諾和諾德 (NVO) 2019 Q3 法說會逐字稿

All right.

There's many, many familiar faces here, so I'm not going to do a boring introduction.

We all know who each other are.

But delighted, on behalf of Citi, to have management from Nova with us.

We all heard their results on Friday.

Probably listened to the conference call.

So a few remarks then hopefully a very productive Q&A session.

So without further ado, Karsten, over to you.

Thank you.

Great.

Thanks, Pete, and thanks for Citi for hosting this Q3 launch event.

So I am Karsten Munk Knudsen, CFO of Novo Nordisk.

With me, I have Camilla Sylvest, our Head of Commercial Strategy and Corporate Affairs; and Mads Krogsgaard Thomsen, our CSO.

So I think you've met all of us before.

So we brought a deck of slides.

Many of you would have seen them before.

So we'll kind of just to get this started.

We will go through them in a somewhat snappy pace, and then we'll get to Q&A after that.

So just making this work.

Forward-looking statements, so you know the drill.

The future is uncertain, and same goes for Novo Nordisk, even though that we have one of the more predictable outlooks than many companies in terms of market share, stability and the like.

Highlights for the first 9 months.

So in short, fast sales growth driven by IO and our GLP-1 business.

R&D events, Rybelsus would be up by far the biggest R&D event in the third quarter, approval of Rybelsus.

And then in terms of financials, so the 5% top line delivered 5% operating profit growth.

And then we had currency tailwinds mainly through the U.S. dollar but also some other currencies.

So 11% reported operating profit growth.

Outlook.

We narrowed our guidance for top line by raising the floor, so now between 5 and 6. And operating profit growth unchanged despite the fact that we take a DKK 1 billion impairment charge in the third quarter.

So with that, I'm handing over to Camilla to go through our commercial performance.

Thank you, Karsten.

So here's the sales growth of 5% and how it's split per region.

You see IO growing 11%, and in IO, all of the regions are delivering growth.

So Region Europe with 6% growth, mainly driven by GLP-1 franchise.

Then we have Region AAMEO, that is driving 16% growth, a lot of it coming, of course, from insulin.

And then Region China with 13% growth, a big majority of that coming from, of course, insulin as the base but also with strong growth on Victoza.

And then Japan and Korea also driving growth, 6%; and Region Lat Am driving 26% growth, a big part of that also coming from our Obesity franchise.

And then in the U.S., you see a decline of 1%, but the underlying growth is, of course, coming from GLP-1 and also Obesity.

And then we have, as you would see on the next slide, a negative growth in the insulin space.

So here, you see the same 5% just distributed by therapy area, and you see that insulin is growing 8% in International Operations.

Strong market growth in many of the regions here also, and we are gaining market share as well.

In the U.S., the market growth is more flat, it's flattish.

And you are seeing, because of the price pressure, a minus 15% total growth.

On GLP-1, we see 20% growth; in the IO, 28%; and in the U.S., 17%; and a strong volume growth in both of these commercial areas.

Obesity, 50% growth year-to-date, big part of the growth coming from IO and also with high growth rates and 29% growth also in the U.S. And then biopharm, that is now back to growth with 4% and relatively flattish in the U.S., but in IO growing 6%.

So when we look at how that all turns into market shares, you will see on this slide that we are expanding our total diabetes market share with 0.8 percentage points on an MAT basis.

That comes from primarily the U.S. But when you look at insulin, we are driving insulin market share in both North America and IO and thereby driving our total insulin volume market share.

When we look at our GLP-1 market share, you will see that recently, on a rolling 3-month basis, we are also driving market share in -- on a GLP-1 basis, but not on an MAT basis yet in all regions.

But on a total level, all sums up to 0.8% market share in Diabetes.

Here you then see our -- the results of our Ozempic launch so far, and you see that Ozempic now has 37% NBRx in the U.S., so just 1% point short of Trulicity in the NBRx segment.

But on the total GLP-1 market share, TRx, you see that we now have a 2-percentage point lead.

