area of high unmet need and signi!cant burden for patients and health systems. In particular today, we'll talk about the role of eosinophils in asthma, COPD and chronic rhinosinusitis and the unmet need in refractory chronic cough. Low T2 disease, which I'll explain later, also remains poorly addressed and this will be one area of BD focus for us.

Next slide, please. For over 5 decades, our ambition in respiratory has been to prevent and treat disease by reducing signs and symptoms, addressing treatment resistance and slowing disease progression. We addressed symptom relief with bronchodilation, reduced exacerbations with a !rst-in-class biologic targeting IL-5 and are now on a path towards clinical remission. This means helping patients who su\$er from asthma retrieve 4 things: to be exacerbation and OCS free and to live with symptom control and stabilized lung function. Ultimately, our goal is to change the course of disease, slowing it down, stopping progression and even reversing previous damage.

Let's hear more about this from the patient's perspective. [Presentation]

## Tony Wood Executive

Thank you. Turning now to biology's IL-5. Understanding IL-5 biology is important to achieve even more for patients and a better appreciation of its role create more opportunity. So let's step back and talk about this biology and it's important in treating eosinophilic disease. Next slide, please.

In our development of Nucala, we've shown that T2 in#ammation is a key driver of exacerbations in severe asthma. What do I mean by T2 in#ammation? This is how the immune system reacts to pathogens and antigens. It's especially important in the lung where a large surface area allows massive exposure. A key immune cell in T2 in#ammation is the eosinophil and IL-5 is a powerful pro-in#ammatory cytokine that is responsible for the maturation, proliferation, activation and migration of eosinophils.

Importantly, IL-5 is also involved in the healthy resolution of in#ammation including the programming of wound healing versus !brosis. With Nucala, we've shown the clinical e\$ects of modulation of T2 in#ammatory response, deepening our understanding of the role of eosinophils and laying a foundation to better understand the science of disease modi!cation. Next slide, please. Our leading work on clinical remission with Nucala creates unique insights to the next phase of innovation and indeed our BD focus. In a post-hoc analysis of our REALITI-A study, Nucala demonstrated that 37% patients with severe asthma achieved all 4 components of clinical remission.

Let me remind you of those again, that is exacerbation and OCS free, symptom control and stabilized lung function, which was sustained for over 2 years. Following this, we have a comprehensive evidence generation program to further examine clinical remission and explore T2 biology at the interface of healthy wound healing and !brosis. Next slide, please. As part of the life cycle management, we're also continuing to evaluate eosinophilia in COPD. Our previous Phase III trials, metrics in the trail was the !rst to attempt demonstration of e"cacy with a biologic in COPD, showing a reduction of 24% in moderate and severe exacerbations in patients with high blood eosinophils of greater than 300 cells per microliter.

As you are aware, only one of these studies demonstrated statistical signi!cance and we received a complete response letter for the application. To address this, we initiated a third trial taking into account FDA feedback. Our MATINEE Phase III trial focuses on patients with higher eosinophilic counts, which have been associated with higher exacerbation rates and poorer outcomes. It's also a pure of COPD population and includes patients with emphysema. Seeking to con!rm the e"cacy of Nucala in COPD, MATINEE is expected to read out at the end of next year.

A positive outcome will inform our approach for further life cycle management for depemokimab in COPD, a potential best-in-class next-generation IL-5 treatment for eosinophilic led disease. Let's talk about that next. Next slide, please. Next slide. Building on insights from Nucala, we're developing depemokimab with more potent pharmacology, a longer half-life and ultimately, an improved twice-yearly dosing schedule.

In Phase I, in the higher doses tested, we saw a sustained reduction in eosinophils for 6 months following depe treatment. Backed by extensive PK/PD modeling, we progressed straight to Phase III in 4 indications: asthma, chronic rhinosinusitis with nasal polyps, eosinophilic granulomatosis and with polyangiitis and hypereosinophilic syndrome. We expect depemokimab to demonstrate e"cacy bene!ts due to improved adherence from twice yearly dosing. And our clinical plan will review outcomes associated with remission. We'll build on this with a comprehensive real-world evidence program starting at launch.

This is just one example of to mark risk-taking in R&D, using our knowledge from Nucala and

accelerating development of the next-generation biologic in 2 versus 7 years. We'll have our !rst readout in asthma in SWIFT-1 and SWIFT-2 in the !rst half of 2024. With that, let me hand over to Luke.

## Luke Miels Executive

Thanks, Tony. So the core assumption that we have when we look at how this market is likely to evolve in asthma and COPD is that we're going to see a high usage of biologics. And as you can see on the chart here on asthma on the left and COPD on the right, this is going to drive a material level of growth. And it's our central hypothesis that depemokimab is going to be a key component of that adoption. Next slide, please.

And if we look at what are the components of depemokimab that are likely to drive that change in practice, I mean it is intriguing right now. And this sort of reminds me in sort of mid-90s when you look at TNFs and other biologics. We're in the foothills of early stages of an adoption of biologics within asthma and of course, more recently, nasal polyps. Interestingly, with ENTs seem to pick this up quite quickly in contrast to where we are today with biologics in rheumatoid arthritis. So I think it's fair to say that we'll ultimately see this level of penetration over time, particularly as you see physicians, HCPs in training adopting biologics and having that experience right from the start of the clinical practice in pulmonology and allergy in particular.

If you look speci!cally at the pro!le of depemokimab and of course, we've tested this intensively, we'll go through this shortly. But if you look at in a nutshell and why we think this is going to shift practice, I mean the !rst one is lower frequency of shots. You're going to have better compliance and so that's likely we know consistently across multiple targets and multiple settings, this is likely to translate into better e"cacy. The other interesting element, as Tony mentioned earlier, this concept of clinical remission and embedding that in our studies and this is something that could be very attractive to physicians in the medium term. And then ultimately, when we look at the other elements, speci!cally at the patient level, I mean, real-world practice, we look at arti!cial things, which are clinical trials.

