Amgen Inc. (AMGN) Earnings Call Transcript & Summary

My name is Erica, and I will be your conference facilitator today for Amgen's conference call from the American Academy of Allergy, Asthma & Immunology Annual Meeting. [Operator Instructions] I will now introduce Arvind Sood, Vice President of Investor Relations. Mr. Sood, you may begin.

Thank you, Erica. Good morning, and good afternoon, everybody. Thanks for joining us. So earlier this morning, we presented important data in a poster form on tezepelumab, and this was presented at the American Academy of Asthma, Allergy and Immunology Meeting. This data is from our Phase III NAVIGATOR study evaluating tezepelumab in severe asthma patients. To review this data in additional detail, I'm joined this morning by Dr. David Reese, our Executive Vice President of Research and Development; as well as 2 of our clinical investigators who Dave will introduce and who can provide a broader perspective on how they see tezepelumab fitting within the armamentarium of therapies to treat severe asthmatic patients. Dave will make a brief presentation reviewing the data, following which we should have plenty of time for Q&A. So with that, Dave, I'm going to turn the call over to you.

Well, thanks, Arvind, and hello, everyone. It's a pleasure on behalf of Amgen and AstraZeneca to review the data today. And as Arvind mentioned, we should have plenty of time for question and answers. For those of you who are following along in the associated presentation, if you turn to Slide 3, you'll see a brief agenda. As Arvind mentioned, I will provide an introduction on the disease state, just to make sure we're all level set on the patient population we are talking about today in the NAVIGATOR study, a brief summary of the data. And then we will have plenty of time for the Q&A session. We're very pleased today to be joined by Professor Andrew Menzies-Gow from the Royal Brompton Hospital and Imperial College London; and Professor Michael Wechsler from The Cohen Family Asthma Institute and National Jewish Health in Denver, Colorado. Both are international experts in this disease as well as principal investigators within the tezepelumab program. If you turn to Slide 4, I'd like to start with some background on the disease state, which should provide context on the patient population that we've studied in NAVIGATOR with tezepelumab. Asthma is primarily a disease of inflammation. It is increasing in prevalence on a global level driven by demographic factors such as urbanization and pollution as well as presumably a host of other factors. In the United States alone, more than 25 million people are affected by the disease. It is a complex heterogeneous disease biologically that can be triggered by multiple environmental factors, allergens, pollution, cigarette smoke and activates in downstream multiple inflammatory pathways. The clinical hallmark of the disease include cough, chest tightness and shortness of breath and exacerbations, which drive much of the health care burden of the disease, may require systemic or oral corticosteroids for treatment and may prompt visits to emergency departments and even hospitalization. Now the patient population that we're talking about today are those with severe and uncontrolled asthma, and severe asthma is typically defined operationally and clinically as disease that is -- that requires high-dose inhaled corticosteroids plus at least one other asthma controller medication. These patients may potentially also require oral corticosteroids either chronically or on an intermittent basis when flares of the disease occur. Despite available treatment, many patients have inadequate responses to or are not eligible for various reasons for current biologics and fail to achieve sufficient clinical control, suggesting that there remains significant unmet medical need in this disorder. If you move to Slide 5, it talks a little bit more about that unmet medical need, particularly in those with uncontrolled severe asthma. Uncontrolled asthma is defined as the persistence of asthma symptoms despite appropriate treatment. Patients with uncontrolled asthma may experience one or more daytime symptoms, may require the acute use of relievers more than a couple of times per week, often have -- waking at night due to symptoms. And in fact, many of these patients experience a significant sleep disruption. Of course, this then can be associated with limitations on daily activity. And as mentioned, those experiencing frequent exacerbations may require long-term systemic oral corticosteroids, which can be associated with various side effects. Finally, again, the disorder is a common cause of emergency department visits and hospitalizations globally. This is one of the principal drivers of costs and burdens on health care systems around the world. The disease can also, in some instances, be associated with mortality in patients who have uncontrolled disease with frequent exacerbations are at greatest risk of this outcome. Severe and controlled asthma accounts for probably on the order of half of all asthma-related costs. And as you see at the bottom of this slide, there are about 2.5 million patients in the U.S., the largest 5 EU countries and China and Japan with 1 million alone in the United States. Turning to Slide 6. Why do we believe tezepelumab can be a significant breakthrough for the treatment of this disorder? Where does it play a role in treatment? Well, I think this is driven by the fundamental biology and mechanism of action of the agent. Tezepelumab targets TSLP, or thymic stromal lymphopoietin. This is a cytokine that is derived from epithelial cells typically released at the time of tissue injury or inflammation. TSLP then recruits and activates a host of downstream signaling pathways involving IL-4, IL-5, IL-13 and Immunoglobulin E. So it is essentially an upstream master regulator of the inflammatory cascade. This really formed the