Novo Nordisk A/S (NOVOB) Earnings Call Transcript & Summary

Good afternoon, and good morning, everybody joining us on the phone. Thank you all for joining us. My name is Keyur Parekh, and I cover Novo for Goldman Sachs. My pleasure to have the management team here post the full year results yesterday. So without any further delay, Lars, passing it over to you for any introductory comments and we'll move to Q&A from there.

Yes.

Just one logistical thing. On the Q&A, if you -- because it's being webcasted, please wait for the mic to come to you before you start asking your questions. Thank you.

Thank you, Keyur. And thank you to Goldman Sachs for hosting us here today, and thank you all for coming. I'll start out by some introductions and go through our sales performance. Then Mads will talk to R&D, and Karsten will close on financials and outlook. We will be talking about the future. So we have to bear in mind that forward-looking -- these forward-looking statements that the future might not exactly turn out as we have [ presented ] here today. On 2019, this is the layout from our Capital Markets Day on how we have set our strategic aspirations. Overall, 2019 was a very strong year for Novo Nordisk. We're very pleased with our commercial execution, also the progress in R&D. We'll be reviewing that with you today. So from this slide, I'll just highlight that '19 was a year where we reached more patients than ever. So we reached 30 million patients. So we take that very promptly and are keen on focusing on delivering good products to those. If you look at our sales development, top line growth of 6% overall, driven by strong performance in International Operations growing by 11%. So compared to a historical level, this is a significant step up compared to the 5% to 6% we grew for some years. And you can see across the region, strong contributions coming from all regions. U.S., North America grew by 1%. You know the underlying composition that there's a very strong, [ Tier 1 moment ], which is then dragged down by insulins. So overall, despite the 1% growth, we are also very confident and satisfied with our U.S. performance. If you look across the therapy areas, very strong GLP-1-based growth, both in the diabetes category and obesity. And then here, we see continued positive momentum on insulins coming from International Operations, but the drag from pricing in the U.S. pulling that down to a 0% -- a [ 3% ] decline. If you look at biopharm, healthy growth. We see that we are holding up based on NovoSeven and a good contribution from our hemophilia products and also Norditropin. So we'll get back to that. If you look at performance of the GLP-1 franchise, here, we have on the left, U.S. new-to-script market shares, and we can see where we are pulling away with a 57% share of new scripts, really fueled by continued strong performance of Ozempic and now also Rybelsus kicking in with a healthy initial uptake. The total -- on the total scripts, we can see it's flattish, but we're starting to pull away from the dulaglutide performance. So we are very pleased with, one, the overall market growth but also our performance [ margin-wise ] in that segment. If you look at the European markets where we have launched Ozempic, we can see that now we're clearly turning around the GLP-1 performance in those markets. And when you look across the regions in International Operations, very strong share of growth coming from GLP-1s in region Europe, 76%; but also strong contribution from China, Latin America and Japan and Korea. If you look at obesity sales, we see a continued strengthening of our position in the overall obesity market. It's not a large market. So we're taking a large share of that, but we see a constant strong addition from the Saxenda franchise. And that's growth both coming from IO and North America. If we end up here on biopharm, we see that NovoSeven is actually doing better than we had forecasted some years back. There's still breakthrough bleeds for patients on HEMLIBRA and NovoSeven plays a role in that. And then we see both NovoEight and our long-acting agents are actually adding to growth, and so is Norditropin. So just a couple years back, we made biopharma a dedicated business unit with a dedicated leader in the executive team, and we can see that, that strengthened focus actually pays off in our commercial performance. So we're pleased to see that biopharm is back in growth. So with that, I'll hand over to Mads for an update on R&D, starting with Ozempic.