So that, of course, is as strong as both Ozempic and Victoza.

And you, of course, also see that this is a mix of the decline in Victoza and the uptake of Ozempic that at least saw stronger performance in the GLP-1 segment.

In Europe, we've also seen good launches.

Ozempic, you see here the uptake curve market share across Europe and generally a very consistent pattern and a very -- product that has been very well received by our customers.

You see, in some countries, a very fast uptake and others, as we expected, and that mainly has to do with market access.

So countries where we have to negotiate market access on a regional basis takes a bit longer to get to the same market shares as it does in countries where we have a 1-payer system.

But generally, this is in -- going according to our expectations and looks very well.

It also means that in Europe, we are now having our share growth very close to our total market share, so 48% compared to our market share of 56% and likely to increase as we are rolling out in even more countries also the rest of this year and next year.

On Saxenda, you will see the 50% growth year-to-date.

We have now launched in 45 countries, and we continue to invest in Obesity.

In Obesity, we, of course, are more focused on how we can develop the market rather than our market share as we already have a lead in market share and value of more than 50%.

So what we are focusing on is really how we can make sure that we drive the building of the market.

And there are a number of points in that, that we might be able to talk about later, but it relates, of course, both to patients and to physicians and also payers and policymakers to really open up for the treatment of obesity in larger scale.

A strong growth from both of our commercial units, so North America and IO.

Then on biopharm, with 4%, really driven by, you can say, a slightly better performance on NovoSeven and then also on NovoEight and Refixia.

There are also, really, the growth drivers that we are looking at in the biopharm market, and that all sums up to the 4% growth year-to-date.

And then I'd just like to hand over to Mads to talk a little bit more about the future in terms of the status on Rybelsus.

Thanks, Camilla.

So as you are all aware, Rybelsus is approved.

We are starting to launch it in the specialty segment.

We have a nice label.

It's much appreciated by those who have reviewed all the clinical data, including the medical staff at some of the PBMs, of course.

We are basically awaiting a CB decision on the PDUFA action date, which is late January.

That goes for this company, also goes for Ozempic.

They were simultaneously submitted and are reviewed by the same review teams at the agency.

And then in Europe, we are expecting a first-half approval, or hoping for a first-half approval with the European Commission and Japan around midyear by the PMDA.

So no surprises on Rybelsus, and we are extremely encouraged about the future prospects.

Now moving into the late-stage 3 program concizumab.

As you are aware, has potential to be the first-in-class TFPI -- sorry, antibody.

So it's a basically cross-segment agent that will be supposed to work in hemophilia A, B and inhibitor patients, i.e., without any restriction in terms of which kind of hemophilia patient suffers from.

And that's also the way we are developing it in the explorer7 program that is ongoing, and the explorer8 for noninhibitor patients that will be ongoing over the next month or so.

We are going to enroll a lot of patients across these segments, about 265, in a number of countries.

And the first data should come in early parts of 2021 followed by as fast submission as possible.

It's initially developed in a single-to-use auto injection pin device for once-daily, low-volume injection with a very fine needle, and that's what is the basis for explorer7 and explorer8.

We then look at emerging markets.

Rather than take you through all of the things, communication, why is that you should follow over the next quarters, I would argue this, of course, the status on the CB indications for Rybelsus and Ozempic in January of next year.

But as we move more into the year, it's, of course, the exciting data across the Obesity portfolio.

The STEP trials supporting semaglutide submission for Obesity once weekly emerging kind of Q2-ish, midyear-ish next year as is also the AM833 data where we will also generate multiple dosing data together with semaglutide in the Phase Ib trial that reads out ahead of the Phase II trial.

Other than that, we are, of course, next year, hoping for approvals on somapacitan as the first once-weekly AGHD product.

And then we can always talk about, of course, the NASH situation because also Q2 next year, a lot of excitement, and hopefully, around our own semaglutide NASH data where both the biopsy study, the MRI study and a fibrosis, a 4-study readout almost simultaneously as does also the Gilead combination trials with a non-steroidal FXR agonist and their ACC inhibitor both in combination, loose combination with the semaglutide.