biologic hypothesis as to why tezepelumab could be effective in a broad range of patients with severe and uncontrolled asthma. So let's now turn to the actual data that were presented from the NAVIGATOR study earlier today. If you turn to Slide 8 in the associated presentation, you will see the study design and the patient eligibility criteria. This was a standard population of patients with severe uncontrolled asthma. The population consisted of adolescents and adults ages 12 to 80 who are nonsmokers. They had uncontrolled asthma as defined by scores on the asthma control questionnaire, and this is fairly typical for a patient population with severe uncontrolled asthma. They had to be on medium or high-dose inhaled corticosteroids plus at least one other controller medication. And as you see, about 3/4, in fact, were on high-dose inhaled corticosteroids. Patients also had to have experienced at least 2 exacerbations in the prior year, and 40% had experienced more than 2 exacerbation. The study design was quite straightforward. Patients remained on background standard therapy and then were randomized to receive either tezepelumab every 4 weeks or placebo every 4 weeks. Tezepelumab at a dose of 210 milligrams. Just over 1,000 patients, 1,059 patients were randomized in the NAVIGATOR study. Turning to Slide 9. You see a summary of the data. The trial met all of the primary and key secondary endpoints. We're actually thrilled with the clinical data that we observed in the NAVIGATOR trial. The primary endpoint was the annualized asthma exacerbation rate, or AAER, over a year's worth of therapy. And as you can see, in the overall patient population, there was a 56% reduction in the AAER and a 41% reduction in patients with baseline blood eosinophil counts of less than 300. So this is a patient population for which there is, in particular, significant unmet medical need. The trial also met all secondary endpoints, measured physiologic parameters, quality of life, symptomatology. Specifically, it improved the pre-bronchodilator fractional exhaled volume on 1 second at week 52 by 130 milliliters. This is considered a robust and important clinical improvement compared to placebo. And then as you see here, using various standard scores, such as the ACQ-6 and others, there were improvements in quality of life and symptomatology. Additionally, the AAER was reduced irrespective of baseline eosinophil count and FeNO level. FeNO is the fractional exhaled nitrous oxide level. It's a marker of tissue inflammation in asthma or allergic status. And then finally and I think very importantly, from a health care system perspective, the AAER rate associated with hospitalizations or emergency department visits was reduced by nearly 80%, a significant finding. On Slide 10, you see a graphical representation of the primary endpoint, again, showing a 56% reduction in the AAER in the all-comers population and a 41% reduction in those with baseline eosinophil counts less than 300. On Slide 11 is displayed a forest plot. And this shows the general consistency of results across different groups of patients defined by various biomarkers, such as different eosinophil count cut points; FeNO levels are again a measure of inflammation; or Immunoglobulin E status, 2 specific perennial allergens. This is a marker of the allergic phenotype. And as you can see in all of these groups, there were significant reductions in the AAER favoring the tezepelumab group. On Slide 12 is an illustration, graphical illustration of the improvement in FEV1. I think a couple of important points to note here. First, improvements were noted as early as 2 weeks and were sustained in general over the course of the year-long treatment period. On Slide 13, turning to safety. I would say there's nothing remarkable. In terms of the safety profile, we anticipate an upcoming publication in the not-too-distant future where there will be detailed safety tables. There were no meaningful differences in adverse events between treatment groups, including serious infections. And as you can see here, in fact, the rate of serious adverse events was higher in the placebo group driven primarily by an excess of asthma-related adverse events. There were no cases of Guillain-Barre syndrome or injectio- associated anaphylactic reactions. So in summary, on Slide 14, we believe tezepelumab provides a potential differentiated first-in-class investigational therapy for patients with severe uncontrolled asthma. It blocks TSLP, the first epithelial drive cytokine that has been targeted. And in NAVIGATOR, we demonstrated clinically meaningful reductions in exacerbations, regardless of patient groupings or biomarker segregation. This was associated with significant improvements in lung function, asthma control and various quality of life and symptomatology measures and, again, importantly, from a health care system point of view, reductions in exacerbations associated with hospitalization or emergency department visits. No meaningful differences in safety. And these data, in combination with the prior Phase I and Phase II data, will form the basis for regulatory submissions in the U.S. and EU that are planned in the coming months in the first half of this year. This is not the end of the development program for tezepelumab. A Phase II study is currently enrolling in chronic obstructive pulmonary disease, and we plan to initiate with our partners a study in chronic spontaneous urticaria in the coming months. As you know, this is a partnership between Amgen and AstraZeneca. I can say on a personal level, this partnership has been functioning very well, and we're looking forward to making this medicine available to patients with severe uncontrolled asthma. So with that, I will finish the prepared remarks, and we will move into questions and answers. We should have plenty of time here. Erica, if you can just remind everyone again and then tee up the queue for Q&A.

