諾和諾德 (NVO) 2018 Q1 法說會逐字稿

Haven't met before.

I cover Novo Nordisk for Goldman Sachs based out of the European offices in London.

It's a pleasure to have the Novo team here.

Mads, Peter, Karsten, thank you joining us.

I'm going to pass it straight over to Karsten to make some opening comments, and then we'll go to Mads and then to Q&A from there.

Karsten, all yours.

Thank you, Keyur, and thank you to Goldman Sachs for hosting this Q1 lunch meeting for the Novo Nordisk quarterly results.

And it's good to see all of you.

And we brought presentation pack for today, and we'll go through that reasonably fast, and then we'll take Q&A from there.

Forward-looking statements, I think you've seen this before.

The future is uncertain, and we have no chance to change that.

So that's what we'll continue to live with.

In terms of highlights for the first 3 months of 2018, then we reported what we believe is a solid first quarter in terms of results.

So we reported a underlying sales growth in local currencies, up 5%, which was driven predominantly by international operations with 70% share of growth and North America with 30% share of growth.

It is worth noting that the 5% online sales growth had a positive impact from timing of Brazilian NovoSeven tender shipment that impacted the quarter by around 1 percentage point on -- in terms of global sales positively.

In terms of our operating profit, then the 5% sales growth turned into 6% operating profit growth, also measured in local currencies.

We did have a one-off impact from sale of shares -- sale of 2 million shares in NNIT, an associated company, with a gain of some DKK 122 million corresponding to a 1% positive impact on our operating profit growth.

So in local terms, the results looks solid.

However, given the development on the U.S. dollar compared to the Danish kroner, then in reported terms, unfortunately it looks significantly less nice.

So our sales growth in reported terms was down 5%, and our operating profit growth was down -- operating profit was down 8%, so a 10 and 14 percentage point negative impact from currencies in the first quarter.

We are most likely going to see several negative impacts in the second quarter, not as big but still substantial, again driven by the U.S. dollar and then moving into the third and fourth quarter.

Then the comparator in terms of U.S. dollar, Danish kroner rate is more meaningful and not as damaging in terms of the reported numbers.

Key highlights in terms of R&D.

We now have Ozempic approved in Europe and Japan on top of the U.S. approval that we got in the fourth quarter of last year, which is important since this is our next-generation GLP-1 product that we have high hopes for in the years to come.

We can come back to that later.

Then we finalized our first Phase IIIa trial for our oral GLP-1, oral semaglutide, the trial we call PIONEER 1, which was against placebo with good results.

And then finally, for our basal insulin Tresiba, we got FDA approval for an update to our label related to hypoglycemia and reduction in severe hypoglycemia, which is a key differentiator in the U.S. basal segment for Tresiba.

So those were the highlights.

Moving on to the regions and regional performance.

Then if you take the right-hand side, the 5% local currency growth with international operations growing 8%.

Of the 8%, 2% is related to the Brazilian tender I mentioned before.

So back to this 5%, 6% sales growth in international operations that we have also seen historically.

Growth in international operations are driven by -- mainly driven by Latin America, China, and AAMEO.

So those are clearly our growth regions as in prior years that continue to drive solid growth.

Japan and Korea are down in terms of growth.

Part of it is the comparator effect from the first quarter of last year linked to a restocking of a growth hormone following a recall in late 2016, but still Japan has declining sales growth linked to the price reductions that we're seeing in the Japanese marketplace.

North America, 3% growth.

Significant growth in the GLP-1 franchise in North America with GLP-1s within diabetes growing 21% in local currencies and then also Saxenda in obesity providing substantial growth in the first quarter.

This we can see on a global scale here when we look at the therapy split on our products.

You can see that the majority of our sales growth in the first quarter is coming from GLP-1, being Victoza.

We have now launched Ozempic in the U.S. The sales in the first quarter are somewhat benign at DKK 69 million, which predominantly are driven by stocking linked to the launch of -- in the U.S. of Ozempic, so -- but all very solid sales growth of GLP-1s, 20-plus percent in the U.S. and 12% outside U.S. Then on top of that we have Saxenda, our obesity product, the GLP-1-based obesity product also growing nicely and driving 23% share of growth.

Hemophilia, positively impacted by the NovoSeven tender in Brazil, but we do see solid growth of our Novoeight franchise, also leading to a biopharm -- total biopharm growth of 1%.

If you adjust for the tender, then biopharm sales are down 5% in the quarter, which is in line with the adjusted underlying sales growth for the full year of last year.

Then we have started reporting our therapy split in a slightly different way.

So instead of looking at modern insulins and new generation insulins, now we're looking at the 3 segments within insulins; the long-acting, the rapid-acting and the mix segment.

And what we're showing here is that in the long-acting segment, we are seeing nice market share gains over the last few years linked to the penetration of Tresiba and the launch of Tresiba across markets globally.

The most significant development over the last 12 months is actually that we have taken 3 percentage point market share in the U.S. basal segment.

So good penetration there at a reasonably low level of 30% when you compare to the other insulin segment.

So good momentum in getting market share in the basal segments.

This we also see here in the U.S. marketplace, where we see a continuation of Tresiba market share, now at 12%, and a slight decline on Levemir, but overall, our U.S. market share is up now as you see it in -- measured in TRx to 34.8%.

What we're looking at currently in the U.S. and it's too early to comment on, but the updated label based on DEVOTE data and getting the hypo benefit for Tresiba into the label, making our sales representatives able to detail based on a hypo claim also, hopefully will help drive continued share uptake for Tresiba in the months and years to come.