When we look at Nucala in the real world and this is consistent with longer acting drugs, and we've seen this in psoriasis. And of course, people on the call are very familiar with those analogues. These do materially improve the experience of patients and ultimately, the e"cacy. And it's interesting when we look at Nucala and it's true for other biologics and asthma, if you go 12 months out, typically, you lose 6 out of 10 patients at that point. And that's a complex component, but a big part of that is the burden on the patient from multiple shots and physician visits.

So again, to go down from 12 shots to 2 per year, for example, the complexity around copays, the Part B versus Part D element, just the logistics, the physicians' o"ces to administer this and navigate reimbursement pathways. These are all elements, which simplify the experience for the patient and the physician and should drive a better outcome in terms of e"cacy.

Next slide, please. Now the other element here is so not only are we going to see sort of broader penetration and expansion that use through a simpler o\$ering, which depemokimab is. When you look at the life cycle program for Nucala, it's taken about 8 years from the initial indication to get through to this set of indications you can see here with nasal polyps being the most recent. In contrast with depemokimab because the targets have been derisked and we've just taken a more proactive approach to life cycle management, it's going to take about 2 years from the initial indication in eosinophilic asthma right through to HES. Now obviously, then we'll have COPD, which is derisked hopefully with Nucala and what we've learned with the Sano! program.

So it's going to compress this uptake. And so physicians will be able to use this in a broader number of patients in a broader number of settings and we'll also be able to obviously, of course, interact around the label with di\$erent physician groups who could in#uence the usage of this product. And it's a very broad o\$ering, as you can see, in contrast to the alternatives and physicians on the right-hand side of the slide. Next slide, please. And this is re#ected in the market research.

So this is -- obviously, you can imagine we do a lot of work around the target medicine pro!le and what we expect that we'll realize through the clinical program and I think this !rst line for me was very compelling. So we asked 200 physicians on a score of 1 to 7, how do you score the pro!le of depemokimab in eosinophilic asthma? And 1 being no di\$erence, 6 and 7 being de!ned as highly bene!cial. So you can see there 73% of physicians picked either a 6 or a 7 score, which is again, I think, very compelling and you don't typically see this level of favorability for a product at this point. Then when we look in di\$erent market research around, okay, where would you use this product?

What type of patients would you employ depemokimab in? And it's interesting, you see here 57% today would look at bio-naive patients, which is, I think, makes sense on many levels for which I'll get into shortly. But intriguingly, I think also encouragingly 66% of them said, I'd also consider switching current patients to depemokimab and so I think there are 2 things they're not mutually exclusive. You've got an existing bolus of patients who, as I said, we're losing a lot of these patients after 6 months, after 12 months. So they could be right for rechallenging with a simpler, more e\$ective agent.

And then when you ask patients themselves and you can see here 6 out of 10 say that a 6 month injection would be attractive. But I mean, I think this relatively common sense. Most people would rather have 2 injections a year rather than 12 or 6. And then when you look at how they would behave if their HCP recommended, you can see 87% would indicated they would be open to that. And that becomes important because there is a material number of patients that are resistant to biologics because of fear of needles.

It's not an inconsequential number. It's about 1 in 5 depending on the data that you look at. So this, again, when you aggregate all of these elements, you start to see an attractive product. If we go to the next slide, please. One of the questions we often get asked, meeting with investors, okay, well, is this just going to ablate Nucala?

And it brings back memories because when we were launching Trelegy, I had a lot of similar questions. Are you just going to cannibalize your own business? And what we've shown in the same target population of physicians is that, that's not the case. And I think it's more than 60% of our business on Trelegy is coming from non-GSK products. And I think we can

replicate this here with depemokimab and in many ways, actually more con!dent about that.

And the stats here tell the truth. So if you look at about 50% of the business will come from IL-5s, either Nucala or Fasenra. When we look at the usage of Fasenra, the main driver at least in the stats that we collect, the market research that we collect is frequency of administration. So I think this population is ripe for targeting either those on Fasenra or physicians that favor Fasenra because of the frequency of administration. But interestingly, it's really only about 1/4 of these patients that physicians would imagine either not using Nucala and replacing with depemokimab or shifting patients o\$ Nucala and employing depemokimab in its place.

And you can see there the [indiscernible]. They're also right for targeting in this area and represent a sizable portion of the patients. And also, I think what is exciting here is that you've got a sizable population, which is naive and not been treated before. And you can also see that 20%, I was mentioning earlier, that's really largely driven by fear of injection. So I think these are encouraging numbers.

We have to obviously achieve this pro!le in the clinical program. It is compelling. Next slide, please. So pulling all this together, so you've got a product, which is simpler to use, o\$ers the potential in the real world of demonstrating sustained e"cacy with simpler ways of administration, simpler ways of navigating reimbursement, lower burden for commercial patients and Medicare patients in terms of reimbursement and co-pays. And when you build this picture up and look at the interplay between Nucala and depemokimab.

This is our estimate, of course, it's a forecast of what the pattern would be. So in 2022, once depemokimab introduced, we would see some erosion as the market research earlier would indicate. But we expect that actually to be o\$set by the successful implementation of the Nucala COPD label as we enter a new subset of patients. And if you remember the !rst slide I presented, there appears to be a high level of interest and that's certainly been what we've seen in some of the analyst notes around the news that Sano! has signaled over the last couple of days. And then you look at depemokimab is really driving the bulk of this uptake in severe asthma.

And then the life cycle, so EGPA, HES, nasal polyps and then COPD are incremental above that. But we expect the bulk of this business to be driven by that classic eosinophilic asthma, severe asthma patient, where there's still a very high unmet need and about half of these patients still adequately controlled on their current treatment. So that's the full picture there. And if we aggregate all of that, you're looking at about GBP 3 billion or more than GBP 3 billion in revenue for depemokimab at peak. And when you also include Nucala there with this relatively stable base business of Nucala, then that !gure is above GBP 4 billion.

