Hello everyone. This is Thomas Kudsk Larsen from the Sanofi IR team. Welcome to the Q3 2025 conference call for investors and analysts. As usual, you can find the slides on sanofi.com. Please turn to Slide #3. Here, we have the usual forward-looking statements. We would like to remind you that information presented in this call contains forward-looking statements that are subject to substantial risks and uncertainties that may cause actual results to differ materially. We encourage you to read the disclaimer in our slide presentation. In addition, we refer you to our Form 20-F on file with the U.S. SEC and our French Universal Registration Document for a description of these risk factors. As usual, we'll be making comments on our performance using constant exchange rates and other non-IFRS measures. Numbers used are usually in millions of euros. And for Q3 2025, only as we state otherwise. Please turn to Slide #4. First, we have the presentation, and then we take your questions. we aim at keeping it all to 1 hour, including questions. For Q&A, we have Olivier, Brian, and Thomas to cover our global businesses as well as Roy, our General Counsel, and Brendan, Head of Manufacturing and Supply. [Operator Instructions] With this, I'll hand you over to Paul on my left-hand side.

Thank you, Thomas. Thank you, everyone for joining us today. Our growth momentum continued in Q3 with EUR 12.4 billion in sales, up 7% over last year's high base of comparison. Sales growth was primarily driven by our new launches and the performance of Dupixent, which reached EUR 4 billion in quarterly sales for the first time. After the third quarter performance, we are confident in the business outlook for the remainder of the year and reiterate our full year 2025 sales guidance. This positive outlook includes our expectations for the business in the U.S., our largest market by sales. We continue to work with the administration and policymakers in the U.S. and around the world on policies that improve access to treatments, lower prices for patients, and improve health systems and protect science. Our recent announcement on the expansion of our patient affordability program, offering improved access to all our insulins is a great example of this work, which I'll discuss in a little more detail later in the presentation. Now let me highlight the contribution of our new launches, which have been a significant driver of this quarter's strong performance. Our launches delivered EUR 1.8 billion this quarter, grew more than 40% and now represent 15% of our total sales. To put this in perspective, our launches represent almost half of Dupixent sales this quarter, demonstrating their significant contribution to our growth. We've strengthened our commercial portfolio with 3 new additions, Ayvakit, our medicine for both advanced and indolent systemic mastocytosis, the first Sanofi sales of Nuvaxovid offering an important protein-based and non-mRNA alternative for COVID-19 vaccination. And Wayrilz, our new BTK inhibitor designed as a multi-immune modulator. This innovative medicine provides a new option for patients with immune thrombocytopenia that extends beyond their platelet count needs to…

Thank you, Francois. I'm pleased to share our Q3 pipeline achievements with progress in many programs. We received regulatory approvals for Wayrilz in the United States in ITP and Tzield in China. Further, we received regulatory submission acceptances for Dupixent CSU in children in the U.S. and in the EU. The FDA nominated Tzield for the new Commissioner's National Priority Voucher program to speed up the review. Wayrilz ITP was submitted in Japan and Sarclisa subcutaneous was accepted globally in myeloma. Our Phase III programs have delivered successful readouts with amlitelimab meeting the primary endpoint in the Phase III study in atopic dermatitis dose 1 and Fluzone High-Dose in people 50 years and above. We also commenced dosing the first patient in new Phase III studies, 1 of the 2 studies for lunsekimig in COPD and Wayrilz in sickle cell disease and warm autoimmune hemolytic anemia. Finally, Wayrilz is emerging as a multi-immune modulation platform in rare diseases with an approval in the U.S. and a positive recommendation in the EU for ITP, multiple designations across new indications, further strengthening our rare disease portfolio. Please turn to the next slide. Moving to dermatology. Amlitelimab met all primary and secondary key endpoints in the first Phase III study in AD. The data demonstrated clinically meaningful improvement in several measures of skin clearance. As an example, looking at the vIGA measure, the efficacy progressively increased and showed no plateau at 24 weeks. Further, amlitelimab offers patient-friendly quarterly dosing. There were no new safety concerns identified in this study. OCEANA AD is a comprehensive program, including 5 Phase III studies in adults and adolescents, biologic experienced patients and across different geographies. We anticipate full data to report out throughout 2026. Brivekimig, our TNF-alpha and OX40 ligand nanobody achieved its primary objective in a…

Okay. Thank you, Houman. We'll now open the call to questions. [Operator Instructions] Now we will take the first question. Please go ahead.