Erica, before you open it up for Q&A, I would just like to remind all our listeners that the slides are available on our website. And if you want to download the poster, the poster is also available on the AAAAI website. So with that, Erica, why don't you go ahead and review the instructions for asking questions, please?

[Operator Instructions] Your first question comes from Kennen MacKay with RBC Capital Markets.

Again, just coming back to really the potential positioning for tezepelumab in the market. I'd love your perspective on sort of where the most unmet needs, especially as it relates to eosinophil counts and from the presentations, how you're seeing differentiation versus Dupixent and any of the other biologics that are already out there.

Great. Thanks, Ken. And I'll just take the last part briefly. I mean we think that based on the data that we presented across the range of patients regardless of eosinophil count or other biomarker categorization, this is a potential first-line therapy for a broad range of patients. But I think it would be great to get Professor Menzies-Gow and Professor Wechsler to lay in here on how they think about this in the unmet medical need, how you would think about the potential use of tezepelumab in the clinic. So perhaps, Professor Menzies-Gow, we can start with you and then we'll move to Professor Wechsler after that.

Of course, and thank you. That's clearly an incredibly important question. I think it's important we remember that although we have, depending on where we are in the world, up to 5 different biologics, that severe asthma isn't fixed and not everyone with a high peripheral blood eosinophil count responds to one of the currently available biologics. And actually, most people with severe asthma have multiple drivers for their symptoms. So it's unusual to have a purely eosinophilic patient with absolutely no IL-4 and 13 signal. And many of our patients overlap with a component of allergy, a component to the eosinophilia and a component of IL-4 and 13-driven disease. So it is logical that upstream epithelial drugs can actually target multiple. And I really wouldn't want for people to think that tezepelumab is going to be positioned only in those late patients because, clearly, there's huge unmet need there and there are no other alternative therapies. But when we look at the impact of tezepelumab on the annualized asthma exacerbation rates, lung function and the PROs, you also see a very effective improvement in those these patients with a higher blood eosinophil count as well. So I suspect if you're asking me as a clinician based on this data, over time, we could see tezepelumab as first line for many, clearly not all because there's no biologic that will be the optimal choice for everyone, but for many patients, I could see this being the first-line choice.