So very key inflection points on the NASH portfolio, on the Obesity portfolio and then the regulatory milestones that I've alluded to.

I think I'll let it be there and hand over to you, Karsten.

Yes.

Thanks, Mads.

And briefly on financials.

So again, 5% top line growth.

In terms of our gross margin, then in local currency terms, we see our gross margin eroding by some 110 basis points in same currencies, the main driver being U.S. pricing and then the impairment charge that we had, which is then being offset partially by product mix and productivity and manufacturing.

Certain distribution cost up 5%.

Here, we are clearly investing significantly in driving growth and collaborations.

We had some 400 reps in China during the third quarter in anticipation of the Toshiba reimbursement, come 1st of January 2020.

R&D, 6% down.

We have a lot of swings in cash flow of one-offs between impairments and severance last year and the inventory revaluation in Q1.

So net all that out, we're just slightly declining on R&D.

The main factor there is that we're basically -- we had a high trial load with the PIONEER trials last year, which then tapered off.

And now we're in the process of ramping up clinical trials and are closing in on setting a new kind of company record in terms of patients on Novo Nordisk trials come soon, all the trials.

So then again, 5% operating profit growth and the 6% positive currency impact, which is then offset by net financials, 3.1, in hedging costs, predominantly driven by the U.S. dollar and CNY.

Our hedging is, I would say, more tight than it's been for a long time between, what, the case we get above the line and our hedging costs, meaning that our emerging market currencies are -- actually, when we look at growth in IO, then reported growth is higher than our local currency growth.

So no loss on emerging market currencies there.

Then one note on income taxes.

You see a lower income tax rate, and we're updating our guidance, as you see on the next slide.

So there has been a tax reform in Switzerland, and we have adopted that.

It has made a change to our accounting for taxes, our deferred tax assets.

It is a noncash adjustment, so that's important.

So our cash taxes are not impacted this year or the years to come, but it has an impact on our tax rate this year.

Net-net, diluted earnings per share increase of 2%.

That takes us to our outlook, the top line outlook in my intro.

So we are raising the floor, so between 5% and 6%, so in line with the top line generated in the first 9 months.

Currencies are unchanged.

They're actually a notch better than when we guided in Q2 back in August, which I'll come back to.

Operating profit, unchanged despite the fact that we have DKK 1 billion of impairments, which, in magnitude, corresponds to 2% operating profit on a full year scale.

Financial items is increased.

Our losses increased from 3.5 to 3.9, and that's basically the slightly stronger FX that is surrounding further up because we have more decimals here in financial items.

Effective tax rate, I just covered, is

(technical difficulty)

assets.

Impairments, the DKK 1 billion I covered.

And then free cash flow increased by DKK 1 billion, which links to the impairment.

So had we not had the impairments, we would have upped our operating profit guidance, and you see translated in our free cash flow generating that DKK 1 billion stronger.

I now can go through all of the key takeaways on this slide, but I'll leave this slide up there, and then I'll kind of turn around and move to our Q&A.

So Pete?

Pete Verdult, Citi.

Just a few clarifications from today's presentation on Friday, 2 for Mads, 2 for Karsten.

Just look, CMD coming up just in terms of the expectations, 5-year targets updated.

Nothing on 2020.

And only a comment on access, is there something meaningful to say?

Is that the way we should be thinking about the CMD?

The second one for you, Karsten, this 1% hit from the data thresholds.

It's in our numbers, it's being talked about.

But am I right in saying you have to go into the low, is it an absolute 100% certainty that, that plays out?

Or is there some question mark there?

And I'll pause, then I'll kind of ask my question to Mads.

Okay.

Yes.

So [you asked] a number of questions.

So to start with the latter one.

So donut hole, I'd say in -- it's kind of frustrating with all the discussions around patient and population in the U.S. And then the translation is so that patients are actually being put under more pressure come 2020 because they have to stay longer in the donut's hole.