Thank you very much, Lars. Talking about Ozempic and Rybelsus, I think you're all aware of the fact that we got a cardiovascular indication in the label for Ozempic, with a beautiful 26% base reduction stemming from the SUSTAIN 6 trial. Whereas for Rybelsus, we've got the headline data in but no indication because the size of the PIONEER 6 did not justify that in terms of the robustness. But that is why the SOUL trial is now ongoing, first of all, to gain us that label for Rybelsus, the CV indication claim, but also to be able to bridge peripheral arterial disease data and kidney disease data that are confirmatory secondary endpoints in the SOUL trial to bridge them towards the data from the 2 dedicated Ozempic trials, allowing us also, hopefully, label claims in those areas. The Rybelsus European opinion came just a few days ago, as you're also aware, and that's going to be a really nice label with all the superiority data against competitors. The detailed cardiovascular data and so on will all be included in that label. Now the news of this quarter in terms of real R&D excitement, in my view, is the once-weekly insulin icodec, also known in the old days as LAI287, an insulin that has been optimized to stay in the circulation and act with full potency, one might say, against the insulin receptor on a once-weekly basis. It has been up against insulin glargine U100 in the key Phase II trial, and these are the data that have emerged, a very, very strong reduction in A1c with good target achievement and you can say even a greater A1c lowering numerically by 0.2% versus glargine U100. We can always discuss why that might be. And then the hypoglycemia rates among both products were, compared to the excellent glucose control, actually on the low side. This has made us optimistic that we can go into Phase III towards the end of the year with what is both a big, big, big convenience benefit but actually also aiming to seek clinical differentiation against what is arguably still the world's most used basal insulin, glargine U100. In terms of news flow throughout this year, well, diabetes-wise, the excitement will, of course, be surrounding starting Phase III for icodec. We'll be getting the high-dose sustained forwarded data for Ozempic, but also getting, importantly, Rybelsus approved in the next quarter in Japan because this is one of the biggest OED markets in the entire world. As we then look into the obesity situation, we have a very exciting inflection point in second quarter, where the STEP program, 1, 2, 3 and 4, should -- if things go well, all read out. And we'll keep you updated on that. But so does the amylin Phase II monotherapy data. And around the same season, we will also see the emergence of multiple-dosing combination data at different ratios between semaglutide 2.4 and increasing doses of AM833, the once-weekly amylin analogue. As we move into biopharm, it's not up here, but concizumab, actually, towards the end of the year or around year-end, we'll hopefully be able to complete the first data set in the niche indication hemophilia B with inhibitors, where we have achieved a breakthrough status with the FDA. Mim8 is our next-generation Factor VIII mimetic antibody that has entered Phase I on January 24, to be specific. And somapacitan, we are targeting approval for adults with growth hormone deficiency during the course of the second half of the year in both Europe and U.S., whereas, we will submit it in Japan during the course of this quarter. And finally, on other serious chronic diseases, as we call it, NASH is an area of also a very exciting inflection point, with biopsy data from the 310-patient-sized Phase II semaglutide NASH study will emerge, as will also [ lose ] combination data with the FXR and ACCI compounds from the Gilead Company, with whom we collaborate on a clinical basis. So with that, an exciting year full of data readout, hopefully good ones, but time will show. That's a forward-looking statement. And then over to you, Karsten.

Great. Thanks, Mads. Clearly, a super exciting year in terms of pipeline readouts, so we can come back to that in the Q&A. So for full year results, I brought a slide with a lot of numbers. I know I'm among friends in terms of reading the numbers, but we delivered a very strong set of accounts for 2019, totally in line with our guidance, or even slightly better. And then there's sell side, frankly also. And then what would that mean? That would mean that, in local currencies, then our sales growth in the fourth quarter was 6%, our OP growth in the fourth quarter was 6%. Full year sales 6%. Full year OP 6%. And free cash flow increased 6%. So I think there's a magic to the 6%. But going into the P&L, then we had some currency tailwinds. So our reported sales growth was 9%. So upping our [ accrued ] sales by DKK 10 billion year-on-year. So a very successful year there. You heard about the commercial performance. Sales and distribution cost, up 6%. So a lot of investments in driving growth in IO. And then in North America, it's a lot of reallocation between from insulins on to GLP-1 business and on to our obesity franchise and growing that. R&D, we have some nonrecurring events with the priority review, March 2018, and some inventory revaluation. Adjust for that, then we're growing our R&D spend roughly in line with sales. All in all, 6% local currency operating profit growth. Then we have our hedging losses, the 3.9%, of which the majority is linked to hedging losses, the strengthening U.S. dollar and the CNY, tax rate of 20% and, hence, the resulting net profit. So then that yield, a very strong cash generation. So our cash to earnings is 88%. So we convert almost 90% of our net profit into cash. So more than DKK 34 billion in free cash flow generation. The DKK 34 billion we returned to shareholders between dividends and share buybacks as we've done in prior years. So we're very disciplined around returning our free cash flow to shareholders. And as you see here, also into 2020, that we are raising our dividend per share compared to -- for the 24th consecutive year. And then we just announced a [ DKK 17 billion ] share buyback program for the coming 12 months. And so basically, again, returning all the free cash flow we're generating to shareholders in 2020. Then formally, we have the long-term financial targets that served us well for many, many years. You saw our strategic aspirations that last [ cohort ] that is moving into that space. So -- but just formally, now we're closing out our long-term financial targets. We met the targets as we show on the right-hand side of the slide in terms of operating profit growth, operating profit to net operating assets and our cash conversion. This doesn't mean that we'll not be tracking, and they will stay in fashion about the cash conversion, et cetera. It's just we're not issuing formal long-term financial targets in that respect. 2020 guidance, totally in line with what we communicated in terms of outlook for 2020 at the Capital Markets Day. Then we have guidance of 3% to 6% local currency sales growth and 1% to 5% on operating profit growth, the delta in growth rates basically being an investment here in investing in Rybelsus' launch, especially in the U.S. and will simply roll out on a global scale. So that's why we have a fantastic opportunity in 2020 really to get those products out. Then nothing major on tax rate. On our free cash flow in the bottom, DKK 36 billion to DKK 41 billion, so midpoint, up DKK 4 billion compared to 2019. So really strong cash generation. And part of the cash generation is also that our CapEx is coming down from a level of DKK 9 billion, now down to DKK 6.5 billion in 2020. And that's basically the story of our U.S. [ debut ]. So our API manufacturing project in North Carolina, getting closer to finalization. So most of the constructions are done. Of course, work is still ongoing, but now we are moving more into, you would say, qualification, validation mode, more so construction. So you should expect our CapEx level to be at this level or perhaps even lower in the years to come. So with that, over to you, Lars, for final comments.