Great. And Professor Wechsler, your point of view?

Yes. So I'm really excited about these data, and I find that this therapy, tezepelumab, has broad efficacy. You asked you about the biggest unmet needs. I think there's -- I think severe asthma, in general, remains a big unmet need. Many of the biologics that we have, they reduce exacerbations by 50% to 60%. But there's still that 40% to 50% of exacerbations that remain not addressed. Many of those are non-type 2 exacerbations brought forth by viruses and nonallergic phenomenon. One of the important things to remember about asthma is how dynamic it is and that there are many different contributors, whether it's allergic or nonallergic, viral, irritants. And we currently don't have any therapies that are targeted for the nonallergic, non-type 2, non-eosinophilic patients. So this would certainly fill that void. And I think it would complement the other biologics that we have in terms of having an additive effect. So I think there's broad efficacy in both the eosinophilic as well as the non-eosinophilic patients. It could be the first-line biologic therapy because it is so broad, and it works upstream of the other cytokines that we target, interleukin-4, interleukin-5 and interleukin-13. So I think we'll see how it gets taken up in the -- in clinical practice. But there's very good reason to believe that it should be an optimal therapy for patients with severe asthma.

Our next question is from Robyn Karnauskas with Truist Securities.

I guess big picture, you mentioned synergy or -- with different biologics out there. Obviously, have you started experimenting with that in the clinic at all at the moment? And then second, like when you think about positioning in your clinic, how does -- what is the impact of not being able to get rid of steroids? I think you had a trial that did not work or in a steroid-sparing trial in severe patients. How does that influence your -- the appeal of this drug? And how do you incorporate that data point into practice?

Yes. Robyn, perhaps I'll address briefly the first part. In terms of potential combinations, that's something that's under consideration. We talked a little bit more in terms of thinking about that over time. In terms of the SOURCE trial, the oral corticosteroid sparing trial, we'll ask our investigators to comment on that in a moment and how they think about this. I would note that in this trial, the patients -- less than 10% of patients required oral -- chronic oral corticosteroids. That rate does seem to be declining perhaps because of the widespread availability of biologics. It'd be great to get our clinicians' point of view here on this point. So perhaps Dr. Wechsler, we'll start with you first this time, and then we'll go to Professor Menzies-Gow.

Yes. So the top line results of the SOURCE study have been presented and do not meet its primary endpoint. We're going to have further data analyses. And at the time of presentation and publication, we'll share more data in that regard. I will say that, obviously, patients on oral corticosteroids represent a huge unmet need. Oral corticosteroids used chronically, it was associated with significant morbidity. Patients who are on oral corticosteroids can often fraught with complications like cataracts, glaucoma, diabetes, osteoporosis. So we want to try to minimize the amount of oral corticosteroids a person gets and certainly for patients with severe asthma. There have been concerns that the study design or the patient population that was tested in the SOURCE study may not have been adequate to tease out those differences. It's very hard to do those studies in the U.S. I can say that there are not as many patients who are oral corticosteroid dependent. Part of that is because we've had other biologics to the low-hanging fruit of patients who would be most likely to respond, have probably been treated with other therapies. That being said, I think I would still utilize this therapy in patients with -- on oral corticosteroids because of its broad effects and the fact that it seems very effective in patients who are not on oral corticosteroids. If you remember that in those kinds of studies, there can be -- a lot of patients tend to be overprescribed. And so as you taper down, there's going to be a strong response in the placebo group as well. And so people who were on 20, 25 milligrams a day, they probably don't need that much. And so it's fairly easy to get those people down. So it's harder to see an effect compared to drug in those patients.

Andy, what are your thoughts?