But that's a broader reality.

I think the likelihood of that law being changed or amended at this point in time is low, given the fact that also all the preliminary discussions around health care costs in the U.S., I think it's hard to see legislation passed with the -- which will give pharma an easier time in 2020.

As to long-term financial targets, just to reiterate, so our approach is that our current target on a 3- to 5-year horizon, we're currently in year 4. And historically, what we've done is that we have updated our targets.

When we reach them, or in the one case back in '16, when we had, in fact, we couldn't reach them.

So that's our approach.

I hope to see you all at the CMD.

2020 guidance, you should expect to get in February as you got last year.

And just quickly for Mads.

The Ozempic reformulation we're hearing elsewhere, that you actually improved by -- it's about 20%, which is quite impressive.

Anything you're willing to acknowledge there would be interesting.

And then just on those amylin readouts, just to clarify, monotherapy and combination with sema, what are we actually going to see next year and the timelines?

Okay.

First on the sema, oral sema, we have not disclosed what levels of improvements we are seeing, Peter, in terms of the specifics surrounding bioavailability of the new formulations.

If you go to ClinicalTrials.gov, you would see there are 2 formulations present, one called C and one called D. And they are patented or patentable innovations around the overall concept of making these semaglutide/snack-based tablets in the way that optimizes bioavailability.

But we've not disclosed the improvements.

What I can say is that there are significant improvements across all semaglutide value chain from the very early parts of it, the 8-cubic meter fermentation tanks and the process yields we are getting out of those facilities, initially in Denmark later on to be in the Clayton facility all the way through the purification down to the [tailing] process.

It is an array of activities that are taking down the cost of goods sold.

And since you have to realize that with the semaglutide, we are broadening out the label with all the various comorbidities, including cardiovascular complications, renal complications, fatty liver, over time, et cetera.

It will be highly advantageous for us to optimize the asset we have where I do believe that the patents we are also seeking on the newer and the optimized formulations will take us all the way towards the end of the [30s] and, in essence, be a formidable barrier for other entrants into the area.

So that's the short version on why 2023 did not meet the hurdle this time around.

In terms of all the other things, amylin, in particular, well, the multiple-dosing study is like 12 weeks of cotreatment with a fixed dose of semaglutide and increasing doses of amylin and log both on a once-weekly basis or at least intended for once-weekly in Phase III to seek out the optimal ratio to be able to coformulate or to coadminister these with -- in Phase III.

And those data are actually coming in ahead of the 26-week of B2 data in the AM833 Phase II stand-alone monotherapy study that is destined for Q2 next year.

And that means that our formulation experts will have the abilities to start making the Phase III formulations ready for Phase III even before the readout on the monotherapy, ideally making it possible for us to have a simultaneous Phase III program where the AM833 moves from Phase II to Phase III, and the combo leapfrogs over Phase II and goes from Phase Ib and directly into Phase III, aided by the fact that semi is already on the market and amylin has conducted Phase II.

All of that is going to happen before the end of second quarter next year.

Good.

We go here first, and then move over.

Simon Baker from Redburn.

Firstly, on Slide 9, you have the chart of launch performance of Ozempic by country.

If I cast my slightly shaky mind back to a similar chart you gave for Tresiba, that looks a lot better performance.

And is that a function of the profile of the product?

Or is this lessons learned from the Tresiba launch?

And how should we think about this impacting on Rybelsus?

And one for you, Mads, on Obesity.

You said that ultimately, what you need to do is approach the level of surgical solutions to obesity.

What discussions at this stage have you had with payers and KOLs to just really stress test that hypothesis?

Is that really going to be sufficient?

Because I can see some of the preliminary data that you've shown us.

You can get there.

But the volume of surgery is a lot lower than potentially a pharmacological alternative would be.

So how certain are you that 20%, 25% will be sufficient to crack that market?

So Camilla, if you start with the uptake curves.

Yes.