Yes. First question from Sachin Jain from BofA.

So first one, I wonder if you could just update us on the tole SPMS regulatory debate and confidence resolving any questions the FDA has had with that delayed PDUFA? And the second one, just to make sure there's no confusion in the market given the debate at 2Q, can we assume that your wording of profitable growth for '26 means EBIT and EPS ahead of sales?

Okay. Thank you. Do you want to provide some clarity on the regulatory piece and the SPMS tolebrutinib?

Yes. Thanks for the question. Pretty straightforward. As you -- as we reported earlier in the year, the FDA requested an extension. We've submitted data sets with the FDA, continued conversations and look forward to the PDUFA December 28.

Thank you. And Francois? François-Xavier Roger: Just on profitable growth, I confirm that the idea is to have as we go down through the P&L, all items growing faster than at the upper end of the P&L, which means basically gross margin growing faster than sales growth, BOI growing faster than gross margin and EPS growing faster than BOI. We have achieved that each and every single quarter this year. We'll do it in 2025. We expect to get there in '26, and we are working in the same -- to deliver the same objective for the following years as well.

Next question is from Luisa Hector from Berenberg.

Perhaps I could ask the obligatory U.S. policy question. Do you have any updates on your conversations with the U.S. administration? Is there -- is it more complex for you with shared assets like Dupixent? And then should we be concerned about a degree of silence right now? Or is it simply that there is more of a bottleneck in terms of a long queue to speak with the administration, maybe Trump's busy agenda. Just any color on that sort of ability to communicate. I'll stop there.

Thank you. I think our answer is pretty straightforward, which is we've had ongoing dialogue with the U.S. government and indeed multiple governments since before we received the letter back at the end of July, I think it was -- we focus on making sure that people understand the value we can bring. And those conversations have continued throughout this whole process. So I can't really comment on bottlenecks or other implications. But I think that's probably as much as we can say at this point.

Yes. Next question, Shirley Chen from Barclays.

Can you guys hear me?

Yes.

So I have a question on Beyfortus. So as you guys said, so the Q3 orders have been impacted by inventory carryover from last season. Could you please give us a sense of how Q4 ordering trends are tracking so far? And are we still expecting a 3Q, 4Q equal split? And also given the competition that is ongoing, like how do we see the clesrovimab competition is going to impact the 4Q number, given we know that Merck actually entered the market at late-ish in 3Q. And also, we saw very strong growth in ex-U.S. market. And what do you see the opportunities there? And also, what is your overall perspective for Beyfortus in 2026?

Okay. So the 4 questions, Tom, for you.

Beyfortus rich question. Thank you very much, Shirley. Maybe a couple of points to address these different elements. Thanks for giving an eye to the Beyfortus performance in Q3. As usual, a few points of highlights. First of all, overall for Beyfortus, you saw that we increased our performance versus last year. And we have done that notably by extending the penetration to about 40 countries. Now in your question, there were some geographic elements to it. On the U.S. part, yes, absolutely, you're correct on what's going on in the U.S. And yes, we do confirm the guidance that we have provided last quarter, that Q4 we expect it to be roughly in the same order of magnitude that Q3 for Beyfortus overall for global Beyfortus performance. The second point, I think you wanted to mention a couple of points on the environment and the competition. Without talking about the competition specifically because we don't do that, maybe I can say a few words on the fact that actually, if you look at last year 2024, first year of full supply and Beyfortus was a blockbuster. You remember that if you look at the U.S. performance, the total, I will say, coverage rate of all babies in the U.S. was around 55% for all RSV solutions available, roughly 55% of babies were getting some form of RSV protection. We're really focused in 2025, and I think we will be focused in 2026 to increasing that vaccination coverage rate. We expect 2025 to land around 70% vaccination coverage rate in the U.S. And that's -- therefore, we're welcoming all efforts from everywhere to make sure that we increase the importance of awareness. What we are seeing in the signals in the U.S. is that by far, Beyfortus is the favored product, the favorite prescription from U.S. prescribers simply due to the fact that it's highly differentiated with an extended half-life of 71 days. It makes it by far the longest-acting monoclonal antibodies for the prevention of RSV. And it has an incredible real-world evidence behind it. So that's very important moving forward.