And we also do have a very solid market access for basal insulin, so more than 80% unrestricted access for Tresiba in the U.S. marketplace.

Then moving onto GLP-1s, as you saw, our key growth driver in the first quarter.

Overall -- and this slide covers the U.S. marketplace -- overall, when we look at the market growth, market growth measured in volume, above 20%, those solid market growth which is driven both by continued uptake of Victoza and then Trulicity from Eli Lilly continuing to penetrate the markets.

And that you see in the middle chart in terms of total number of scripts.

So Trulicity penetrating, but Victoza also continuing to increase in volume.

Victoza has had the support from the updated label in the second half of last year with the cardiovascular benefit from the LEADER trial that we got on label and have been promoting for the last -- a little bit more than 6 months.

And now it's hard to see on the chart, but you see a small red dot there showing that Ozempic is now launched and measured in total TRx.

When you have hundreds of thousands of TRx, then a couple of thousand in a weekly TRx count or in a monthly count is not something that makes it on the chart, but hopefully, we do expect a significant uptake there.

And when we look at the weeklies and the NBRx share, we see a nice penetration for Ozempic.

With that, Mads, then I'll hand over to you and covering PIONEER 1 readout in the first quarter.

Thank you, Karsten.

I have a slight issue here and that is with the cable I think.

It should be okay.

Can you hear me?

Okay, thanks.

So when we met like a quarter ago, I think there were some concerns or anxieties surrounding 2 elements of the PIONEER program.

One was actually what would the efficacy profile look like and the other was what does the GI tolerability profile look like.

And in both regards, I think it was a great relief and excitement at least for me to see the PIONEER 1 data readout with the excellent efficacy data.

We are in a situation where if you take 14 milligrams of oral sema once daily, you will actually have an 80% likelihood of hitting the ADA target of going below 7%.

That is unprecedented for an oral anti-diabetic medication if you look into the literature.

So I would call it extremely compelling hemoglobin A1c data.

And when you look at the weight loss, it is not only the size, 4 kilograms up to 26 weeks, is the fact that there is a totally linear relationship over time and no sign of waning, i.e.

in the majority of the studies that are now going to be conducted for 52 weeks, you should expect or I should expect to see curves, where actually the weight loss that is unique for sema to continue week after week, will also continue unabatedly in these 52 weeks trials, at least until towards the end of these long-term trials.

So very excited about the efficacy profile, boding well, I believe for the future readouts in the PIONEER program.

And from a GI tolerability perspective, it seems like the titration regimen where you start with 3 milligrams instead of 5 milligrams, we kind of underestimated the potency of oral sema in Phase II, I believe, that has really turned out for the better.

I.e., we are seeing only up to max 16% nausea occurring in the population even at the high dose of 14 milligrams, which if anything, is on the low side as compared to the injectable counterpart studies, LEAD-3, DURATION, SUSTAIN 1, et cetera.

So I think we are entering a very exciting journey and that leads me directly to the next slide, which basically shows the story of the biggest program Novo Nordisk has ever done.

More than 9,000 patients entered into 10 trials, reading out typically after 52 weeks, some even 72 or up to maybe approx 2 years if we take the cardiovascular safety trial and everything is going to readout this year.

We are happy to say that to the extent that we're able to give you an update at the Orlando ADA meeting in June on some of these, you will be most welcome to witness these updates.

We'll do our utmost to see what can be done in that regard, but I think to be honest we're off to a really good and encouraging start of the PIONEER readouts.

Now that would then lead to a submission next year and launch in '20.

Now I think Karsten alluded to DEVOTE.

I mean, I've been around heading up diabetes research for, I don't know, 24 years or so and in my time neither we, nor Sanofi, nor Lilly has ever been able to get a label on the most feared acute complication of insulin therapy, namely severe hypoglycemia that can lead to hospitalization and death quite frequently, unfortunately still.

This is the first-ever documentation in a superiority claim in an insulin label of such a benefit.

40% event rate reduction, superior on severe hyperglycemia degludec versus insulin glargine U100, something that is going to be the basis of very emotive and very I think robust selling efforts going forward in the U.S. where we have not had a label with this documentation until about a month or so ago.

Exciting times also for Tresiba.

Now I'm not going to go through all of these milestones, because we would like to preserve time for the discussion.

Just repeat that triad market approval is now existing for Ozempic; U.S., Europe, Japan.

Maybe mention Ryzodeg being submitted in China, because China is a huge -- NovoMix is one of the biggest Chinese drugs in the whole market.

Ryzodeg is, as we know, distinctly superior and maybe a prominent player in the future of China.

LAI287 is a long-acting analog of human insulin that has just completed successfully multiple dosing in Type 2 diabetes up against Tresiba with, I would say, equipoise both on efficacy and safety, which bodes well for this once-weekly insulin.

And then maybe I also mention that in obesity, we have a once-weekly version of human amylin that based on a 7% weight loss after 8 weeks of therapy, basically is up there in the same league as semaglutide.

When I interpolate, which is a naughty thing to do, but if I interpolate the data on to the weight loss curves that we've seen for semaglutide 2.4 milligrams or what corresponds to that, this is up there at the most aggressive titration and weight loss we've seen even for sema in obesity.

So -- and since animal data suggests that the activities working by 2 different mechanisms of action, one in the hypothalamus, one in the hindbrain, seem to be additive, we could be going for a -- up to 30% weight loss in a combined product in human beings, all other things being equal.