So the one we're looking at here is Europe, and yes, we have learned a lot of thoughts from our Tresiba launch as we do from all of our launches.

But what has been significantly different for the Ozempic launch has been basically, of course, the product profile and the innovation that we've seen with Ozempic; and then secondly, also, our approach to market access where you could say, for Tresiba, we went in with a significant premium originally in Europe and many years ago.

With Ozempic, we are actually offering, you can say, head-to-head, a better product at a relatively similar price level.

And that, of course, makes the market access, to a large extent, faster.

But it also means, from an execution point of view, when we have to advocate the profile of the product, that then, of course, there are a number of advantages both on the benefit side but also on the cost side relative to the benefits.

So what that means for Rybelsus is that we now have it approved in the U.S., and you should expect that we are going to do sort of an access-related uptake, just like what we did for Ozempic also.

And that means that when we have access, we will go all in on Rybelsus in terms of detailing the product.

But right now, we are in the middle of the access negotiations, and as you know, we cannot share any details with you on that.

But that's the approach that we will take for Rybelsus also.

And then over to Obesity.

I'll just quote (inaudible), one of the leading U.S. authorities on obesity.

His team, now that you mentioned bariatric surgery, because it is a good point, they have investigated what amount of the respondents who come to the bariatric surgical clinic then asked, "Are you willing to undertake the most dramatic one where you disconnect the stomach from the rest of the GI tract?" That's the one that works really well, but it's associated with a lot of malabsorption and other risks to the body in doing that.

But that being said, it's only between 10% and 20% who are asked the question, "Would you take this bariatric surgery?" who actually end up getting it because the rest shy away from it.

Our dream is, of course, to have a similar level of efficacy in a once-weekly ultrafine needle injection so that people -- it'll be like Ozempic but offer a surgical-like level of weight loss.

What is it that the patients need?

Well, we have spoken to Camilla's team and my team many times, advisory boards, including key opinion leaders, whether they be bariatricians, endocrinologists or high-prescribing GPs, and they're all saying that for the obesity market to truly open up, this 5% to 10% weight loss that we are seeing with today's therapies, including our own wonderful Saxenda, it's disappointing to most, over time.

That's why the data is so short because when they, in an unaided manner, attend the clinic and asked the question, "What do you expect?" they typically expect 15%, 20%.

So they are basically not informed well enough about what to expect with today's therapies.

If and when semaglutide comes of age, potentially even spiced up with AM833 over time, then we are getting whether patients will not be disappointed.

The therapies will be easier to administer on a once-weekly basis.

They will keep on a constant weight loss over longer periods of time, and Camilla will have developed further patient support programs to further enhance the adherence to the therapy.

So to me, and I've said this many times, I'll repeat it today, to me, the OpG pharmacotherapy area is only in its infancy.

And pending success for some of the data next year, I think we can make this a fantastic thing for patients and, I guess, for ourselves in the many years to come.

Thanks.

Wimal?

Wimal Kapadia from Bernstein.

Just firstly on Obesity again.

Mads, is there any way of actually identifying the best responders?

So if we look at the Saxenda clinical data, you see some patients, you can lose more than 10% of their body weight.

So is there any biomarker strategy that you could potentially develop that really identifies patients that will really lose a huge amount of weight with your existing products?

Second question on -- for you, Mads, again is on the anti-TFPI.

So Bayer announced the stopping of their products last week.

Any comments there on -- I mean I should ask the company themselves, but how is your product different?

And just your thoughts around that.

And then one for Camilla, just on Obesity again.

Could you just give us an update on how your discussions with regulators are going outside the U.S. for potential reimbursement?

It's still very early, but are you having those discussions or the feedback you are getting, et cetera?

First, Mads.

Yes.

So if I start on the Obesity question.

It is a good point, Wimal, because...

(inaudible)

No.

Yes, I'll take the first one, the Obesity one about the biomark, and the Bayer thing is, yes, also very interesting.

I'll get back to that.

But on the Obesity thing, we are collecting, with full consent, all relevant biomarkers of what makes a super responder a super responder and a nonresponder a nonresponder.