Thank you, Karsten. And we will, throughout the year, keeping ourselves honest against the strategic aspirations. And you heard Karsten talk to the targets on the financial dimension, Mads went through what to expect from our pipeline. Commercially, we will continue to drive our diabetes performance more towards the 1/3 market share aspiration we have defined, more than doubling of our obesity sales and getting a sustainable growth profile from biopharm. So maybe just a comment on the sustainability part. We are launching here in January this year an expanded affordability program in the U.S., really trying to take away the burden of insulin affordability as it's being discussed a lot. So we have launched this $99 program, where you can get what is equivalent of 1-month supply of insulin for $99. We've also launched a half price for one brand, NovoLog and NovoLog Mix. And then we have put in place an immediate supply option. So any patient in desperate need of insulin can go to a pharmacy and get very easy access to Novo Nordisk insulin. So we feel really good about that. And I think it's important to show our commitment in a challenged health care system where people are falling between the cracks. On the environmental side, I'll also say that, from this quarter, we're also utilizing renewable power in the U.S. on all sites, manufacturing labs, et cetera. That means that, from this year, all powering of manufacturing facilities in Novo Nordisk are based on renewable energy source. So all products leaving our strategic sites were based on green electricity. So we feel good about that. With that, we go to the Q&A.

Thank you. It's Keyur Parekh from Goldman Sachs. Two questions, please. The first one is, I think Camilla mentioned in the call yesterday that 70%, 80% of the scripts you're getting for Rybelsus is coming from the orals. And that's obviously kind of very early data. I think we're all aware of that. But as you think about how that's tracking versus your expectations, is it kind of broadly in line? Is it better than you expected it to be? Is it -- do we expect 100%? Just help us frame that early data. And then secondly, Mads, I can't remember a time where Novo had optionality on, whether it be obesity, NASH, concizumab, somapacitan, Alzheimer's, next-generation insulin. As a senior management team, what does that optionality bring with it in terms of the risk but also the discipline needed to make the right decisions? And how are you doing that differently from what you might have done in the past?

So let me start out with the source of Rybelsus business. You can say our positioning of the product is clearly early on. But one thing is the marketing campaigns and the positioning. Another thing is, of course, what happens in the market. So it's early days, as you say. Based on our experience from having launched Ozempic, I would probably have expected a bit more mixed sourcing because that's what we saw. So I'm maybe positively surprised about how we see it right now. But I think we should be very cautious and understand that it's still early days. And I think you will have some physicians that are first movers, probably based on excitement around the product. And then you have more steady-state source of business, which could look somewhat different. But it's clearly the positioning we're aiming for, I'm not sure that exactly will last forever, but it's for sure, a very good start. Then Mads on how we look at R&D and the broader set of options we have.