And that was a forward-looking statement.

In terms of biopharm, the intravenous version of N8-GP has now been submitted both in U.S. and in Europe, and we are now into multiple dosing what we call Phase II-like study for the subcutaneous counterpart of N8-GP, which has this unique bioavailability after subcutaneous administration.

And then to help our friends, (inaudible) on the biopharma front, we actually have gotten 2 new labels, Prader-Willi Syndrome and Idiopathic Short Stature in the U.S. for Norditropin, which will give us hopefully a strengthened leadership position.

And we've licensed in a sickle cell disease program from a small, small, small company called EpiDestiny, but with a great promise, albeit at this point very small and fragmented data, it looks promising and Novo Nordisk has taken upon itself to develop this for the global community.

And with that, my last slide really just shows the news flow for the rest of year.

Basically speaking, growth velocity data in kids with Somapacitan, proof-of-concept data with Concizumab in hemophilia A and inhibitor patients, the subcutaneous antibody with a new mechanism of action and a total readout of the PIONEER program are probably the most noticeable things during the next quarters of the year.

Then over to Karsten for a wrap-up on financials and other stuff.

Thanks, Mads.

So just going through the financials.

We covered most of it.

So 5% sales growth in local currencies with a 10 percentage point negative currency impact for the first quarter.

Gross margin is declining by some 70 basis points.

This is all related to currencies, so our gross margin in local currency is flat, covering that we are impacted by price mostly in the U.S., which we're able to offset through product mix and productivity.

Investments -- sales and distribution costs are up 5%.

This covers that we are actually investing significantly both in North America and in international operations, more than the 5%, but as we wrote last year, then we had some legal provisions in the first quarter of last year that we don't have this year.

So significant commercial investments in the first quarter.

R&D, growing in line with sales.

You've seen kind of the activities Mads just described being the main drivers in terms of R&D costs.

And then on other operating income, DKK 351 million.

Of that DKK 122 million is related to the NNIT shares I called before.

That leads to the 6% operating profit growth.

Currencies are hit, so we lose to the tune of DKK 1.9 billion on currencies on operating profit.

Remember, we have that above the line, while we have hedging gains in net financials.

So of the DKK 1.9 billion we lose on currencies, we are able to recover DKK 1.1 billion roughly in hedging gains or some 60% in the first quarter.

Income taxes are down in terms of the effective tax rate by 90 basis points.

This reflects -- this is in line with what we guided and reflects the impact of the U.S. tax reform, which was implemented as of 1st of January this year.

All in all, this leads to a net profit up by 6% and earnings per share up by 8% following the continuation of our share buyback program, which corresponds to roughly 2% of shares on an annual basis.

Currencies, I covered the main currencies.

So the U.S. dollar is really what is impacting our reported numbers.

But in terms of side currencies, we also see negative impact on the Argentinian peso, the Turkish lira, the Russian ruble and also the Iranian rial currency.

And do remember, we're not hedging, that's why we've called non-hedged, but the side currencies, the hedging costs are simply too high to cover that.

So they're not hedged then and hence no gains in net financials for those.

This then leads us to our outlook for the year, where based on the passage of time, then of course uncertainty goes down and with the strong first quarter realized, then we are narrowing the guidance range both for sales and operating profit in local currencies.

So we take the floor up by 1 percentage point to between 3% and 5% sales growth in local currencies and operating profit growth between 2% and 5%.

Currency impact is slightly less than when we guided in connection with full year, linked to appreciation of the U.S. dollar, which then in terms impact our hedging gains which are correspondingly lower, being reduced from DKK 2.5 billion to at now DKK 1.9 billion.

And then the remaining part of our guidance remain unchanged.

So I will not go through our closing remarks but rather take us directly to Q&A.

So Peter promised he would help guide -- moderate it.

There is a mic.

And it's, of course, very important that you state your name and affiliation please and, even more important, try to stick to 2 questions.

It's Michael Leuchten at UBS.

Two questions, please.

One for cost, and just going back to your gross margin.

If I think about the products that I think are high margin products, you don't have Ozempic in that, that's now been launched.

You have price declines in the basal insulin.

Yet you have an underlying flat gross margin.

How do I think about that mix for the rest of the year and then also going forward, because previously the commentary was, do expect some pressure on the gross margin and I'm not seeing that in a quarter where I would have thought to see some pressure?

And then a question for Mads.

We've seen some interesting data coming out of the EASL on fibrosis in NASH.

The data suggests that you can maybe treat fibrosis a bit better than previously.

Does that mean that GLP-1s could actually have a role later in NASH?

Your commentary previously has been that the GLP-1s would have to sit in the earlier stages of NASH?

So I will start out responding to the gross margin question.

So when we dissect our gross margin development, then it's basically a price, productivity and product mix.

The pricing impacts will be there for the full year, then you can say there might be some phasing on the details, because recall, we had an adjustment in the fourth quarter of last year, so there might be something between the quarters.

But fundamentally we'll see the pricing impact for the full year linked to the U.S. price developments.

Productivity, we should also see for the full year and the same for mix.

So mix, it's Victoza, it's Saxenda and Tresiba being what has kind of the most material impact from a product mix perspective, slight, and then offset by NovoSeven.

Ozempic is too small to kind of meaningfully impact gross margin this year.

So for the full year in underlying terms, broadly speaking, neutral gross margin is what we're looking at.