Yes. Next question, Matthew Weston from UBS.

Two questions, please. Francois, I'm going to buck the trend. Most people ask 2026 guidance questions. I'm going to ask a 2027 guidance question. Slide 29 flags the Regeneron R&D reimbursement stepping down. It's something that's been familiar, I think, for many people for a while, but maybe not fully reflected in consensus. The step down is actually much greater in '27 than in '26 and your slide implies about EUR 0.5 billion EBIT gap in '27. I'm just trying to understand, are there any additional levers within manufacturing gross margin or the business that could help mitigate that? Or that's something that we've just got to get prepared for even though it's 1.5 years away? And then one for Houman. You set out your enthusiasm for amlitelimab. Lilly has just reported the findings from the Phase III ADjoin extension study for EBGLYSS which looked at 8-week dosing and showed very limited erosion in efficacy. I'd be very interested if you think that limits the differentiation for amlitelimab where you were aiming for your 12-week to be differentiated?

Okay. Thank you. Maybe Francois, do you want to catch that and then maybe Brian and Houman... François-Xavier Roger: Matthew, I think that this is a good question for 2027, indeed, and which is what I presented in the last call at the end of July, the fact that in 2026, we will have -- we will lose about EUR 300 million of R&D reimbursement from Regeneron. And as I said a few minutes ago, it will be entirely offset by the additional royalties that we will receive from Amvuttra. In 2027, indeed, we will have a much more significant amount of R&D reimbursement in terms of decrease because we will lose EUR 800 million from one year to the other. And this will not be fully offset in 2027 by the additional Amvuttra royalties, which will be around EUR 300 million, which is in one of my appendices in this presentation as well. So we'll have a gap indeed of about EUR 0.5 billion. We will be able to cover it with other exceptional items or, let's say, nonrecurring items? The answer is no. That being said, I mean, we will continue growing at a reasonable pace as well. So -- and I said it last time that if there was 1 year where we were not sure about increasing our profitability and deliver this profitable growth, it would be 2027 for that same reason. That being said, and I repeat what I said last quarter, which is we expect our BOI to increase in absolute value in '27 in spite of this impact. So we will be able in absolute value to cover the gap, the EUR 500 million gap. And most probably, I believe that we may be in a position, still early to say to even increase our profitability in terms of BOI in '27 as well. I say we may be still very early. We have not even completed our 2026 budget. So I want to be careful. But directionally, there is probably a possibility largely as a consequence of the growth leverage that we will get from superior growth.

Okay. Thank you. So maybe Brian, first and then Houman.

Yes. I think -- a great question. Thank you very much. I think as you look at the marketplace, we've always said this is a marketplace that's going to continue to grow. The bio penetration rate is extremely low, just over 14%. So more assets coming into the marketplace will only help the marketplace grow. As you can see from our growth today in Dupixent, a market -- a product that's already on the marketplace, it's benefiting from other therapies coming to the marketplace. That said, one thing that we've said consistently for a very long time, these IL-13s are incomplete therapies, and we've seen them on the market now. Actually, if you look at lebri, that's been on the market for nearly a year now as single-digit share. So we're seeing while it's helping grow the marketplace, it's really not taking very much share from Dupixent. Dupixent, we believe, still has a very strong profile. And each of these still have these dosing ranges that are either 2 weeks or at the best, 4 weeks is what we're seeing right now. So we think that there's a lot of room for a much more durable dose in the marketplace, assuming we finish the regulatory trials and get it to the marketplace for [ nema ] or for...

Just, Matthew, thanks for the question as well as the bio pen comment, which is important. This is a totally novel mechanism, an apical node in immune response with not only durability, but multiple differentiating factors. We remain, based on the data actually for the K1 study, encouraged by the potential for amylin.

Yes. Next question from Seamus Fernandez from Guggenheim.

So just a couple of quick questions. Can you just update us on riliprubart and just kind of timing dynamics around that? It's unclear if that's still on track for second half 2026 readout in CIDP. So I just wanted to clarify that. And then more broadly, just hoping to get a better understanding of when we're going to learn more about the programs that have had kind of unfortunate outcomes and where a number of programs are under review, including the oral TNF, including itepekimab. It just seems like there's a lot of secondary analysis exploration going on and holding on to assets as part of the pipeline. And I'm just trying to get a better understanding of when we're going to know the advancement or elimination of some of those assets that haven't quite lived up to at least investor expectations.