To date, I have to disappoint you that we have not pinpointed on the GLP-1 side of things exactly what that is.

Maybe there is more for continuum.

It's not like a bimodal distribution, that much I can tell you, because we have data from literally many, many patients nowadays.

But we are working on segmenting the population.

So it's not going to be like necessarily personalized medicine, but it is one way of -- we are starting to pick up clusters of responders, not by genetic analysis, but by a number of, you can say, agnostic information that we have in the patient journals that are starting to dissect out who is expected to respond better.

These are things we'll publish going forward once we're ready for it, but it's not a single biomarker-based solution.

For amylin, we know from the heyday, if there ever was one, of the Amylin Corporation that also, both pramlintide and devlin side, the not-so-good and the slightly better amylin agonist they had, they also had responders and nonresponders, and that makes me believe that we need to work also on AM833 to dissect that out.

So we have whole group built in research but also in translational medicine that works only on these topics.

With regard to the bias TFPI antibody, theirs was targeting both the clinical domain 1 and the clinical domain 2, i.e., the domains that are responsible for factor VII and factor X binding, respectively.

And that's, unlike ours, which is a K2 domain only, a clinical domain 2. Now it's getting a little bit technical.

But that also means that they are different pharmacologically.

The other thing is that TFPI inhibitory antibodies, they initially get cleared by target mediated clearance.

And that means that you have a constant rate of removal, whether the dose is high or low, but once you saturated the TFPI binding sides with the antibody, then you move into a saturation -- not a saturation, but actually a first-order kinetics, the classic monoclonal antibody kinetics with a half-life of 11 to 14 days.

What that means is that you get rapid, rapid, rapid accumulation of the product in the body.

And the reason why Novo Nordisk has decided it's easier to go for once-daily already in Phase II, is that we want to control the level of the concizumab antibody in the circulation to a much higher extent than we can achieve if we give a huge dose once weekly because there, big parts of the time, the exposure rates will be chemically in the red -- sorry, at least in the yellow to orange zone where you can start seeing procoagulant activities maybe going a little bit array.

So we're taking more precautious approach.

And I do have to say the feedback from the patients using a oral injector FlexTouch Pen with ultrafine needle, it really is not an issue taking it once daily.

But what it does is it gives a much more smooth patient exposure and hits hopefully not the risks that apparently we're seeing in the 3 cases of the bio patients.

Because of the variability of the CSPI castration completions, is that another factor?

It is -- Wimal, that is a really good point.

You've started a tough -- that is bioavailability among patients.

Not within patients.

Within a patient, you tend to be same.

But between patients, that intra-patient variability in KPI levels does fluctuate quite a bit.

That's a further aggravating factor, that if you are one of those where you easily saturate, boom, things happen.

So that is an element in why we also decided for the -- by the way, we're trying to investigate, together with some leading academic clinicians, what is the underlying basis for this huge heterogeneity in TFPI levels because no one truly understands it.

But it could be a factor in responsiveness and/or safety concerns.

I am optimistic, by the way, about the prospects of concizumab because of the things that we've done.

Okay.

Camilla?

And back to obesity and your question about our status on working with regulators about reimbursement.

I can say that there's just a few countries in the world where we actually do have reimbursement at this point in time, and as you know, the commercial segment in the U.S. is one of them where we are close to 80% coverage in that commercial segment.

However, outside the U.S., this is going slower.

And you will also see that the markets where -- that are used to having no reimbursement for pharmaceuticals such as in the Middle East and also Korea, Brazil and so on, are relatively doing much better on our obesity franchise than traditional markets like Europe where we have a tradition of having products reimbursed.

So what are we doing about that?

We have established what we call the OPEN network, which is a network between payers and policymakers for, right now, 8 countries of the world that meet and discuss how do we deal with obesity as a chronic disease because, of course, one can imagine that from a payer point of view, it can be quite scary to recognize obesity as a chronic disease and then knowing that in some countries, 1/3 of the population then suddenly has a chronic disease and what do we do if we have to treat all of those?