Yes. Thanks, Lars. So first of all, Keyur, governance-wise, just both Lars, Karsten and I, we sit on the same product development committee, implying that the optionalities are always weighed against each other, also in cost terms. When that is said, I would agree, because I've been around for a few years. And I would agree we have many optionalities right now. And some of them are like must-win optionalities. Insulin icodec, it's not an option not to develop, empower and make the right program for that great molecule as fast as possible. Then there are other ones where you can decide to make a bit of an inflection point evaluation such as NASH. If semaglutide pans out in a good way in NASH, then I think we are en route to something that could be the emergence of new therapy area. In the, hopefully, less likely that semaglutide does not pan out with strong data, Q2 or summertime this year, then that must tell us something about, at least, it's a disease that is more tricky to treat than we had envisioned from the get-go, because sema is perceived by many outside observers to be one of the more promising agents. So -- and when it comes to biopharm, as Lars alluded to, Ludovic Helfgott is doing a phenomenal job as EVP of Biopharm. We have promised him financial reasons to really optimize the assets we have and build new ones. And that means that concizumab, somapacitan and so on, they are also must-win battles, maybe not at the same magnitude as the GLP-1 analogue. But in relative terms, they will help us grow biopharm in the years to come. So we are weighing things against each other over time, but there are some things we simply must do. I don't know if you have more to say.

Yes. I think, from a strategic point of view, we are in a cycle where we see strong commercial momentum from launching a number of products in many markets. So I feel good about, say, the short, medium term and also into the longer-term growth prospects. So it's in that period of time, we need to make sure we seed the options for the long term growth of Novo Nordisk. So you'll see that we are expanding in that area. And then, obviously, at a point of time, we need to make some choice and double down on what do we make the big bits on. So I think it's a very positive situation that we have that in-house capability to actually create these projects because it is the most attractive way to make and return to our shareholders that we do this based on in-house, complemented with selective external sourcing. Good. Maybe we should keep the microphone, and then we'll move over there. Well, we'll take Simon first. And then you're going to -- maybe in the break, then move it down. There's a mic.

Simon Baker from Redburn. Two questions. Firstly, on Saxenda. It looks like that not only the sales growth but the breadth of territories in which it's launched is proceeding very, very quickly. But I was wondering if you're coming up against kind of a problem of your own making. The fact that you've got high-dose sema potentially just around the corner, which, on the face of it, should be a knockout competitor for Saxenda, is that a feature in any of the negotiations? I'm also wondering if that was an issue with the NICE decision in the U.K. around Saxenda. And also, does that have any broader implications for that rollout? And then secondly, on hemophilia. It's probably a bit early, but I wonder if you could, as you did for HEMLIBRA, give us some idea of the expected timing and magnitude of the impact on your hemophilia franchise from June 30?

So maybe if I start with some perspectives on obesity. And Mads, you can get into the NICE process and potentially also a perspective on gene therapy in hemophilia. So we have a situation today where in the U.S., it's largely employers who opt in, so there's reimbursement for Saxenda. And in most of the remaining territories, it's private pay. So that's an interesting situation that it's individuals who have tried a lot of things to lose weight and find that the profile of the product is actually justified by the price. Having said that, we also have to admit there is a lot of turn in patients, so the stay time is not high. And I think that's [ then attractive ] to consider the potential launch of semaglutide in obesity because there you have a continuation in the weight reduction, at least throughout the trial we've seen so far. And now we are doing the Phase III in a longer period of time. So I think the stay time is a function of the continued weight loss you see. Because as long as you lose weight, there would be an incentive to stay on medication. So from a business and a sales point of view, noting that we see a rotation of patients, I think there's a significant, say, upgrade potential in cannibalizing the Saxenda business with semaglutide, if you then inherently increase the stay time, what I know, 50 potentially more percent than what we have today. So that's for, let's say, the existing business. And then in terms of unlocking the market, it's clear that if we can double the efficacy, that will drive a lot of attention, and that will help further create the awareness that obesity is something that has to be treated based on medical intervention. It will create one more strong vehicle to actually have a platform for educating physicians and also in terms of bringing it to public payers. And now we're slowly getting into the NICE and the U.K. situation. Obviously, we also have more bang for the buck in that discussion. So I like to say that it's a positive. When you develop a market, you need a constant flow of education investment and use to do that. And I think sema business has the potential to do that. Then on the NICE process?