And then, Michael, so actually, as you are aware, what we're looking into in the Phase II trials for semaglutide in NASH is predominantly stages F2 and F3.

So up to but not including true cirrhosis of the liver and since we don't have any data yet, I think it's maybe early to speculate.

We have, of course, the Lancet paper from the LEAN study with Victoza in a single-center done by Dr. Phil Newsome at the Birmingham, showing that you have this nice remission again from the F2 and F3 stages of the disease.

What we have more data on is actually some of the mechanisms that GLP-1 seems to use when it counteracts fibrosis in animals.

These are studies that we both do with a collaboration partner on NASH models, but that we also internalize in-house where we look at all kinds of signaling mechanisms that lead to collagen deposition and hence fibrosis that is reversible originally or to begin with and then turns irreversible over time at least in humans.

And there, we are actually seeing a targeting of several anti-inflammatory pathways from NF-kappaB, all the way through to cytokine inhibition, MEK1 inhibition and things like that.

So I think part of the -- if there is an anti-fibrotic mechanism that would then justify that sema or another GLP-1 could also be used at late stages of the NASH condition, it would probably be a mix of different effects, some of which are actually either occurring at the level of the kind of sinusoidal macrophages or stellate cells or even infiltrating leukocytes.

So it's a very complicated story.

I would agree with you that I think we should not underestimate the anti-fibrotic effects in different conditions such as kidney disease but in particular also liver disease with the GLP-1s, and we are working more.

And so I might be more crisp later on.

I think we should move to the host, to be polite.

Keyur Parekh from Goldman Sachs.

Karsten, do you want to just clarify, there was a lot of confusion yesterday on your comments about the outlook for the basal insulin market.

So can you just clarify, do you expect basal insulin segment for Novo to grow in the U.S. this year?

That's one.

Do you expect it to grow globally?

That's two.

And then any commentary on longer-term trends for the basal insulin franchise?

Yes.

Thanks for that question and clarification, Keyur.

So when we look at our first quarter results, then in local currencies, our basal insulin sales are down 3% globally.

That covers a 13% growth in International Operations and a 10% decline in U.S. So if we take International Operations first, we have solid momentum on our basal insulin franchise and continued penetration behind Tresiba but also start to find some countries, for instance in France where Xultophy is penetrating strongly, so continued momentum in the basal franchise outside the U.S. In the U.S., the 10% decline there has to be seen in the light of the slide that -- let me just click back to the market share slide that we have here.

So when you look at that slide, the dark blue line, the market share development in the basal segment in the U.S. then we see a solid momentum in terms of gaining market share in the basal segments.

The main drivers behind that are twofold.

One, it's of course launching Tresiba, penetrating into the marketplace, so having the latest-generation basal insulins with the benefits associated with Tresiba being recognized in the marketplace.

So that's one.

And then two, from a market share perspective, then the formulary changes that has happened over the last couple of years in the U.S. where certain formularies have been displacing Lantus and putting BASAGLAR on, which has actually had a positive share impact on our business -- our basal business because then we've been able to capture higher share on those formularies.

So in terms of volume growth and share penetration in the U.S. marketplace, then we are continuing to track well and perform there.

Then what comes together with the volumes is, of course, the pricing component when we report sales and how do we get to increasing volumes and growing market shares 3 percentage point and then to a 10% sales decline.

That's pricing in the basal segment, and that is a function of the fact that BASAGLAR launched and some of the formulary changes I called out just before, so which means that the price point in the basal segment is going down and that is the driver behind our lower U.S. sales.

But we had strong market access with more than 80% access behind Tresiba.

We have the DEVOTE label coming in, so we are committed to continuing to take share in the U.S. marketplace.

And maybe, Keyur, if I can just add, it's quite clear that we've been waiting to kick off these global hypoglycemia campaigns, both unbranded awareness and also branded awareness specifically.

That's something we've never done and there is a clear underestimation of the size and the magnitude and the severity of the hypoglycemia problem in Type 2 diabetes.

And the fact that we now have a kind of globally, you can say, ready-to-go campaign that initially is kicked off unbranded and will then trickle over into Tresiba specific 40% severe hypo reduction, that will allow our sales force, first of all, much more strong an emotive selling and also much more awareness among -- one of the problems has actually been that many physicians and patients have underutilized basal insulin, so they've given so low doses, so that you don't see the hypoglycemia problem.

And here you actually stand with an insulin where we can make the argument that you can stay in a very nice zone of near-normal glycemia without going into hypoglycemia.

And that should not only capture more patients, but also give more assertive dosing of the insulin at higher unit doses and controlling glycemia better.

So this is actually a multifaceted problem we're targeting, but that will all lead to the benefit of the patients and of our market share and consumption.

And then just to close off with your medium-term question.

So we're not guiding on the medium term, as you know, for specific brands.

The pricing will continue to be under pressure in the U.S. basal segment due to the market dynamics and market entrants, but that doesn't change the fact that we're driving to gain share and get Tresiba accepted as the leading product in the U.S. marketplace.

So I think we have good momentum and we keep pushing that.

Simon Baker from Exane.

Two questions.

Firstly, for Karsten.

You said yesterday on the call that the 2019 impact from the shrinking of the donut hole was about 1% to 2% of group sales, which is about $200 million to $400 million.

The Congressional Budget Office estimates the benefit from the government of that 20% shift from them to you as an industry is about $400 million, which would suggest either that estimate underestimates the cost to industry or your estimate is on the high side.