But of course, as within many other disease areas, there is an opportunity to treat some, you can say, a subsegment of this population first and thereby avoid some of the incurred costs that obesity actually has on society.

And we know that if we start with some of the more severe segments, then, of course, there will be savings in the health care system to then work with the others.

So those are some of those discussions that we are having.

We also know that obesity is a significant risk factor for many other serious chronic diseases.

And just by bringing down the weight with 7% to 10%, we know that we can significantly reduce those risk factors.

So those are some of the types of discussions we are having with payers and policymakers.

At the same time, we are also working with more risk-based contracts in some countries, meaning that if a certain population is not getting the risk, the weight lowering, that we've seen in clinical trials, then such a regulator would not pay the full price for Saxenda.

So it's possible to share that risk.

So those are some of the initiatives we are taking.

But it does take -- it is a slower process and it does take time.

But nevertheless, of course, we also do see that patients themselves are willing to pay generally more for obesity products than other types of chronic treatment drugs.

And Wimal, it's also like with the SELECT study, nobody -- when the UKPDS came out, there were like, sulphonylureas, metformin and human insulin.

Now there's like 60 products for type 2 diabetes, and the pricing went up radically when people showed that you could extend the lives of people without invalidating complications just by intensive therapy.

As you know, the SELECT trial is targeting that exactly same situation to try to show that when obese people with high BMIs live 7, 8 years shorter just like diabetics, then we can do something about it with the pharmacotherapy.

And that should also help in Camilla's efforts to position obesity as what is, a serious, progressive disease that warrants medical attention.

But those data, of course, will not help us until they come out.

Okay.

And let's move to Keyur, and then we'll go to (inaudible).

Keyur Parekh, Goldman Sachs.

3 questions please.

Camilla, the first one for you.

Can you help us understand what the discontinuation rate on Ozempic is?

So far, it's very early days.

One of the points Lilly have made, historically is patients data in 1 (inaudible) is meaningfully higher than that on Victoza.

So whatever way you can share with us on what you are seeing with discontinuation rates for Ozempic, that would be helpful.

Secondly, as you've been speaking to payers on Rybelsus, what is the biggest pushback you've had?

And should we think of whatever access you -- are they thinking of this as another GLP-1?

Or are they thinking of this is an oral antidiabetic agent?

And then Karsten or Mads, you historically said that at a price point similar to injectable GLP-1, we should think about the steady-state gross margin on Rybelsus being equal to the corporate gross margin.

Obviously, Henrik's been working on improving kind of at facilities, et cetera, over the last 24 months since you gave that guidance.

So can you update us on, as we stand today, does that corporate gross margin level still hold?

Has it gotten meaningfully better from that?

Yes.

Camilla, if you can start?

Yes.

So yes, on discontinuation rates and stay time on Ozempic, it's a bit of early days to exactly give you good reasons, as it hasn't been on the market for that long, the exact data on that.

However, we do expect, based on what we've seen from real-world evidence data and in general that we will see same stay times as we see for the other GLP-1s.

We have also looked at who is being prescribed Ozempic versus Trulicity.

And it looks like from an HbA point of view, HbA1c point of view and weight profile point of view and so on, it's a relatively similar population as well.

But it's a little bit early days to give you stay time insights on Ozempic at this point in time.

And Keyur, just one medical comment.

As you recall from the UKPDS, the HbA1c changes year-over-year in the same patient is typically 0.3% to 0.4% increases year-over-year.

And as you know, from the SUSTAIN 7 trial, there was approved doses 0.4% HbA1c benefit of Ozempic versus Trulicity.

So what they might win a little bit on the convenience of the device, we will win on the longer clinical stay time driven by greater efficacy.

And then on Rybelsus and payers, it's, of course, in progress.

So I cannot give you many details on it.

But from the initial contact that we've had with payers, it's clear that the benefits of Rybelsus is well understood and also that it is a quite novel treatment because it is a GLP-1, but it is in a tablet.