Yes. So what the NICE issue was there, the draft assessment, in February, the complete assessment will be done, and we are working closely with NICE until then to clarify a number of things and hopefully come out with a good assessment from NICE by February. But that is really too early to predict. I would say one thing along the line, what Lars is alluding to, that if you look at the facts that Ozempic and Victoza, despite their vast differences in terms of clinical efficacy are basically sold at around the same prices, it is true that the health care utility cost gains that we can drive in obesity trials with semaglutide, even maybe on the cardiovascular side, vis-à-vis the SELECT trial, et cetera, they will, of course, not come at twice the price of Saxenda, even though we're not talking pricing right now. But by definition, you will get more bang for the buck also when you have your reimbursement [ assessment does ] by the health technology assessors.

Good. On gene therapy in hemophilia and how that will change the market?

Yes. So there are some kind of scientific technological factors, and then there's some psychological ones. If you look at the science and technology of it, we are all aware that BioMarin has submitted the first [ Hem A ] gene therapy product. The big question is, of course, we're not -- at least I'm not an expert in how you do -- how do you actually price such a product. But by definition, it's a one-off treatment. So you have to have an assessment for how many years will these patients essentially be bleed free, i.e., have factor VIII levels at a decent level, let's say, above 10%. And what then does that cater for in terms of price? That's an assessment that has to be done on a case-by-case basis. We are all of us aware that it's only for adults with grown livers because it is episomal, you can say, plasmid delivery. It is not chromosomal integration, that will only arrive with gene editing years into the future. The other one that one has to bear in mind is some conservatism among the hemophilia treaters because these are hematologists that, first of all, are quite scientific. They individualize. They think a lot about how to treat each patient and which doses and which dosing frequency. They're a very, almost artistic bunch of people who feel it's an art to treat well. Of course, each boy -- or sorry, not boy. Each man, they take and put on gene therapy. They lose a customer. They'll only have them for follow-up, and they have to be convinced that it's long-term, safe, et cetera. So these are some of the reflections that they would be doing on a case-by-case basis. So I think our assessment is that it is not an overnight, big thing. But of course, the better the gene therapies and gene editing therapies become over time, we will, of course, see that this is something that takes off in immunogenic diseases like hemophilia.

Michael, did you get the microphone?

It's Michael Leuchten from UBS. You were kind enough yesterday to give us the number of countries that Ozempic is launched in, both U.S. and in Europe. So as we think about 2020, '21, how does the launch curve look like for the countries outside Europe, outside the U.S., so IO, but not Europe? What does the cadence look like there?

So it's without going into specific numbers because it's almost also like adding oranges and apples, then we'll be continuing to roll on many countries to come. Most recently, we launched in Germany, just not too many days ago. So I think we'll continue to launch. And of course, we have [ Japan ] also coming up. So we have a number of countries coming. In China, you should expect some years into the future because, first, we have to submit and get approval and reimbursement and so on. So going from 26 markets to close to 100 markets, where we are with Victoza, will take place over the next 4, 5 years or so.

But I can add that we have in the committee -- Development Committee also discussed the notion that we are and have to become better at relatively swift rolling out when we've done trials in countries. Historically, sometimes it's been a bit too many years between the first and the last launches, but it does take years.

That was feedback from the Chief Science Officer to commercial. Richard?

Richard Vosser from JPMorgan. Just following on Ozempic. Just in the U.S., where you're having the Rybelsus launch, where are the new patients for Ozempic coming from? Are they still coming from sort of first 2 injectable patients? Or are there a greater proportion that are using it on top of insulin? And then second question, just on Rybelsus and the new formulation work and the manufacturing improvements. Just where are we with the new formulation? I think we've had Phase I data, but where do we go from here and how the manufacturing is improving?

So if I talk to source of business for Ozempic, Mads, you could talk to the next-generation formulation. So as you say, we have launched Rybelsus. So it's not that source of business for Ozempic has changed significantly. That's still from other GLP-1s. It's in addition to GLP-1. And it's also some patients migrating from current OED. So it's not that we can say that the launch has actually changed that significantly. But again, it's early days. So we actually believe that there is a place for both products and is supported by at least a few -- it is, in essence, [ into weeks ] of experience we have. Mads, on...