So I'm wondering if you can give us some view or thoughts on how you arrived at your 1% to 2% estimate.

And then a question for Mads.

One of the hot topics across the industry at the moment is gene therapy.

I'm just wondering if you could give us your updated thoughts on where we stand from a diabetes point of view with gene therapy, where Novo stands, the approaches we've seen data on this year in preclinical stage on things like alpha cell reprogramming?

So in terms of the impact from the changed regulation regarding to the donut hole and the pharmacy of that, then I've been around this kind of technical calculation for quite some time, because the coverage in the donut hole was implemented as part of Obamacare back in 2010.

So that is where pharma was told to cover 50% of the doughnut hole.

So the way for us to kind of -- I understand your data discrepancy and I'm not going to comment on how the other side has calculated that, but just to give you a couple of data points in terms of supporting facts on how to assess it on our side.

So first of all, you can simply -- you could take the -- what our competitors have been out calculating, that is clearly independent, right, what Lilly and Sanofi have been out saying.

Then you calibrate that with their exposure in diabetes and the relative sizes.

Then I think then you'll get a number that's not too distant from somewhere within our range.

So that's one data point.

Of course, we didn't have that initially when we calculated our own exposure.

So how we calculate our own exposures?

We know how much we paid for 2017, so the 50% in 2017, we paid the last invoice around 1st of February.

It's a quarterly invoicing, so we had the hard number for 2017 based on the 50%.

And then going from there to the 70% and then you adjust for utilization, segment growth and prices, et cetera, then that's how we got to our estimates.

And then, of course, you can discuss whether the estimate is right or not, but of course there are still uncertainties about what formularies, who gains what formularies for next year and so on, but I think this is kind of the broad range we're working with that at this point in time.

For you to get a feel for -- if you wanted on a rainy afternoon in London to try and calculate back into this, then one way to kind of triangulate your way into it is basically to say our U.S. net sales, if you take that last year, then you adjust for our gross to net ratio of the 64% you see in our annual report.

Then you get to somewhere around [DKK 150 billion] in gross sales in the U.S. Then you take somewhere between 25% and 30% being in Part D, and then you take the share of Part D volumes that are used in the donut hole, which would whatever, be around 20% or so.

And then, then you get to somewhere in the range that we indicated.

And then on gene therapy, if I may rephrase your question a little bit, because gene therapy per se in diabetes, despite the very significant advances that we fortunately for the patients are seeing in diseases, such as hemophilia and a few others, they have the commonality that they are typically so-called monogenic traits or monogenic diseases where you can actually wrap the DNA sequence for the gene in question into a viral capsid and boom, away you go and potentially you have a curative therapy for that disease.

But if I lump together all the monogenic versions of type 2 diabetes or diabetes per se, it only accounts for maybe 3% on a good day of the entirety of the diabetes universe, the LADAs and the likes of it.

And that means that they are simply too small to be the targets in my view for gene therapy.

Now if what you meant was curative therapy, then I would actually lump together gene therapy, cell replacement therapy and even ex vivo gene editing and combinations thereof.

And there, it's a different story.

Because if you look at diabetes, we have actually documented in a very robust and a very, you can say, industry-conformed, FDA-compliant way that we can actually cure rodent diabetes with a stem cell-based approach, which has been flying under the radar for the past 22 years or so.

And that is something that we're really diving into now, looking into our stem cell -- embryonic stem cell technology platform as something where we can, in a matter of few weeks, replicate whatever -- what the nature has taken 9 months to do in utero, so to speak.

And that might for type 1 diabetes, if we can make sure that we also have some insulation against the immune system, actually be a curative type 1 diabetes therapy for those folks who really deserve it.

So this, I think, you'll hear more about, but gene therapy, not so much in type 1 diabetes.

You mentioned the thing in your last sentence that I forgot what it was -- alpha-cell reprogramming.

Yes, that is actually something that is interesting and we're looking into.

You have to bear in mind there are big differences between alpha cell, beta cell reprogramming in rodents and in humans but interesting science that we can discuss maybe a bit offline.

Wimal Kapadia of Bernstein.

So if I look at the realized price for Levemir for 1Q '18 and I compare it versus the full year for '17, on a per unit basis, is down 19%.

And if I adjust Tresiba for the one-offs, rebates and contracted -- uncontracted volume, it's also down 19% versus the full year of '17.

Is minus 19% a fair reflection for the rest of the year in the basal market?

And then the second question is for Mads.

If I think about oral sema and comparing it versus the SGLT-2s, how should I think about what the key -- the most important characteristics are?

So if I had to rank HbA1c weight loss and tolerability, now how important are each of those, how would you rank those?

Because I guess the number one question we had post the PIONEER 1 data was that the weight loss didn't look that great.

So is that the third of the 3 in importance?

Thanks.

U.S. basal insulin segment pricing.

So first, when you combine TRx data from (inaudible) or Symphony Health with ex-factory data, then of course there are quite some steps in between and some uncertainty.

So doing that analysis, you always have to take that just with a grain of salt.

So it is directionally correct what you're getting at with the size.

We're not commenting specifically on realized price in specific segment, but it's directionally correct.

Do bear in mind that for Tresiba, we had 2 effects that we just need to adjust for the Q4, and the non-contracted business.

So that's why if you take the total basal business being down 10% in the U.S. then adjust for these special items, then in the first quarter we are more around minus 6% down for the basal business in net sales.