Next slide, please. So now before we go in depth into the clinical and commercial arguments for camlipixant, I think it's appropriate that we give you a perspective from the patient. So you'll hear from [indiscernible] is su\$ering from refractory chronic cough. So we'll just now go to the video. And then after that, I'll come back and we'll get in with Tony into some more debt into this program.

[Presentation]

## Luke Miels Executive

So I mean thank you to [indiscernible]. Hopefully, that was illuminating for you and gives you more of a sort of practical demonstration of what the burden is like for the patient su\$ering from refractory chronic cough. If you look at the numbers, I mean, when we did do the BELLUS deal, a number of investors contacted us and sort of how big is this population? I don't hear people coughing around me every day. What's there after here?

And you can see on the top left-hand side, if you just look at people who have been coughing for more than 8 weeks in the larger countries, you can see a population of 28 million. That's not a population that we use in our models. We then take a more strict slice where you look at individuals who've been diagnosed with simple for more than 1 year and you can still see that's a very material number of 10 million people. When you look at the pro!le and the burden that these patients have. So typically 2/3 of them are females in 50 to 65 years of age.

You can see 500 coughs a day. Obviously, some people are a lot higher than that. Some people are lower than that. Urinary incontinence is very typical. So the social consequences of that, of not only coughing, but the psychological downstream e\$ects of that and depression, you can see here is in about half those patients because of the anxiety impact on their partner [indiscernible] and also just the social isolation that comes with coughing and particularly !ts of coughing.

And of course, in more extreme situations to see some of the elements below on this slide in terms of vomiting and fainting. So it is a disease with a lot of burden. And if we go to the next slide. If we look at the treatment pathway that these patients typically have to navigate and this may be partly because there are no licensed therapies available. But typically, it's a disease of exclusion.

So when someone presents with chronic cough more than 8 weeks, the !rst re#ex of the physician, of course, very sensibly is to try and remove or exclude more serious diseases such as lung cancer, asthma, bronchitis, COPD, IPF re#ux, of course, is very common as a reason for cough. And ultimately, if they get there, there's no underlying disorder that's identi!able then they are classi!ed as having refractory chronic cough. Now the challenge here is, if you look at the bottom right-hand side of the slide, typically, these patients have to navigate 3 or more specialists in 50% of the cases, sometimes it's more than that. And even if they are diagnosed with refractory chronic cough, they tend to rotate di\$erent options because the therapy is here, you can see 3 or more therapies. So di"cult to get a diagnosis and frustrating to get an outcome that is satisfactory for that patient.

If we go to the next slide, this is direct surveys of physicians. So A, this is just a serious disease. Again, going back to the original question that I mentioned at the start of this section is this some sizable disease? Is it a real disease with material burdens that would encourage treatment. And you can see 91% of people, physicians managing refractory chronic cough say it's extremely burdensome.

And then if you ask them, well, what are the options that you have for treating these patients, you can see 3% are satis!ed with the options. So 97% say they're very unsatis!ed. And then when you ask them, okay, if assuming the drug is well tolerated and there is a new treatment

approach, would you employ it in these patients, even if you were not deeply and fully familiar with it. And you can see over 3/4 of pulmonologists and allergists who treat the bulk of these patients would say, yes, I agree or strongly agree in terms of trying a therapy.

So you've got a high unmet need, you've got a sizable population. You've got physicians who are frustrated with the options available. So it is ideally prime for an e\$ective and welltolerated treatment. And with that, I'll hand over to Tony.

### Tony Wood Executive

Let me now describe the biology of camlipixant. P2X3 is a validated biological target implicated in cough re#ux hypersensitization. We know that activation of P2X3 receptors increases the excitability of airway sensory nerve !bers contributing to the excessive cough. In addition of P2X3 receptors, therefore, could normalize the overactive cough re#ex and reduce cough frequency. In this medicine club, we also know that selectivity for the P2X2/3 receptor has proven di"cult with a consequence of taste turbine, given the presence of P2X2 on sensory taste nerves.

Camlipixant's high selectivity for P2X3 therefore, provides potential e"cacy and tolerability bene!ts. Camlipixant is a thousand fold more selective than its current competitor, so has the potential to be a best-in-class medicine for RCC. In order to assess the true potential of camlipixant, a placebo-adjusted reduction in cough frequency is used. As illustrated here, camlipixant demonstrated a clinically relevant reduction in cough frequency at all doses tested in the Phase IIb SOOTHE trial. Importantly, across all doses, there are low rates of taste related adverse events of less than 6.5% and 0 discontinuations related to test.

Next slide, please. Looking ahead, data from our Phase III CALM program for the 25-milligram and 50-milligram doses is expected in the second half of 2025. Following the FDA's advisory committee on gefapixant on the 17th of November, committee members acknowledged the signi!cant unmet medical need in RCC. But suggested that a more robust di\$erence from placebo and the removal of potential confounding by taste e\$ect would have been more compelling. They also suggested looking at di\$erent endpoints that better re#ect the patient experience.

These evidence requirements identi!ed by the committee are well considered in our program for camlipixant. In fact, the camlipixant Phase III design was optimized post deal close to ensure we capture the true e"cacy of the molecule. The optimization was done in close collaboration with the FDA and includes a number of key elements. First, prior to dosing, all patients recruited into the CALM program must be adjudicated by an independent panel. This panel look closely at the patient's medical history to con!rm the diagnosis of refractory chronic cough.

Second, the dosing phases of the trial includes a placebo run-in period prior to randomization in treatment groups. We know that cough trials typically experience a placebo e\$ect and by including this placebo reninphase, we will reduce both baseline variability and anticipated placebo e\$ect. Third, we're enriching the trial with patients with a higher cough frequency. And !nally, the high selectivity of the asset is expected to reduce o\$-target e\$ects like dysgeusia, which could lead to unbinding. With these points in mind, we're

con!dent that the Phase III program can assess the true potential of camlipixant.