Yes. So what happened last year was that we had the advent of the first Phase I data on new formulations. And we, of course, do not review details about what are those formulations. But in essence, we are constantly evolving our technology of how to make these tablets in a way that either gives greater viability and/or less molecules of whatever ingredient in the tablets, such that our dear colleague in a group can produce either more or cheaper or both. And we will come to yet another inflection point later in the year about which specific one do we then take into, let's say, the trial that is needed to -- for approval of a new -- you don't do a full program because there's nothing new, it's only the way in which we formulate existing ingredients.

Thank you, Mads. We had a question over here from Pete.

Peter Verdult, Citi. Two questions just for Mads. Tirzepatide started their CVOT yesterday. What could you say, if anything -- SOUL started last year, but can you say anything on recruitment rates? And what's the latest as to when -- or the earliest we might see a readout? I realize it's an event-driven trial, but anything you want to finalize or sharpen up there? And then secondly, look, you've always been very clear where your M&A focus is. And you have a framework that's been very principled and very [ Novo-esk ]. But in the current environment, we all see any asset with a pulse that's in demand. The valuations are probably not the sort of valuation Novo is used to paying. So should that basically mean that the likelihood of you actually doing anything M&A-wise is very low? Or are you willing to sort of loosen up your principles?

So on the CV side of things, yes, tirzepatid started their CVOT, we are -- and your question was more as to SOUL and timing and so on. The SOUL, here, we are a little bit on home turf because it's the same population as we've treated in both the leader in the SUSTAIN 6 and the PIONEER 6. So I can make my own predictions about the severity of their cardiovascular condition and diabetic condition. So this is a trial where we are typically a little bit on the conservative side in terms of how we power for both, you can say, event rate, but also the benefit that we achieve over and above standard of care. So hopefully, that caters for a good outcome of the SOUL trial. And since it is, as you correctly state, Pete, it's event-driven. That, of course, the more events that occur faster and to the benefit of Ozempic -- sorry, Rybelsus, the more that will help us in the duration of the trial. And then, of course, we also have to bear in mind that there are the other confirmatory secondary endpoints, the composite renal endpoint and also looking into peripheral arterial disease, PAD, both of which are indications of dire need of new medicines. So hopefully, it's the totality of data that will help the brand going forward by also bridging to Ozempic as we discussed earlier.

So on M&A, if you look at the broad spectrum, we are not looking to do late-stage deals because with the pipeline of products we have in development and our current marketed products, we believe we have what we need to have to sustain growth, short to medium term. And if you [ buy late ] these assets, that's typically based on some degree of desperation because the price you pay to get it can only be justified by that. If you then look at what we just spoke about that we're trying to broaden out our research efforts to build future growth platforms, then it's clear that we are more than maybe in the past looking for external innovation. In many cases, that can be accessed in terms of licensing deals, and that is based on that many companies find that Novo Nordisk can be an interesting partner. If you sit with the technology, we have a deep disease understanding, biology understanding. Combining that in a partnership is attractive in terms of generating the value of that technology for a partner. But if there is an area where it takes M&A to get access to it, we are also willing to do that. So it's the aim, it's the purpose of what it brings us that informs what type of structure we would be considering. But we are willing to do what it takes. But as you say, we are also disciplined. And the margin we have is a reflection of very strong organic in-house value-add to our R&D activities, and we are a strong believer in that. You have to be really sharp on what is the value-add from buying something and paying a follow-up, including a premium. Then you have to be really sharp on what are the synergies.

Chung from Crédit Suisse. First one on Rybelsus, touching upon one of the questions. Of the oral products, are you able to clarify which are SGL -- if your -- if these are simply SGLT2 or DPP-4 failures? Or are people moving immediately after metformin? And secondly, just on international Victoza, you still see healthy growth this year. We should see Ozempic entering. Should we use the U.S. as a reasonable comp in terms of what Victoza does for the year?

So if I start on the source of business and maybe Karsten, you can comment on Victoza. So I think it's too early to start interpreting down to that level of specificity in terms of what class we are sourcing from because I think the first few weeks, if you interpret too much on that, I'm afraid you run the risk of getting to wrong conclusions. So we can see that it's both naïve patients, but it's also from across -- the all categories. But again, I think we should be very cautious of that. So maybe when we meet a quarter from now, there's a bit more robustness to those data.

Yes. And on Victoza ex U.S., then what you generally see on products between U.S. and ex U.S., the main difference is the volatility on price. So ex U.S. pricing is more stable. So the Victoza erosion curve would, everything else equal, be slower ex U.S., than in the U.S.