Okay, Houman?

Thanks, Seamus. I'll try and be succinct on this one. On riliprubart, we've updated the timelines at this Q3. And the reality is that the outcomes of the 2 Phase III studies for CIDP are just creeping over the year. This is purely a patient recruitment phenomenon, and we look forward to seeing the results of those studies. You'll remember the Phase II studies were extremely encouraging. And then on your second point about when you'll see the data and when we'll make decisions, obviously, those data sets will all be presented at the relevant scientific congresses. I've already announced today that we will go forward subject to regulatory approval with our partners in itepekimab. So there's no tardiness there. Obviously, we have to take a regulatory opinion before we move forward into a replication Phase III. And then for the other studies, I've already alluded to the fact that we will take a portfolio view on asthma with a number of our assets and figure out what we take forward in asthma. And on balinatunfib, we've just had access to the results. I've outlined the importance of ACR50 and 70 and the fact they're clinically meaningful. And we'll figure out -- as we've always said, we've been consistent in our view that we'll figure out exactly the role of balina in mono and combination therapies, both with our own molecules and partners.

Thank you, Houman. Thank you, Seamus. So great questions. We -- finally, we've seen it with many of our competitors having hits and misses in immunology over the last months that these extra levels of thinking actually are worth doing and stand you in good stead because you really do pick the right patient population. So taking time, I think, is wise for us. Next question?

Yes. Next question from Simon Baker from Redburn.

Two, please. Firstly, on Dupixent. Francois, you said that the gross margin benefit from manufacturing improvements is now fully being captured. I'm just wondering if you could give us some idea of the magnitude of the gross margin improvement this quarter, which is down to Dupixent manufacturing? And then a question on indication opportunities. Houman, you mentioned rilzabrutinib in Graves' disease. That's potentially not a particularly rare condition. So I just want to get your thoughts on the potential you see there. And also the other one in light of Moderna's failure this week is CMV vaccination. I know you've been in this space in the '90s. I just wonder what your level of appetite was for it now.

All right. Thank you, Simon. Francois? François-Xavier Roger: Yes, Simon, on Dupixent, the gross margin contribution from the C3 manufacturing was actually very limited in Q3 itself. I mean this is -- we just took the opportunity to mention that -- we have completed the full implementation of this new technique, which has spread over a couple of years actually, but it did not have a significant impact in Q3 per se. I take the opportunity to mention that our gross margin increased globally for the company by 2.5 percentage points in Q3 and by 1.8 percentage points in H1. Most of the factors are still contributing to it are still relevant for the future to a certain extent. One of them is volume growth. Our volume grew by 12% since the beginning of the year. We expect to continue at a high level. We are obviously benefiting from a positive product mix, including this quarter, Ayvakit, actually Ayvakit is much more significant than the new manufacturing technique for Dupixent. We are also obviously benefiting from the industrial restructuring that we did over the last couple of years. Plus there were some one-offs this quarter -- last year and this year, which did create a little bit of positive impact as well. If we look at it underlying because we were at the high range, once again with 2.5 percentage points of increase in Q3. If we exclude the one-offs and some of the items that will not necessarily replicate each and every single quarter like Ayvakit, for example, you can consider that the underlying gross margin increase that we have experienced since the beginning of the year is around 1 percentage point. This is obviously before any impact, if you want to use that for the future, that does not include any potential impact coming from tariffs.

Okay, Houman in Graves' and then Thomas, CMV.

Okay. Simon, thanks for that insightful question. As you all know, Graves' is a well-established autoantibody-related disorder. The classic long-acting thyroid stimulating antibodies, the TSHR antibodies are important. So number one, it's a canonical autoimmune disorder. Number two, we know from investigator-initiated studies that B-cell suppression is a successful therapy, particularly for the ophthalmopathy. And thirdly, with our unique covalent reversible molecule in rilzabrutinib that has already shown significant promise in multiple disorders, including IgG4 disease and ITP. I think that Graves' is a promising opportunity, and we look forward to taking the molecule forward. You are completely correct that it's not a rare disorder of that sort. It's not super rare. And therefore, I think it's a potential opportunity, particularly the ophthalmopathy.

Thank you. Thomas, CMB.