With that, I'll now hand it back to Luke to wrap up the presentation.

### Luke Miels Executive

Thanks, Tony. I mean, I think what is always compelling and engaging when we look at an asset is that you've got a viable hypothesis, the underlying mechanism makes sense for physicians as to why they would see a e"cacy. And then secondly, why would there be a di\$erentiation between our molecule and a potential option available to physicians within the same class. This is market research that we've done based on the pro!le of gefapixant and also with real-world experience with physicians using gefapixant in Japan, you can see here that taste disturbance is a problem. And so we spend a lot of time looking at the data during the due diligence with ballast to really understand mechanistically, we have a high con!dence in seeing a di\$erence here between gefapixant and camlipixant.

And that then as he presents itself when you look at the ATP preference. So you can see here, 73% believe that camlipixant is the best-in-class molecule. And you can see again, 85% preferred because of the taste disturbance. So even with neutral e"cacy, they prefer camlipixant because of the lower consequences around o\$-target taste disruption. So for me, this is very exciting.

And if we go to the last couple of slides, when you look at this for the consequence structurally strategic to the business, again, we think we can keep Nucala growing and that it is a very durable business and it has a position. But the bulk of practice will shift to these longer-acting products like depemokimab and I mentioned the analogues that we've had before in a number of very similar settings where longer-acting treatments have really shifted practice, we've seen more people treated staying on longer and physicians electing to use these products ahead of the shorter-acting products, and we're very con!dent that will occur with depemokimab. And again, with camlipixant, high number of patients, high unmet need and a compelling pro!le of a product, which is attractive to physicians and should help patients. If we move now to the last slide before Q&A, if we look at the forthcoming catalyst, of course, COPD. You readout there, as Tony covered earlier, we have a high level of con!dence on this based on the insights we got from our own program.

And the recent success that Sano! has further increased that con!dence. With depemokimab, SWIFT-1 and -2 are on track for a readout in the !rst half of next year and then we've got nasal polyps.

Nimble is the switch study where we're taking people on Fasenra and Nucala and switching them across to depemokimab. So an important study for us. And then we've got the other life cycle programs that I said to get this 2-year time frame post launch for depemokimab and then CALM 1 and CALM 2 for camlipixant reading out in 2025. And again, to reinforce, as Tony has covered and I've also mentioned, we have taken the insights from our discussions with the regulator and the public information that's available on Merck and have enabled these programs to be designed to give us con!dence that they're going to read out as we expect with the target medicine pro!le. So with that, I'll go to questions with Tony.

# Unknown Executive Executive

Thanks, Luke. Tony, obviously, great overview of respiratory and particularly the signi!cant opportunities that we have ahead of us for both depemokimab and camlipixant. Given, obviously, just where we are in terms of time, we'll now start the Q&A, but we'll aim to try and !nish around 2:00, obviously, depends on demand. We'll see where we get to. [Operator Instructions] And with that, we'll take our !rst question from Simon Baker.

So over to you, please, Simon.

## Simon Baker Analyst

Two questions, if I may. Firstly, on depemokimab. In reengineering the antibody, you increased the a"nity by about 30 fold. Could you slow the clearance from about twofold. So I was just wondering, how generalized is that approach to other antibodies?

And because I was looking and you've not disclosed much of what you did for probably understandable reasons. And speci!cally on to depe, what do you think is the real drive the 6 month of dosing? Is it the high a"nity or is it the lower clearance? And then on to a more general question on eosinophilic asthma. Could you give us an idea of current testing rates by severity to determine the eosinophils on driver of asthma in mild, moderate and severe?

## Tony Wood Executive

Nick, why don't I get started on Simon's question with regards to depimocumab and extension of the pharmacodynamic e\$ect. Simon, there's nothing special about this that would mean it's solely applicable to depemokimab. It's fairly straightforward. In depemokimab's case, of course, we had con!dence in measuring the e"cacy of the molecule because of what we learned in translational sense from assays from Nucala. But in general, the reason why you see the extended dosing frequency is a combination of intrinsic half-life and also the ability to, let's say, overdose because of the increased potency giving us a longer pharmacodynamic e\$ect as well, both make a contribution that is entirely calculable on the basis of the PK/PD models that we have.

And we're able, of course, to go to that higher dose in the !rst instance because we're so con!dent in the safety of the mechanism. So generally applicable and following very I would call standard but sophisticating modeling behind it in terms of our approach to con!dence and translation.

## Luke Miels Executive

Yes. And Simon, I mean, the bulk of patients that are being managed by pulmonologists, so those at the severe end of the spectrum are tested, as you know, it's a relatively simple blood test. And the guidelines are very clear on that factor. They also encourage multiple testing because we know -- and this is -- we've used this insight in our COPD program that these levels can move over time. And if you look at the background, it's around 85% of severe patients eosinophilic -- the disease is eosinophilic in nature.

So it's very rational to test for that. For the male, the patients, the penetration is less, as you would expect. But it's something that we've actually been trying to shift in the U.S. with

promotion to Nucala in high using high user primary care doctors and had some success on that front.

#### Nick Stone Executive

We take our next question from Richard Parkes at BNB Paribas. So over to you, Richard.

## Richard Parkes Analyst

Nick, hopefully, you can hear me okay. Yes, I'll stick to the 2 questions. Firstly, can you talk about what percentage of COPD patients are likely to be eligible for biologics treatment based on entry criteria to MATINEE and how that compares to the DUPIXENT studies? And just clarify the di\$erences in improvement criteria between MATINEE in the previous study is it just eosinophil cuto\$ or any other di\$erences?