And then if I dwell a little bit on -- you're in particular hinting at the PIONEER 2, in particular [closing] comparator study that is going to report late this quarter.

First of all, yes, let me just show this one.

What you have here is the data from PIONEER 1, and what I cannot show you, because we have to reserve that for scientific publication and conference presentation, but the underlying data behind those columns showing the 4.1 kilogram weight loss after 26 weeks, the 14 mg dose, that is actually composed of a very straight line going like this and PIONEER 2 study is actually for a full year.

And I know from semaglutide as a molecule the way it works on hypothalamus, the way it penetrates to bring the weight, continues to give weight loss that I would expect that over a full year, a more sizable weight loss with semaglutide and one that is more continuous than I would see with an SGLT-2 inhibitor, where what happens is that as the SGLT-2 inhibitor takes down glucose levels, there's less glucose to pee out in the urine, so that comes a waning of the weight loss effect that typically levels off at around 2 to 2.5 kilograms or so in steady state for the SGLT-2 inhibitors.

So I would say absolutely I would put weight in not as the number one, but for many patients it might be number one.

If you are a cardiologist, it might be the cardioprotective action that is atherosclerotic for sema and that is anti-heart failure for SGLT-2.

So they match each other, but in different ways and for different patients.

And in reality, it's a win-win scenario for both agents in that regard.

When I then look at the overall tolerability, I'd actually put it to you that with some intraclass differences between the (inaudible), we can look into the labels and say there are elements of slight risk of diabetic ketoacidosis, some overrepresentation of amputations, some genito-urinal infections, such as candidiasis and bladder infections, in particular in women.

That when you add it all together in terms of safety and tolerability compared up against an oral sema where we have a very low rate of nausea and everything else being at the placebo level, in my view gives, if not the upper hand, then a very strong hand to oral sema.

When I then look at the hemoglobin A1c, the 1.5%, this is a placebo-corrected 1.4%.

This is close to double of what you see for the other orals typically aside of maybe metformin.

So I would not rank them, but I would say the primary end point and also in the statistical hierarchy is hemoglobin A1c control, followed by body weight control and the doctor will typically rate A1c control number one.

The patient will, quite often, because they see, feel and notice the weight loss, they may rate that as number one.

And again.

if you're a cardio, you'll rate the CV protection.

So I think it's an overall benefit-risk profile that in my view clearly favors this as an unprecedented oral agent.

But that is forward-looking disclaimer statement.

We have to show the data and hope to give them to you at the very least at Q2.

It's Mark Purcell from Redburn.

Two sort of broad questions.

The first one on the GLP-1 U.S. market, there's been a dramatic acceleration if you look at NBRx.

I think Lily talked from a TRx perspective volume growth going from 23% to 27%.

Obviously, 2 factors have changed, one is the CBO label for Victoza, the second one is increased promotion behind Ozempic.

So to the best of your ability, how much do you think is coming from the CBO claim?

How much of a game changer is that?

Because clearly, if you're shifting promotions from Victoza to Ozempic through the year, you'll shift promotion off that label claim, so I'm just wondering if that's a wise idea.

That's the first question.

The second one is on DEVOTE and the message that you now can promote on the label.

How broadly is this going to be promoted and how important is this for Novo Nordisk?

You presented some interesting data last year at EASD when you were looking at the proximity of severe hypo to cardiovascular event.

Obviously, we've got the CAROLINA study coming up this year as well.

Everything points to a CV benefit or disbenefit, rather, for [sulfonylureas].

So how forward are you going to go after that?

Should you be doing a trial of oral sema versus SUs in high-risk CV patients and showing there's a huge opportunity to shift that 19% of the volume to banal agent, which is not killing people?

And I was going to slip in one if I may, priority voucher.

Can you give us an idea of where this has come from, how much you paid for it?

And is there an opportunity to use this and file early on oral sema and get ahead of the contracting negotiations in 2019?

I know that's not the official comment, but it seems that you could file at the end of Q3, beginning of Q4, which does get you in front of payers next summer using your priority voucher?

So I'll start out with the GLP-1 market in terms of volume and the volume development.

And I think the 27% you're alluding to is something you have from some of our American competitors.

But when we look at the market data, you can always discuss, but it's in that neighborhood.

So we see strong underlying volume market growth.

We are out saying 23% on a MAT basis and I think Q1 is slightly higher.

Then trying to decipher what comes from what is of course something and the question we also pose internally, not only in the U.S., but when we look at German performance et cetera, and it's simply trying to separate hot and cold water.

So how much is the label update, how much is better productivity from the field force because they have something new to talk about, how much is increased DTC level.

So I think it's safe to say that we see this continued 20-plus percent volume trend, very high level of DTC and continued field force pressure from us and Lilly.

And then the impact from the Ozempic launch, you can always speculate and you see that across classes whether a launch will increase market growth et cetera.

I think with the current level of market growth and the market size being where it is, I think you should not expect oral market growth to come up further, but of course having a new product into the market and perhaps sustained market growth for some time to come.

So that's on the volume market perspective.

Then on the priority review voucher, if we take, first of all, the price we pay for it, we're not out saying that publicly, but I think when you look at the deals that are public, then we paid a market price which is pretty much in line with what you can see out there.

And if I want to give you a little bit more guidance and if you read through the details of our accounts on intangible assets and so on, I think you can get a pretty good feel for what magnitude we're in.

But we're in the $100 million type range.

Then what product it will be used for and et cetera, there you can say from a strategic point of view, the good part there is we don't have to decide that when we get to submission timing.