CMV [indiscernible], not much to say. The news is very recent, as you know very well, Simon. So I will not comment and we need to see the full data set. The only thing I can refer to is indeed you know better that quite a while ago, a few decades ago, we had worked on this antigen. It's a difficult target while we had reached some, I would say, interim efficacy our assumption at that time was that it will not be sufficient to reach a protection level, and that's why we had invalidated this program quite a while ago. Sadly, overall, because the field of CMV vaccination is an important field, and we will welcome a vaccine against this devastating disease.

Next question from Richard Vosser from JPMorgan.

A question on Dupixent and just giving us a little bit more on the development around COPD and the -- also the gross to net, how that developed in the quarter and how we should think about that in '26, but also how the COPD launch is going, seems to be developing a little bit better this quarter. And then second question, just on the Inhibrx data. Just you to file, I think you need some of the more long-term safety data that you called out from the open-label extension. Just wondering what, if anything, is being looked at in terms of that safety data from the regulators? Is there anything of interest that they want to see or rule out?

Okay. Brian, Dupi?

Yes, Richard, thank you so much for the question. And first and foremost, I think the really strong overall growth that you've seen is really coming from all different sources of growth. If you think about it, first, our foundational indications, we continue to grow bio penetration in things like asthma, atopic dermatitis, EoE, nasal polyps. But also, we've moved into about a year ago, we launched in COPD, and we've really seen a strong success actually in COPD, one of our fastest -- actually, it's our fastest respiratory indication as far as growth rate goes. So that plus CSU plus BP, you can see now 8 indications deep into the U.S. Our sources of growth are coming from everywhere and of course, launching around the world. It also has actually created this really strong momentum we've seen with 26% growth this quarter and reaching over EUR 4 billion in sales. As it relates to gross to net, obviously, that's captured in there in reference to our sales growth. This is something that we've monitored for a long time. Actually, we see that as we go into additional sources, if you will, there are different payer groups, we obviously will provide discounts to get into different access for different patient populations. But again, this is something we've known for a long time, and it's captured in our long-term guidance for Dupixent.

Okay. Houman, Inhibrx?

Excited to have done this acquisition, an example of our disciplined capital allocation policy. And it's exciting that the preliminary data has proved so promising superior both in Q3 and Q4 to some of the standards of care. You're correct that we're interested in the long-term extension study, which is open label for the safety data. There is no specific, if you remember, this is a fusion protein like an antibody. There's no specific side effects we are looking for, but you'll have noted from the press release that the safety and tolerability were in line with that, which was expected.

Next question from Michael Leuchten from Jefferies.

Two questions for Francois, please. One, I just wondered if I could provide you a little bit more on the seasonality comment on gross margin. You said the Dupixent process is now tucked in and you return to normal patterns. Just wonder what you meant by that? And then if most of the driver of the gross margin increase in Q3 still hold, just wondering why you opted not to offer an increase to guidance for this year.

Okay. Thank you. Francois? François-Xavier Roger: As I said, the 2.5 percentage points of increase that we had in our gross margin in Q3 is probably not necessarily a good proxy for 2026. So I do believe that we have some factors that will stay there like our volume growth, like, for example, even the product mix, these issues are structural, but we won't get another increase next year of 2.5 percentage points in gross margin. So it will be probably closer maybe to what I said, underlying maybe 1 percentage point, benefiting from volume and product mix. The other thing in terms of guidance, I'm glad you asked the question. So we do confirm our full year guidance, which is high single-digit growth -- sales growth. Today, after 9 months, we are at 8.8% year-to-date. So do expect some increase from where we are in -- at the end of September. Business EPS growth is low double digit, even excluding the benefit of share buyback. We are excluding share buyback at 9.9%. So do expect there as well for the full year that we will go up from where we are as at the end of September. So just to clarify as well, our full year guidance assumes basically that Q4 will be the best quarter in sales, BOI and EPS growth this year. We will do better than we have done in any of the quarter. Q3 was anyway a bit softer because of the comps, as we said earlier. There is just one thing I want to take the opportunity to mention one thing, maybe what the Street does not always estimate is the profit sharing with Regeneron. That includes, by the way, both Dupixent and Kevzara because there is probably an understanding that it grows in line with sales of Dupixent. It doesn't. Let me just give you some further color. For example, if we look at Dupixent sales in H1, they grew by 21% and the Regeneron profit sharing grew by 32%, so 11 percentage points faster. If I do the same analysis for Q3, Dupixent sales grew by 26%, which is remarkable. And Regeneron profit -- the profit sharing with Regeneron grew by 37%. So another 11 points higher than sales. So once again, it is a profit sharing. It's not linked necessarily fully to sales, and there is about 10 to 11 points of growth in terms of difference between the 2 concepts.