And then secondly, one of the advisory committee concerns over gefapixant was not just the dysgeusia but maybe that high discontinuation rate really re#ects that what bene!t patients were experiencing was modest and maybe just re#ecting an imperceptible change in cough frequency. So how con!dent are you that you will ultimately have a pro!le where you'll have a more meaningful clinical bene!t in addition to less incidence of dysgeusia.

### Nick Stone Executive

Thanks, Richard. Tony shall come to you !rst.

## Tony Wood Executive

Yes. Let me -- and Mike want to comment on the general population considerations to COPD, but I'd make a couple of points in answering your question, Richard. And let's talk !rst of all about the di\$erence in patient populations between the DUPI studies, [indiscernible] ongoing studies. The important di\$erence to stress is we've included a broader COPD population that includes on emphysemic about 30% of the COPD population. What we've also done in response and the MATINEE design was in response to the CRL.

So as well as higher EOs as you've indicated, we've also been careful to recruit a pure COPD population.

If you look at the impact of both of those things and what that might do in terms of relative reduction in exacerbation rates compared to the Dupi studies, you can !nd a sort of indicator of that in some of our published Phase II data that shows that when we look at a more bronchitic population, which is the one that Dupi addressed. Those individuals have more reactive and responsive airways. And with that in mind, we achieved in our Phase II studies looking at that selected population e"cacy in terms of exacerbation reduction, which is very much in line with the sort of results that you're now seeing reported out for Dupi.

I mean what I'd also emphasize, of course, is that we also have extensive patient safety to the use of IL-5 in that population as well. We're out at now more than 100 weeks. And we don't have the hypereosinophilia observation that comes with Dupi. So I'm con!dent in terms of our design addressing a broader population, achieving e"cacy and continuing to underscore the role of eosinophilia and COPD, which is actually something that's important to stress

when we look at the latest date for Dupi, that, if anything, just increases my con!dence that this the eosinophilia or is an approach in COPD is important. I'll pause there and let Luke add what he wants to in COPD before we go on and talk about the magnitude of clinical bene!t in the CALM studies.

## Luke Miels Executive

Yes. I mean the only thing I would add is peak patient population at the time of expiries is between -- well, it's over 100,000 and it's under 250,000 because of emphysema, EOS cuto\$, et cetera. So it's a speci!c population, but that's the number range you should model for COPD.

## Tony Wood Executive

Yes, Richard. And then in terms of magnitude of clinical e\$ect and what's deemed as signi!cant relative to placebo in a refractory chronic comp. There are a couple of features that I just want to reemphasize in terms of our design that I think play into the reality of e"cacy. So !rst of all, let me go back to the 3 features that I mentioned earlier in the presentation. It's important to recognize that cough is quite a heterogeneous disease.

And therefore, our focus to an independent committee adjudicating the entry of individuals as presenting with refractory chronic cough is important. The placebo running that looks at the trajectory of baseline e\$ect is also important. And then lastly, as I mentioned, we've recruited into CALM 1 and CALM 2 in a way in which we're enriching for those individuals for which we think the mechanism is likely to have the greatest therapeutic e\$ect. So with all of that together and then I'd just direct you back to the e"cacy that we saw in the Phase II SOOTHE study, which is indicative of a higher reduction in chronic cough frequency than is the case in the general population. So this is a matter of a better study design, a better molecule and ultimately, that allowing us to target it at a likely more responsive patient population who will see greater clinical e\$ect.

#### Nick Stone Executive

Thanks, Tony. We take our next question from Andrew Baum at Citi. Over to you, Andrew.

### Andrew Baum Analyst

A question for Luke in relation to depemokimab. Given the method of administration, it's obviously going to be done by the clinician, could you just remind us for Nucala and the incumbent competitors. What percentage is self-administered versus physicianadministered? And given this will require a buy and bill dynamic with the clinicians, how problematic is that in transitioning patients from one to the other?

#### Luke Miels Executive

So Andrew, for self-administration, Dupi is 100%, which is, I guess, expected. Fasenra based on our !gures, it's about 80% self-administered. Nucala is 73% on the latest numbers we have and then [indiscernible] 55%. But of course, the self admin, I think from [indiscernible] came in February. In terms of moving patients back and forth, that certainly in#uences some of the assumptions that we make naive versus switch patients.

If you look at the total insurance spread linked to that, we think it peak about 1/3 of these patients, just over 1/3 of these patients are going to be commercial. So that Part B, Part D dynamic is important. But I think the bulk of the business for this product ultimately will come from Part B, B for Bravo. And we've got some thoughts there around access and making that very compelling for the physician and the patients as well.

## Nick Stone Executive

Next question is going to come from Je\$eries and in particular, Peter Welford.

## Peter Welford Analyst

So 2 questions. Firstly, just on Nucala. I'm curious when you look at the graph you showed of the peak sales potential, I guess, given everything you've said, is the 0.5 billion to 1 billion peak in COPD, which is on risk-adjusted, is that conservatism on your part? Or is this something we should be bear in mind when we consider why, I guess, given the size of the population you just talked about, you see that as an eligible COPD peak sales for the drug. And I guess, related to that, what have you seen for Nucala to do with potential either biosimilars or I guess, possible IRA impact?

And then secondly, on camlipixant, just on the Phase III, I guess to do with the trial design. Firstly, the unexplained chronic cough element, I think another thing that was introduced by the FDA at the advisory committee was whether or not potentially by including unexplained cough your sort of high, I guess, an undiagnosed condition. Is -- could the study potentially be analyzed in the separate populations, respectively. So refractory versus unexplained or any thoughts, I guess, is how you can potentially aside the FDA's concerns on that issue?

### Nick Stone Executive

So !rst question for you, Luke. And then second question for you, Tony.