And of course the value of the voucher, if you take oral sema, then there's a value in terms of getting into, for instance, Part D contracting season for the following year, which would entail that the value of the voucher is higher, if we can get an approval in whatever, Q1-ish to get into the next year.

There's also a value in the voucher in terms of sema obesity potentially, simply linked to the potential launch timing of sema obesity, compared to Victoza -- sorry, Saxenda patent expiry.

So that's kind of -- we're doing the financial modeling and saying, how much more the value of peak sales in the various scenarios are we getting and then we'll make a decision when we get closer.

And then, Mark, first of all, just a quick notion on CV and GLP-1.

It is so after all that the 6.6% versus 8.9% event rates are actually mentioned in the Ozempic label in the U.S. And according to the Freedom of Speech Act, you are able to actually leave behind the paper -- The New England Journal of Medicine Paper on SUSTAIN 6 with the cardiovascular and so on.

So it's not as if Ozempic is devoid of cardiovascular benefit in the minds of even the American physicians.

That's one point.

Doing a cardiovascular trial, sticking now to the CV thing, up against sulfonylurea, I think it's creative of our friends and colleagues to do that.

I would not do that.

I would just prefer to go up against standard of care whatever that may be.

And right now, we are in the -- this very good and positive discussion with the regulators about which cardiovascular outcome trials to do to get a strong cardioprotection label on the various products, whether it be Ozempic U.S., whether it be semaglutide obesity or whether it be oral sema, and we'll provide you an update as soon as we can on that one.

In terms of oral sema, speaking about that, do bear in mind that it's not an enemy.

It's a new route of administration for an existing molecular entity for which reason the anticipated approval time lines would normally be 10 months as opposed to 12 months.

And then if a priority review voucher were to be used, you would deduct 4 months from that.

And of course, as Karsten is saying, that all depends also on the strength of the PIONEER data that fortunately will emerge in the months to come because then it's much more easy for management to make up their mind what should this particular voucher be used for.

Then when it comes to DEVOTE and Tresiba, again, I may be getting old, probably I am, but I have to say I've been waiting for 20-plus years for an insulin to show what the FDA in their guideline write, show that insulin does the job of lowering glucose and has an added benefit and then they say typically in the reduction of the occurrence of severe hypoglycemia.

That's what we've got after 20-plus years of trying.

So I think this is a very strong selling message.

And you're very right in saying that we were out at the EASD meeting, talking to the Hypo-Triad, the fact that glucose variability is reduced with Tresiba for a number of reasons that is associated with a reduced risk of overall, but in particular also severe hypoglycemia to the magnitude of 40% at night -- sorry, overall and 53% at night.

And that is at least associated with improved cardiovascular outcomes, this triad between variability, hypo and CV morbidity.

So that is something that most specialists understand we should not overcomplicate the issue.

But personally, I do believe with all the ECG changes that happen during hypoglycemia, it's highly relevant to speculate that this is what's happening.

And that gives Tresiba a very strong hand in long-term safety.

Sachin Jain from Bank of America, a few questions please.

Firstly, just to follow up on the oral sema weight discussion, if I may.

If I understood it correctly, we're referencing a different time point versus the data we're seeing for PIONEER 1. So can I just confirm the primary end point on weight is at 52 weeks not the A1c 26 weeks, just to make sure I understand that correctly?

And then for A1c, we typically think of a delta sort of 0.3%, 0.4% as the total non-inferiority margin.

Any color on what you would view as clinically relevant as a weight difference as you power the study?

And second question on the short-acting insulin franchise as we head into next year, Sanofi talking on their call that they've entered contracting for '19.

So any early color on what pressure we should expect there?

And you remain of the view that they should be different from the basal dynamics that we've seen.

And then the final question is on the oral sema yield enhancement program, and I know you've talked about this at a very high level of a CMD.

But just any color over the next 2 years, what level of yield enhancements you're targeting, how you're going to measure that and what communication we can expect from that?

So Mads, do you want to start out?

Okay.

So Sachin, when it comes to the primary end point, the hemoglobin A1c, I think you may well argue that doing a study for 52 weeks, there are good arguments in favor of using a 52-week cut-off, but these are also negotiations with the regulatory agency.

And in that regard, they typically do a 26-week readout for A1c outcomes.

And our statisticians, as they looked into this, also had mostly data on the other OADs after 26 weeks.

We had 26-week data after our own Phase II program, the one that was published in The Journal of the American Medical Association.

So I think that's what we've powered it up against and are measuring on.

In terms of the difference we need to see, well, we actually believe and again, here the guidance is a little bit, you can say not easy to understand, because sometimes agencies talk about 0.3% as you state, sometimes 0.4%.

Clinically speaking, greater -- anything greater than 0.3% is of clinical significance, is the view of most physicians I have spoken to.

And that does mean the power we have is strong, around 90%, to detect even an 0.3% difference.

Weight at 26 weeks or at 52?

Because I think the answer to Wimal's question was you're confident on SGLT-2 weight differentiation given duration of trial.

So is the weight end point of 52 or 26, and so weight delta not the A1c?

So Sachin, in the publications we're going to issue, we will have the weight curves and statistical comparisons at week 52, so we'll get the absolute difference and relative difference at week 52.

In the hierarchical testing, do you remember, Peter, is it 26 or 52?

Yes, so it's -- both the primary and the preferred secondary, that is 26.

But this is due to a regulatory discussion that they prefer this.