Thanks, Mike. Listen, I absolutely agree with everything you said. And to sort of put this into context, it's very clear now with all our international guidelines, including GINA, the Global Initiative for Asthma. First, oral corticosteroid should be seen as drug of last resort. So they really should be in that post-biologic space. So I think whenever possible, we should be taking [indiscernible] asthmatics that are frequently exacerbating and treating them with biologic then before we ever get to the point of being on chronic OCS. Now clearly, there are historical patients and that there are various other tracking primary care that aren't necessarily getting towards a biologic. So Mike has really very well described where we are with SOURCE. And we cannot go into any more detail around that. But the reality is this is a drug that acts upstream, that acts on all of the cytokines knocked down by oral corticosteroids. So I see no reason why I wouldn't want to use this drug in a patient who is exacerbating frequently irrespective of their baseline and prednisolone prescribing.

Your next question is from Alethia Young with Cantor Fitzgerald.

And congrats on the very good data. I guess maybe a little bit of a simple question. With such good data, I'm curious if you're going to look to go maybe into moderate patients. And then secondarily, just how do you have the committee between AstraZeneca think about the framework and broadening out indications? I know you named 2. But I guess, kind of how does that framework work? Because if the data are interesting here, it feels like there's a lot of opportunity set being so far upstream.

Yes. Two very important questions, Alethia. Obviously, these data open up questions about potentially broadening the patient population within asthma over time. And that's something that we're certainly working with Professor Menzies-Gow and Professor Wechsler and others on thinking about how to most appropriately do that. And then the question is given the data we've seen here, upcoming data that we'll show you over the course of the year, other indications, particularly those that are eosinophil-driven, it may be served by tezepelumab. And so as I mentioned, chronic spontaneous urticaria is a study that we will be launching in the coming months. We are discussing with our partners at AstraZeneca and we work very, very closely in a quite integrated fashion what additional life cycle management would be appropriate across a range of indications. I would say more to come on that over the course of the year.

Your next question is from Yaron Werber with Cowen.

Congrats on the data. And I have a couple of questions. Maybe the first one, can -- it sounds like there's going to be a publication, so we'll get more insights into the safety. Can you comment, are you seeing any even small nonphysical imbalance on rates of new cancers? And then maybe to the 2 KOLs, any thought -- when you look here, the hazard ratio and the high eos, it's actually 0.3. So it's a 70% reduction, which seems to be the best so far of all the biologics. Any thoughts about that? Does that potentially make it the first-line biologic now?

Yes. Thanks, Yaron. And in terms of the adverse events and numerical difference, well, I can't go into detail there prior to the publication. And we were, of course, just limited by the moderated poster format, which is the format for late-breaking abstracts at the AAAAI meeting. But you'll see detailed data tables. We're very, very comfortable with the tolerability and adverse event profile that we've seen so far. And let me ask, again, our clinicians to address the second part of your question there. Professor Menzies-Gow, if you wanted to comment on that, and I think it was directed, in particular, in the high eosinophil population and the roughly 70% reduction observed there.

Yes. No, thank you. Very important question. I think, first of all, always difficult to compare directly these Phase III pivotal head-to-head studies because the inclusion criteria are always slightly different and they're done at historically different points in time. But that being said, the exacerbation reduction with patients with a high eos is very impressive. And when we look, there's a significantly stronger signal, as we would expect, in the placebo arm in patients with a higher eosinophil count as that does tend to predict increased risk exacerbation. But that is very effectively knocked down by tezepelumab. Whether or not it is best in class, I think the reasons I previously described, I think that's hard to know. But it's certainly as good as any of the alternative agents there, which is why my previous answer was I would consider this to be a potential first-line therapy in high eos patients. I'll hand over to Mike.

Professor Wechsler, anything you'd like to add?