### Luke Miels Executive

So Peter, based on our intelligence, I mean, there's 2 biosimilars in the clinic very early stages in China. I think there's one in preclinical and outside of China. So our modeling is we would not expect to see a biosimilar before 2031. So there will be some IRA e\$ects that we've modeled in that number. Our numbers that we give you are always at the midpoint, so by de!nition, conservative.

In terms of COPD, yes, I mean, it's interesting. When you look at the IL-5 class, it's a bit of a coin toss, but when it comes to physicians favorability between Fasenra and Nucala, it's about 1/4 of the HCPs love Nucala and use it extensively and don't use Fasenra. And it's about 1/4 of the doctors that use Fasenra and don't use Nucala as much. But the other 50% are agnostic. They like the mechanism.

They use both products. So -- there is -- you don't have this embedded group. So that's why we think when you look at the market research there, it lines up with about 25% of the business will come from Nucala. We think because Nucala is going to obviously launch in COPD earlier, we'll pick up those patients. There is a segment of patients that physicians don't want to move and there's also patients that say, look, I'm stable, don't move me.

But the other thing, as I mentioned earlier, is it's a lot more dynamic than we think. And we lose 6 out of 10 patients, as I said earlier, at 12 months. So these patients are moving back and forth between options or out of the biologic class and back into the biologics class. In terms of depemokimab and IRA impact, we've got pretty clear sense that payers will see this as di\$erentiated. And following the core clinical program, our intention is to be very aggressive with the real-world evidence program.

We've got some great data from Nucala, more coming up. It clearly shows, as you'd expect, a correlation with adherence and e"cacy. And we also know that lower shots lead to higher adherence. So if you connect those components up, we should see better results in the real world and working with payers in the U.S.

## Tony Wood Executive

Peter, let me -- yes, it was an interesting comment, I thought, because look, there are 2 ways to look at unexplained. The !rst one is an individual who has not gone through a full diagnosis cascade to identify underlying pathology. As you can see from our clinical study design, there is an independent adjudication committee that is addressing that particular issue with regards to entrants for the CALM 1 and CALM 2 studies. For those who have been triaged and for which there is no then determined underlying e\$ect I think you're in the simple position of then re#ecting that camlipixant is the best opportunity they have ahead of them to reduce remaining cough. I mean indeed, one might look at this as the emergence of a cough class that we might call the neuropathic co\$ee, which the origins of the enhanced coughing frequency associated with a change in the distribution of receptors in the upper respiratory tract.

So in terms of that aspect of unexplained which we can deal with, its accounted for in the Phase III design. In terms of that, which remains of unknown origin showing that we can reduce coughing frequency in that class with an agent that is addressing a particular aspect of potential neuropathic pain, I would argue is an appropriate approach to those folks.

## Nick Stone Executive

Thanks, Tony. [Operator Instructions] Our next question will come from Seamus Fernandez of Guggenheim. So over to you, please, Seamus.

## Unknown Analyst Analyst

So just wanted to go through the opportunity in chronic cough more from an access and payer perspective and how you're thinking about the pricing that would be realistic in the market. I know you're unlikely to talk about price, but maybe just help us understand if you see this as a product that would fall above the specialty tier in terms of pricing or below that relative to the Medicare pricing dynamics. And then also uptake in new areas like this tends to be quite slow from an access perspective. Just hoping to get a better understanding of how you see the launch of this product evolving over time. I see the opportunity from an upside perspective but the pace of uptake, I think, is something that everybody would like to try and understand a little bit better.

## Luke Miels Executive

Yes, sure, Seamus. Yes. I mean the typical patient, about 3/4 of them are commercial patient. So in our models, we -- and you have a lot of insight from this, obviously, from Nucala and Trelegy. So we assume classical abandonment rates of about 1/4 to 1/3 of patients at that point, and I'll get to the pricing range in a second.

Adherence, you should model more than 70% adherence, we believe, based on the market research because this lines up with classical high morbidity diseases, and there's lots of analogues that we use there, ranging from rheumatoid arthritis, complicated diabetes, et cetera. So yes, you've got those commercial patients. We can assist them with co-pays. Now in terms of the pricing range, we are very deliberately pricing below that threshold. So the price is somewhere between 600 and 800 in our models, which our market research and we've done a lot of that with payers.

We've spoken to over 45 payers. It's probably more than that now. Indicates that they are very receptive to the introduction of this because, of course, the unmet need is very high. In terms of the adoption, all the market research from pulmonologists and allergists, again, these are the primary drivers for refractory chronic cough are very clear that they will adopt it and adopt it quickly. We know these individuals very well from our existing business.

The other intriguing thing is we've looked at analogues like triptans with migraine, where you've got unmet need, pretty nonexistent standard of care, and you see very rapid uptake. So acceptable price range, established population. The other thing when you look at the patient populations, we are really looking at the cohort of people who have been coughing for more than 1 year and who are being managed by specialists. So that is under 2 million in the U.S. and similar numbers in Europe.

So when you add all of these things up as well as the other !gures I've given you, that's how you end up in the range of numbers that we've disclosed with the peak forecast. And then ultimately, we've also modeled that there may be 1 or 2 other novel classes in there, but we would expect that the P2X3s get at least 60% of these severe patients.

### Nick Stone Executive

Thanks for that comprehensive answer. I'm going to take our next question from Kerry Holford at Berenberg. So Kerry, over to you, please?

# Kerry Holford Analyst

Thank you, Nick. I have 2 questions, please. Firstly, on depemokimab. Given the high emergency you've discussed, do you ultimately expect bad drug to deliver higher e"cacy and Nucala and Fasenra? Or should we really be thinking in that more convenient dosing regimen as a key di\$erentiator to ultimately improving adherence and results in that way?

And then secondly, I guess my question is kind of on replacement power. So today, you've discussed what we see as a solid late-stage pipeline in respiratory. But I would suggest in contrast that early to mid-stage pipeline in this disease area at GSK looks relatively surpass. So I wonder if you would perhaps disagree whether there were any Phase I, Phase II

respiratory buydown assets. So we should keep an eye on, might be accelerated GSK alternatively, should we be thinking of respiratory as a key area for your business development team considering external opportunities.