Doesn't mean to say that when you publish this and present it the full year data and will payers -- payers typically like to see at least 1-year data because it gives them more about the durability.

So doing 52-week studies are critically important for market access and for health economic outcomes and for quality of life and so on.

But in the regulatory label, it's defined as 26 in dialogue with the agencies.

So then in terms of the short-acting contracting space for 2019, we are in negotiations for '19 now.

So providing too detailed comments for how things are would not necessarily be conducive for our own negotiating positions with the payer.

I think the key difference between the basal and short-acting segment is that the short-acting segment has pretty much gone exclusive already.

So from a negotiations point of view, for -- it's one of our exclusive contracts that Admelog should display NovoLog in the U.S. -- on the U.S. formulary next year.

Then a payer, to make that decision and to make that work, then first of all, the payer would have to figure out what do they do with the mix insulin, because there is no Admelog mix from Sanofi.

So how do they handle that in a contracting segment, because until now that has been always been contracted with the short-acting?

So that's one piece.

The other piece is, since the business is exclusive now, then they would have to flip their entire patient population from NovoLog onto Admelog.

And then it's always a big question for the payer in terms of how many patients will for whatever reason, medical exemptions, et cetera have to stay on NovoLog at full list price.

And then finally, when you are a payer, then you also want to kind of consider the disruption factor.

So if you force many patients to switch their medication from say NovoLog to Admelog, then there's a lot of disruption on a patient basis that they have to change the medicine that they've been using for a number of years, so that's disruption.

And it's hard to say the dollar amount on it, but that's something that the payers are also [post] about.

So to make that equation work, then the payer needs to look at what kind of -- what price is realistic to make that work in an already competitively priced segment.

So just to provide some color around the dynamics.

Then I lost your final question.

Yield of oral sema.

Yield of oral sema.

So as you saw in our announcement then, we don't want to comment on price of oral sema at this point in time for all of the classic reasons in pharmaceutical product development.

That said, we are at full speed in terms of improving yields and that's something that we have done pretty much for decades in terms of optimizing insulin and GLP-1 yields in our current facilities.

So this is kind of that's in our genes, and we keep doing that big time.

And for oral sema, it's even more relevant in terms of a patient population, because taking down the unit cost of oral sema will hopefully make it feasible to launch the product into many more countries than what we are able to do with the current costing of the product.

So I think given the fantastic clinical profile that are kind of emerging in the PIONEER 1 today, then if we can take the unit cost down that will create an immense platform for the company going forward.

In terms of specific timing on when we're going to comment further on that, then I think we have to get closer to when we discuss pricing of the product.

And maybe just a technical comment, Sachin, if you're making devices like single-use devices or pens, there's a lower limit even with volume increase as to how far the unit costs can come down, because metal springs, plastic parts, glasses, they all cost a certain amount of DKK.

What we are working on here is, the traditional Novo Nordisk way of yield optimization, but it's also actually taking technology leaps in terms of even reformulating, making patented versions of something that is further optimized in terms of tablet formulation composition, way of making it, and even the bioavailability of the molecules inside it.

So the dream is of course to make this as competitive from a cost perspective one day as even the injectable counterpart, because here you don't have the device, but you can work on the API and the other ingredients.

But that's a long haul approach that is considered a must-win battle in the R&D labs.

Do we have a final question before we close off?

Thibault Boutherin from Morgan Stanley.

Just a couple of questions on your pipeline asset concizumab for hemophilia.

Can you just confirm the explorer 4 and 5 studies will be pivotal, and as well, what are your expectations for this product?

So split and use daily in the context where we're probably going to have HEMLIBRA approved next year in both patient population.

So what kind of specific patient population or specific situation do you expect this product to be mostly used?

Well, first of all, these are actually Phase II proof-of-concept studies.

So what we are doing is having a target product profile that has to be seen in light of the HEMLIBRA, the ACE910 emicizumab era that we're living in.

So that also means, by the way, that what we've done is actually to make sure that we don't end in a situation where patients who bleed on concizumab cannot be used or treated on demand with one of the other agent.

We've actually specifically started the trials by testing out can you actually co-administer NovoSeven and concizumab in a safe way?

Short answer is affirmative.

We've done that.

There is no drug-drug interference in terms of thrombosis or the likes, at least in those patients that we have tested.

So that is actually the kind of the first readout from a safety perspective.

These are actually, one study in hemophilia A and in the other study, it's hemophilia with inhibitor patients, both hemophilia A and B. And the target product profile is to have a simple subcutaneous administration that is safe and efficacious.

And that means basically meaning a ABR, an annualized median bleeding rate of down in the 0 to 1 range that we are also seeing for -- in concizumab.

Do bear in mind that this is a new mechanism of action.

It's a humanized antibody against tissue factor pathway inhibitor.

So it's kind of disinhibiting the brake that is constantly on the system.

And that means that it will work across segments.

You are no longer relying on the need for Factor IXa to be present on the activated platelet surface.

So in principle, this can be used also in hemophilia B unlike HEMLIBRA.

So the really exciting thing is, to be honest, the explorer 4 and 5 readout.

Because if they are good, then we are moving directly into pivotal trials starting next year and make them as snappy as possible, realizing that we are behind our friends at Roche.

Good.

Thank you, Mads.

This concludes the lunch session for the Novo Nordisk Q1 financial results.

Again, thanks to Keyur and Goldman Sachs for hosting us.

And we look forward to see you again next quarter.

Thank you.