Next question from Florent Cespedes from Bernstein.

Can you hear me?

Yes.

Florent Cespedes from Bernstein. Two quick questions, please. First, on M&A with the massive success of Dupixent and with the mixed news flow pipeline, maybe could you be more aggressive in terms of product acquisitions and maybe kind of a Blueprint like transactions is something that we should see in the future? And maybe a second quick question for Thomas on vaccines. Can you share with us how do you see the trend notably in flu with the vaccine fatigue that we observed across the world? So some color on this would be great.

Maybe I didn't quite catch perfectly the first question, but it was regarding M&A and should we be doing more and more Blueprint, like is what I think I heard. Francois, do you want to comment? François-Xavier Roger: Just a few words. Maybe Paul, you can complete it. It's not really about being aggressive. It's about finding the relevant acquisitions. We have space in our balance sheet. We said that we want to retain our AA rating. In order to get there, we could afford -- once again, this is not necessarily what we want to do, but we could afford investing in BD and M&A currently, something like EUR 14 billion, EUR 15 billion and still retain our AA rating. Anyway, what we are looking for is to meet 3 criteria, basically, strategic fit, which is around our 4 therapeutic areas and possibly white spaces as well. Second, scientific differentiation and relevance and first-in-class, best-in-class. And third, financial return as well without any certainty. So it's less a matter of amount or aggressiveness. It's more about finding the right targets at the right time at the right price.

Yes. Yes, I think that's great. We've been really disciplined. Francois, when he came in, actually went back and looked at all of our acquisitions and agreements to decide whether we allocated capital to his standard. And I was somewhat relieved to find out that we did. There is a huge amount of discipline. There is a huge amount of discipline. And I think you've just heard how Ayvakit has done in Q3. So we're really -- I think we're very good at it, but we have to be a little bit choosy about what we do. I think that's very reasonable. Okay. Thomas, Vaccines?

Thanks, Florent, for the question on flu. A couple of points. First of all, it's early. We're still in October. But indeed, as you had in mind with your question, I think it's fair with the first 2 weeks that we observed a little bit vaccination rate on the soft side when it comes to flu vaccination, particularly in the U.S. So we see a soft this year to date. A couple of points, though, I'd like to highlight when it comes to the 2025 performance that you've seen in this quarter. First of all, we had highlighted it before, but I just want to be clear that it's linked to 2 elements. In Germany, there is a price effect where there is a significant price decrease due to the change or for recommendation. And in the U.S., it's more what you're mentioning, i.e., the soft this year. But in both cases, in this 2025 environment, we are keeping a very strong market share performance overall for Sanofi flu vaccines, but in particular, even for differentiated flu vaccines. And beyond the performance in terms of market share, what I'd like to highlight also this quarter in terms of flu is the progress we're also making on an R&D perspective. You see that we have a positive Phase III in Fluzone High-Dose/Efluelda 50-plus extension. So with the great performance you've seen with the FLUNITY trial and the fantastic 32% improvement compared to standard dose on influenza hospitalization. If we can extend that to people above 50 years of age, that will be fantastic to be associated also with the progress we've made on pandemic flu with, I think, we are the best-in-class H5 seroprotection results. And the move and progress we're making on flu, COVID-19 combination. So flu remains important for us. We are moving forward with teams commercially and R&D-wise.

Yes. Next question from David Risinger from Leerink.

So congratulations on the positive Inhibrx ElevAATe results this week. Could you provide some more color on how you would characterize for AATD, both the normal range and the trough levels? And then regarding net price prospects for U.S. Dupixent in 2026, since your contracting is likely largely complete at this time of the year, how would you characterize the expected net change in pricing in '26 versus the net change in pricing in 2025?

Okay. Thank you. Inhibrx AATD?