#### Tony Wood Executive

Nick, do you want me to get started on that?

### Nick Stone Executive

Please, Tony.

### Tony Wood Executive

So let me just pick that apart if I may, because there's the -- let's start with depe relative to Nucala in the !rst instance. I think the potential advantage on top of the dosing frequency that Luke described is really about, I think, compliance. And if you like, we're moving to a point where a sense of freedom from the disease is going to be important for these patients and you saw that to a degree in the video. I do want to just draw out a distinction because we haven't had a chance to talk about this yet between Nucala and Fasenra. The way I look at the biology here in general and I'll come on and answer your second question about portfolio and BD in a minute in the context of this.

As you can divide it into T2 and low T2 biology. T2 biology is typically that associated with eosinophilia. And in T2 biology, you get to take 2 options at it. You either go for the cytokine IL-5 or you go for the receptor. So this is the Nucala depe versus Fasenra comparison.

And what seems to be the case in this instance that we're learning from the pro!le that we're developing for Nucala that there is advantage associated with targeting the cytokine because of the deeper tissue e\$ects.

And you see that across a range of outcomes when you compare Nucala and Fasenra not only in with regards to example, OCS bearing status in asthma, but also with regards to e"cacy in broader eosinophilia. So I think we have an advantage in the side of the access that we've chosen on IL-5 versus the IL-5 receptor. And to your point and bringing in a longer acting agent is largely going to be associated with compliance. With regards to the pipeline itself, we have a number of earlier assets coming through that in time we'll be able to talk about. I want to stress that we really have focused our e\$orts.

And you can see we've got a pretty active Phase III pipeline, developing a better understanding of the role of IL-5 and T2 disease in moving the treatment paradigm on for severe eosinophilic asthma from symptom reduction to exacerbation reduction and ultimately to clinical remission.

If you then ask -- and one might imagine that you bookend that, for example, in other areas of the biology, we have TG2 molecule that is in Phase I that will allow us to begin to ask questions about the role of development of !brosis building on top of T2 biology, alarmings as the other axis of the important biology in lung disease, either from IL-33 or from BD activities in the alarm and class. I won't describe that in more detail for obvious reasons also become part of our focus. So I would say we have a comprehensive program aimed at

positioning new mechanisms in the context of what we're learning about IL-5 in the clinic. And therefore, where the most important medical need and opportunity would sit and follow-on agents.

#### Nick Stone Executive

Thanks, Tony. We take our next question from Graham Parry, please, at Bank of America. Graham, over to you.

### Graham Parry Analyst

So just a quick question on camlipixant. So you talked about the enriched patient population in your Phase III study relative to gefapixant. And obviously, one of the key things the FDA was saying they were struggling to see a clinically meaningful reduction. So if you can just help us understand the di\$erences in the patient population in your Phase III versus the gefapixant Phase III and then the extent to which that then also reduces the addressable patient population from a commercial point of view. And then secondly, on the Nucala depe study, gefapixant has obviously had a very high bar here, 30 reduction, but in a narrower patient population arguably.

Would you sort of characterize your expectations around the Carla's Phase III study as being that the inclusion of more emphysema patients broaden your addressable population, but perhaps reduces the e\$ect size you're going to see in the study.

#### Nick Stone Executive

Tony, do you want to take the !rst part of camli, and then Luke and then over to the Carla question.

#### Tony Wood Executive

Yes, let's do that. And Graham, I'm not going to get into the details of the Phase III design. That's something that is still subject to broader conversation with regulators and an aspect of our positioning of the agent and I don't want to share more broadly. But if I can just bring you back to the points that I made earlier. The emphasis for achieving greater treatment e\$ect comes from 2 broad approaches.

The !rst one is the trial design. That's very much around ensuring that we are admitting those individuals who are more likely to have frequent cough associated with neuropathic origins playing to the mechanism.

The higher the frequency of cough, the greater e\$ect on is likely to see, and we will look to enrich our patient population with that in mind. We're also taking account of the fact that there is clearly a placebo e\$ect here in the greater extent to which one understands the trajectory of that moving into the study, the greater opportunity to show a di\$erence. And then lastly, it's probably worth emphasizing because I didn't when I answered the question earlier that the vastly improved selectivity of dysgeusia will also help confounding teachers associated with unblinding. But at this point, Graham, I wouldn't want to go any further than that. Perhaps the only other thing to stress in terms of the regulatory feedback.

There was also a commentary on the relevance of the patient-reported outcomes and we have engaged closely with the regulator on that, particularly with regards to the Leicester Cough Questionnaire and you'll see that accounted for in the relative size of the 2 Phase III populations. And then Graham, look, you might have to remind me the questions, Graham. Sorry, sorry, go ahead, Luke.

# Luke Miels Executive

Yes. No, thanks, Tony. So Graham, I mean all of this was largely visible during the due diligence process with BELLUS as well because there obviously has been an early attempt at approval with gefapixant. And as Tony said, I mean, if you look at the selectivity, camlipixant is around 1,500 fold more selective for P2X3 in contrast to gefapixant, which is 3x to 8x. So -- we - when we did the modeling and the forecast that you see and the background to the deal, the same way we did with Sierra, we take a conservative assumption in the population and the label -- and so the numbers that you see exclude these milder and non-neuropathic coughing populations.

And it's very much a severe end of the population with established extensive disease for more than a year.

And that's what we've modeled it on. Just back to Seamus' question as well, I was just reminded of another analogues that we looked at, which is obviously the [indiscernible] analogue where you saw within 3 years, it was around 60% of those patients were being treated with a novel agent. So similar prescriber base, high unmet need, di"cult disease, et cetera.