Yes. I agree. I think it's really hard to evaluate these. Ideally, there would be a head-to-head study comparing different biologics in the same kind of trial. That would be the ideal way. It would be an expensive study to do, but I think that it would be something that would be ideal. I'm confident that this therapy would be effective in people with high eosinophils. I'm also confident the other biologics that target eosinophils would be effective and perhaps as effective in that patient population. The prior question that came out also asked about moderate asthma. I think given the price points of the biologics in general, it's unlikely to filter down to people with less severe disease unless the price point for these drugs came down a little bit.

Our next question is from Terence Flynn with Goldman Sachs.

This is Dan on for Terence. I was just wondering if you could share any preliminary perspective on your pricing and reimbursement and commercial strategy.

Yes. Thanks, Dan. We're not ready to talk about that yet. Obviously, over the course of the year when the drug is approved, we'll talk about that. But I think it's a little too early, and that's something we'll, of course, be discussing in conjunction with our partners, AstraZeneca. So more to come there, but I think a little too early to discuss that.

Your next question is from Geoffrey Porges with SVB Leerink.

This is Na Sun on for Geoff. Congratulations on the data, it looks very good. So you guys are planning to file for approval in the first half of 2021. Will we get detailed breakdown from the SOURCE prior to that? And how do you think the data package from SOURCE and NAVIGATOR will play into the -- to the label when you file for approval? And then another one for the KOL. You mentioned that potential for this to be a first line. Do you think the data is strong enough that you would persuade patients currently on, say, Dupixent to switch to tezepelumab? Or would you only -- in the patients currently on Dupixent, would you only replace it as a -- after they fail off of Dupixent, you would put them on to tezepelumab?

Great. Thanks, and I'll take the first part and then, of course, again, we'll get our clinicians involved. So as Professor Wechsler mentioned, we anticipate full presentation of the SOURCE data at a meeting and meetings later this spring. So more to come there, but you should get the detailed data. How all of this plays into the label, of course, is the source of regulatory discussions. But based on what we've announced publicly, we would not expect, of course, any sort of indication for oral corticosteroid lowering. The switching question is an important one. And so let's perhaps ask Professor Wechsler first and then Professor Menzies-Gow to talk about how they would approach that sort of question in the clinic. So Professor Wechsler?

Sorry, if a patient is doing well on a current therapy, whether it's dupilumab or benralizumab or mepolizumab or reslizumab, I don't see a reason why I would switch. The issue is if they're not well controlled on whatever biologic they're on, then I would consider a switch. Ideally, if costs weren't an issue and if we could validate safety, it would be nice to know whether or not adding biologics one to the other would be effective, but there have not been any studies looking at combinations of biologics.

Great. And Professor Menzies-Gow?

Yes. No, thank you. So I agree with Mike. When someone's doing very well on a biologic, I certainly wouldn't switch. However, we think or we know that the number of people that do switch is less than it should be. So when we looked across the international severe asthma registry, approximately 80% of people stay on the first biologic that they're prescribed. So I think it's more a discussion and a communication within the severe asthma community of appropriately setting the bar of response, so we don't accept mediocre responses or any people who have had a super responsive response or extremely good response should stay on a biologic and not at least consider switching. So I think there is an opportunity to increase the amount of switching that occurs to ensure that people are on the best biologic for them. But I certainly wouldn't consider switching if someone was doing very well on their initial therapy.

Your next question is from Matthew Harrison with Morgan Stanley.

Great. I guess I was just hoping to understand a little bit better. When you think about new patients and positioning, now you have 3 or 4 biologics, and I think we understand what will happen in low eosinophil count patients. But in patients who have high eosinophil counts, how are you going to weigh the different agents? And where do you think teze fits within that -- in that group?

Great. Yes. Why don't we go straight to our clinicians in terms of how they would approach patients and potentially new patients. So Professor Wechsler, perhaps you want to start on this one?