Yes. Without running into any embargo issues, you'll remember that the definition of alpha-1 antitrypsin levels is related to the wrong crystal nomogram. I'll speak in broad terms, standard of care, by and large, doesn't get into the normal range for alpha-1 antitrypsin levels. Our Q3 and Q4 molecules provide very commendable alpha-1 antitrypsin nomogram levels, both for trough and mean dose.

Thank you. And Brian, I think a clever question from David trying to update at the rebates likely for '26 over '25, but I'll let you answer that.

Yes. I think it's a great question, David. Thank you so much for asking it. As you know, we don't typically give guidance on the net price year-over-year. But one thing I will say, as you can see, we've been incredibly disciplined over the years. We're now 8 years into the launch of this asset. And it's been captured in our long-term guidance how we believe the net price will develop over time.

Yes. Next question from Sarita Kapila from Morgan Stanley.

Just on amlitelimab, how should we think about the upcoming readouts? Is there potential for the placebo arm in the COAST 2 trial to behave more normally? Or should we think about it as a pure sister trial to COAST 1? And then just on ESTUARY, beyond the no plateau in efficacy that we saw in COAST 1, what's underpinning the confidence that the long-term efficacy can improve with time?

Great questions. Okay, Houman?

Okay. So I'll take them one at a time. Thank you for the question. But firstly, on COAST 2, it's a precise replicate the study, there are some subtle differences in regional recruitment and execution, but essentially, it's a replicate study and we anticipate and hope that we will get a replicate of COAST 1. As you'll remember also, you've said correctly on ESTUARY, it's a slightly more nuanced study than has perhaps been observed as well as being able to tell us about durability ultimately in a randomized way. It will also give us a sense of dose variation that we will do. You remember there are multiple dose switching arms. So punchline on ESTUARY, which we'll get throughout next year is not only will it tell you about durability, but it will tell you about the relationship between dose and durability. And those are going to be critical in terms of our understanding of the positioning of amlitelimab.

Okay. Thank you very much, Houman. Yes, we'll see. We'll get the data. We'll see how competitive we are. I think from a commercial perspective, we're very enthusiastic, but we'll let the data read out. Okay.

Next question from Steve Scala from TD Cowen.

Two questions. Yesterday, Roche said they were taking patients back from ALTUVIIIO. What is the nature of the patient that is being taken back? And what does this mean for ALTUVIIIO long-term growth outlook? And the second question is, based on the subgroup data presented at ECTRIMS and the language in today's press release, it seems that any tolebrutinib SPMS approval will be in subgroups. What subgroups are likely to be in the final label? And what portion of the overall SPMS market will this represent?

Houman, why don't you deal with that last question because we better clarify that as soon as possible.

Yes. So just to be clear, I'm going to give you 2 facts. The SPMS population is about 170,000 and the PPMS population, 120,000 at no point have we entertained the notion of doing subgroups. The regulatory discussions are ongoing. We don't participate in any further insight during those regulatory discussions.

Thank you. Brian, ALTUVIIIO?

Thanks, Steve, so much for the question. So a couple of things. ALTUVIIIO saw another really strong quarter. It's becoming very clear to us. This is the #1 switched asset. So it's being switched to -- we are the #1 asset that's being switched to in the hemophilia marketplace in hemophilia A. We're still seeing, as we have shared before in the past, we've seen about 2/3 of our switches coming from competitors. Of that, still 10% is coming from Hemlibra. So I can't really comment on what Roche shared. But what we are still seeing is 10% is coming directly from Hemlibra, and we are the #1 asset that is being switched to. Now about 1/3 of the business is still coming from Eloctate, but you can see our overall -- that's starting to stabilize and our overall heme A business is actually growing quite significantly. So again, we're very pleased with the performance and remain committed to delivering our next blockbuster this year with ALTUVIIIO.

Next question from James Quigley from Goldman Sachs.

I've got a follow-up question for Francois on the Dupixent on the antibody pay away. So I think backing out how the BOI margin seems to have been developing for Dupixent. So it looks like last year, it was in the region of sort of 62%. This year, third quarter looks like it reached 72%. Again, working backwards, maybe there's like a 92%, 93% gross margin. So first of all, is that the right ballpark in terms of what we should be thinking about when calculating the pay away? As you mentioned, it seems not necessarily be fully reflected in consensus. And going from a 72% operating margin going forward, how should we think about operating leverage, obviously, going from last year to this year? It looks like there was a big step-up. You mentioned the increase in percentage point growth for the pay away relative to Dupixent. So where could the margin go? And how should we think about that going forward? Or maybe more so where could -- at what point could the margin peak? And secondly, maybe one for Brian as well, a follow-up on ALTUVIIIO. You mentioned about taking share from other Factor VIII therapies. Where are you now in terms of the Factor VIII market from a market penetration and market share perspective? Where do you think you're going to end up? And if that's below 100%, given the data, why would that be the case?