### Nick Stone Executive

Just going back to yes, the question was around Nucala inclusion criteria, it's only for COPD in terms of [indiscernible] and whether it was a narrow patient population.

# Tony Wood Executive

Okay. Super. Let me just make one more point on the PROs for camlipixant as well. They're based on what we've learned from the AdCom as well. So the di\$erent PRO set as well as a trial design to enable a more extensive demonstration of patient bene!t associated with those.

In terms of Nucala and the COPD inclusion, you can think about this as follows. So we have a broader population, Graham, you're right, that is aiming for a broader label. It brings a more di"cult-to-treat population in the emphysemic. But what we've also done has been careful to enrich that study with regards to eosinophil count and to be sure that we've gone for a pure population of individuals. So the both -- the broader population, if you like, the tougher choices associated with that has been countered by a more careful selection on the basis of eosinophils in COPD.

And we've shown, Graham, that high eosinophilia is associated with poor outcomes in that COPD population. So we're con!dent based on the data that we've got in that more general population. And indeed, as I mentioned earlier, the Phase II data in a more bronchitic highyield population that we have at least equivalent e"cacy to dupi.

#### Nick Stone Executive

We've got 2 last questions. I'm going to come to Ben Jackson at Je\$eries !rst and then I'll !nalize with Emmanuel Papadakis at Deutsche Bank. But Ben over to you in the !rst instance.

### Unknown Analyst Analyst

Just on camlipixant, aware of the comments that you just made about trial design, but I was wondering if there's any further thoughts on the speci!c comments made by the AdCom with regards to cough clusters as well as the frequency of cough there. And then, I guess, o\$ the back of that, are there any thoughts longer term about potentially additional follow-on studies or additional indications that could be gone into from here in the same way that Merck have looked at developing their asset as well?

### Tony Wood Executive

Yes. I'll keep it quick, Ben. And Luke, if you want to pick up after the additional indication comment. I'll make it a quick one. The way to think about cough clusters is this is all about the analytical method applied to the cough counter.

We've been -- we've taken careful account of that in our interactions with the regulators. So I -- in terms of the analytical framework that will go forward. So I don't expect to see the desire to recalculate that resulted in one of the trial failures for Merck. And then with regards to additional indications, one can imagine a number of alternative sensory a\$erent based indications like, for example, urinary incontinence, which, of course, is not only a consequence of cough, but a feature of the target patient population here who are likely to be women in their 50s and more who tend to report with a refractory chronic cough. Luke, I'll hand over to you if you want to embellish on that.

#### Luke Miels Executive

Yes, de!nitely. And we didn't model those populations in the valuation of BELLUS. So these are all upside, but our intent to explore them, probably less enthusiastic around pain. And then in the chronic cough population, we're planning an extensive post-approval real-world evidence Phase IV program. So for example, is there an opportunity for primary care doctors to use this mechanism empirically when the patient presents, say, at 8 weeks with chronic cough before sending them o\$ to more invasive or more expensive investigations.

Even if it's just to attempt to provide some relief.

And so there's obviously enormous synergies there with our established infrastructure. And then the same way we want to do the same thing with depemokimab back to Kerry's question because as we've done with Nucala, there are opportunities in the rule that setting to demonstrate higher levels of e"cacy. And that in our modeling, we assume initially launching the target medicine pro!le and all the market research you've seen in these presentations assumes there's equivalent e"cacy, but we think there is an opportunity in a real worth setting and to publish this to assert the bene!ts of the pro!le of the drug, which could result in better outcomes for patients and a better e"cacy claim.

# Nick Stone Executive

Okay. Final question to Emmanuel Papadakis. Over to you.

# Emmanuel Papadakis Analyst

So a couple of quick follow-ups, if I may. IL-5s and COPD, given the rerun Fasenra study resolute and COPD is employing many of the same modi!cations that you've made in MATINEE, is there any reason why we shouldn't assume that will also work?

And have you assumed that in your peak sales assumptions? And then a couple of follow-ups on camlipixant. Tony, I didn't really understand your comments around the population. I mean, the trials are already running. So you're telling us you're discussing with the FDA what precisely will be the population included in the primary endpoint?

And then a connected question. Apologies if I missed this. Are you hoping to show or replicate the 30% approximately placebo-adjusted improvement you saw in the Phase II [indiscernible] study? Or would that only be in a subgroup of the population you enrolled in CALM 1 AND 2?

### Tony Wood Executive

Yes. Let me see if I can clarify that. And then so !rst of all, in terms of Fasenra and its likely pro!le in COPD, I'm back to the point I just made to Kerry that it's important to recognize that across the breadth of performance we see for IL-5 versus IL-5R, IL-5 seems to be the preferred mechanism. So I would suggest that if there is a deeper e\$ect associated with disease modi!cation that we might see an improved bene!t from Nucala. But we will see how that plays out.

In terms of the point that the studies are indeed already running, what I was talking to in that particular comment was the ongoing negotiations we've had with the FDA during the time period before that study design? And what we were keeping in mind that was particularly the point that you make with regards to the distribution of patients who are most likely to bene!t from the mechanism, those with higher cough, those adjudicated to have no underlying conditions. And as a consequence of that, aiming to be able to evaluate e"cacy across a range of di\$erent strati!ed populations. And then, of course, what we're aiming for there is a sort of placebo-adjusted e"cacy that you can see from the Phase II study. But I would expect to see a range across di\$erent populations all of that is accounted for in the CALM 1 and 2 design.

### Nick Stone Executive

Super Thanks, Tony. Thanks, Luke, especially for highlighting the signi!cant opportunities we have ahead of us clearly to all of those that have attended, thank you for your interest and attendance. If there's any follow-up questions, don't hesitate to e-mail us at irteam@gsk.com. And with that, we will now close the call. So you're good to disconnect.

Thank you very much.

Tony Wood Executive

Thanks, all.

# Luke Miels Executive

Thank you.