Yes. So I look at a host of factors when choosing a biologic for a patient. It's not just eosinophil count. It's other biomarkers, including exhaled nitric oxide, including IgE. I also look at other comorbidities that a patient has, whether they have chronic rhinosinusitis, whether they have atopic dermatitis, whether they have urticaria and others. I also look at their insurance profile, insurance coverage and then patient preference with regard to -- with patient preference with regard to frequency of administration, location of administration and other patient-related factors. So there's a lot of things that I consider. And I generally look at the dominant biomarker, if there is one. And now for a drug like tezepelumab, which seems to work across the board in people with high eosinophils and low eosinophils in people with higher nitric oxide levels and lower nitric oxide levels, I would think that teze would be a very good candidate for first-line therapy for my patients with severe asthma. Again, you have to -- you want to try to personalize your approach as much as you can. And everyone's asthma is a little bit different. Everyone's circumstances are a little bit different. But this drug seems to check a lot of boxes in terms of where I would like to target my patient's severe asthma.

Great. Thank you. Professor Menzies-Gow, anything you'd like to add?

Yes. No, thanks. I mean I agree with Mike. It is a combination of clinical history, the individual patient preference and, importantly, the biomarkers. And as I said earlier, it's quite unusual for someone to have such dominant disease with just the eosinophil driving the symptoms, there isn't a component of IL-4 and 13 and nitric oxide, et cetera. And in reality, what we do at the moment is we see the patient, we measure the biomarkers, and then we take our best guess, and that's all it is. And we do in any course 1 trial, assess it at 4 to 6 months and whether they continue or don't. And tezepelumab essentially derisks our approach because we know irrespective of the baseline biomarker count that patients are likely to respond. So I can see for many, but not for all, this will be certainly the logical first-line therapy.

Your next question is from Evan Seigerman with Credit Suisse.

And Arvind, thank you for hosting this call at 1 p.m. and not 8 p.m. But with that, can you guys expand on the lung function improvement you saw by subgroup? Or was the data consistent across subgroups? Or was there some variation that could impact ultimate use of tezepelumab?

Yes. I mean I think you're referring to the changes in FEV1, for example. So yes, we'll get our clinicians to weigh in and in different patient populations. I would note, by way of preparatory comments, that in the very low eosinophil population, for example, there's evidence of biologically the disease may be a little bit different there in terms of things like airway reversibility, more airway remodeling, inflammation that's not steroid-responsive, but it'd be great to have our clinicians weigh in. Why don't we start out with Professor Menzies-Gow this time on the lung function question?

No, of course, a very important question. First of all, clearly, the subgroup for pre-bronchodilator FEV1 isn't in the poster. So we cannot discuss the detail today. It will be in the publication when that becomes available. But 2 points with regard to FEV1 study. I was genuinely surprised how quickly it improved. I presumed given that tezepelumab was an upstream cytokine, that it would take a while just to start to work a bit, like omalizumab takes a while to see any improvement. So I was very pleased to see the early improvement. But then when we talk about the clinical characteristics of these patients, they are very different. And frequently, patients who are highly symptomatic with a lower eosinophil count may be on maximum or even beyond maximum anti-inflammatory, either oral or inhaled corticosteroids, because they have no type 2 signal left. So it's not surprising if you're maximizing out that treatment that is harder to see an improvement in some of these outcome measures than in patients who clearly still have ongoing type-2 inflammation. I'll hand over to Mike.

Right. Go ahead, Mike.

Yes. No. So I think the degree of improvement that was observed overall, I think, is impressive. The absolute improvement that was -- I think you can see from the slides, is around 200 cc, which is a very significant improvement. Compared to placebo, it was a little over 100. I think lung function is just one metric that we utilize to evaluate efficacy of a drug and in conjunction with exacerbations, in conjunction with symptom scores and quality of life scores. So I think the improvements that we see here in the overall population, I think, are really robust and will really reflect the overall efficacy of this therapy. The subgroup analysis, again, as Andy mentioned, will be shown in the publication.