Okay. Well some highly specific questions, James, thank you. I think, Francois, just to give a broad comment to that. François-Xavier Roger: James, I would like to help you, but we don't disclose the margin by products, and so we don't do it for Dupixent or for any of our products. So unfortunately, I cannot provide you any information on that. Obviously, the margin has been improving for Dupixent over time, given that, I mean, we are benefiting obviously from scale, efficiencies, as I said earlier. So our margin has continued to increase.

Okay. And Brian, maybe similarly, I don't know how specific you want to be on that market.

Yes. I don't really get incredibly specific on that. I mean what you see in the heme A marketplace is that you've got the factor marketplace, which is, again, as we said, where we continue to take the most business from is the factor marketplace. Again, this is a factor therapy that obviously comes with increased efficacy. We are taking some from the non-factors, as I mentioned before, 10% coming from Hemlibra, but we don't know, we haven't shared exactly where that might end at the end of the day. But again, as we said, we're the #1 asset that is being switched to. So the progress continues, and we'll continue to keep you updated as it develops.

And then the last question, I think.

Yes. Last question from Peter Verdult from BNP.

Pete here from BNP. Given I'm the last on the call again, maybe [indiscernible] Houman with a few quick pipeline questions. Just on itepekimab human, is it simply a case of the go-forward strategy that you repeat the AERIFY trial? Or will you solely be looking at the former smokers only? Then on efdo, data looks great. You need to gather safe data. But just on the regulatory pathway, if I recall a few years ago, FDA was making some noise about wanting to see respiratory functional endpoints as well as the biomarker analysis. Is there any of that going to be required for efdo? And then lastly, on tole, SPMS opportunity is huge. But how concerned or not should we be about tole getting an onerous label requiring weekly liver monitoring? How much of an issue would that be in your mind commercially?

Okay. Pete, thank you. Last but not least, okay, Houman?

Pete, looking forward to seeing you next week. Firstly, to start. Second, let's get into it. AERIFY-3, broadly speaking, we anticipate that we will have to do some form of replicatory trial. The details of exactly what we do have been guided by both internal and external data sets. We've looked broadly trying to understand why AERIFY-1 and 2 differed. And the exact construction of those trials will be dependent on both discussions with our beloved partner, Regeneron, but also the regulator. So look forward to more on that soon. On efdo as we call it internally dora, the -- just to be clear, you're right that the FDA -- first, FDA may have different perspectives. You'll remember that current standard of care doses to a target of 11 micromolar, which is about 50% the lower limit of normal of the range. And then many patients drop below -- way below that 3 to 5 days. The first point I would make is, and we will disclose the data at a conference, but the Q3 and Q4W dosing that we have is substantially better than that. So we will have a conversation with the regulator as to, a, their intrigue related to those levels, but also what other additional endpoints we'll need pursuant to both natural history data and our open-label extension safety data. And on your third point on tole, SPMS, we're in late-stage discussions, as I said, the PDUFA is 28th of December. And while we anticipate that we'll need some sort of rounds with tole, there's been no further disclosable comment on the intensity of blood draws.

Maybe on that last point, Houman, it's fair to say, isn't it that we learn a lot actually between AERIFY-1 and AERIFY-2. I think you touched on that. So if you like, even better informed. And on tole, that first 90 days of what we did in the clinical trial program and what we're doing in real life, seems to be very practical and responsible. And so I think we're confident we have the right approach. Of course, it's the regulator that will decide what that looks like. Well, okay. Thank you, Peter. Thank you all. Our growth momentum continued in Q3. After 3 quarters of sales and earnings progress, we reiterated our 2025 guidance. Our pipeline delivered important milestones in this quarter, as outlined earlier. And finally, as we're looking forward to 2026, we are confident in our ability to pursue our current trajectory of profitable growth. With this, I wish everyone a good autumn, and we'll now close the call. Thank you.