So from that point of view, it is regarded as a GLP-1.

However, of course, with a potential to address the patient's -- the 70% of patients that are currently being treated with OADs only.

And of course, Rybelsus is not going to address all of the 70 at the same time, but it's interesting to see our learnings from Ozempic, is that it is also being prescribed primarily based on OAD failures and not sort of as replacement of other GLP-1s.

So of course, the potential of accelerating this with Rybelsus is even greater.

Yes.

Good.

And then on Rybelsus gross margin, so what you are hearing and what Mads was commenting on before, that it's not only the formulation but it's the entire value chain that we are improving.

And now you've followed us for a long time.

This is exactly what we've been telling you for many, many years in terms of what we've been doing with our insulin for decades, right?

You remember the new insulin-producing strain with high yields and how we optimized our unit cost in assembly and filling it, et cetera?

So it's -- this is a Novo Nordisk classic for Rybelsus.

So on that front, the assumption still stands that we'll be getting to around a proved average gross margin as we get the volume strength up.

Sorry, Jo?

Jo Walton from Crédit Suisse.

So it's a follow-up really from Keyur's.

Firstly, can I just ask you what's the importance of the CV benefit from the point of view of a payer?

So you're busy having your negotiations on Rybelsus now, but there's this one aspect that you don't know whether you'll have on the label or not.

Do payers care?

Do some of the payers say, "I'll think about it, and I'll give you price X if you get it price Y, if you give don't get it," so effectively pushing back a decision a bit later, waiting for this.

Could you also give us some idea on the infrastructure that you now have in place in the U.S.?

I know that you said you were getting more people with oral-type experience.

So I'm just intrigued to see how much of that you have in place ahead of 2020.

And to go back to Keyur's sort of gross margin question.

That's all very well in 3, 4 years' time where you've got a sizable amount of the stuff coming out.

But the 1st of January next year, you've turned your factory on.

And little dribs and drabs are coming out, and you got all these people busy working away.

Presumably, we should expect to see some material noticeable level of gross margin impact next year as you turn on that.

And one final question for Mads, just on obesity.

How -- what's the rebound when you stop taking these?

And is that a big issue?

And could you perhaps get some people down to a weight where they can take then bariatric surgery because they're just, frankly, too ill and be sort of like a bridge to surgery?

Is that an opportunity?

Okay.

So Camilla, if you start with the...

Yes.

On the CV impact on price and maybe in a minute, Mads can talk a little bit to our expectations on the actual label, but on the impact on price, every characteristic that a product has, of course, makes -- has relevance.

And especially because the treatment guidelines in the U.S. also favors the starting point a patient at risk of cardiovascular issues or not.

And -- however, it's not so that there are sort of markups or true-ups in case we get it or not get it.

That's not how it works.

And also, when we had earlier on in -- in Europe, we had the CV label for Victoza, it's not so that then suddenly prices increased just because of that.

But of course, it gives a broader perspective when we have to position the product and explain the benefits of it, then it all adds to that.

But also in the market later, when we promote the product, then of course having a CV label is important.

We do know, of course, from Ozempic that the HbA1c profile and the weight profile and then the CV profile is sort of the order of which the product is prescribed is what matters the most.

Mads, would you want to say a little bit about the expectations to the label?

Before we move on to...

Yes.

Just only you can say -- yes, only you can say, Jo, that basically, we are in a -- we're also in constructive dialogue with the agency, and I'll repeat that here.

The actual date, as you know, is January 20.

They are running for the Ozempic and Rybelsus.

And the longer time that we do not hear about CV outcomes, the less likely one would argue theoretically that having such an outcome would become over time.

That being said, we're totally prepared for it.

I would say the base case is that for (inaudible), this is a repetition of what I said already, that Ozempic should be in a good position to get a CV indication also based on the SUSTAIN 6 being a bigger trial, even though not a big one, it's still bigger than PIONEER 6, and it's supported by PIONEER 6. Whether the reverse is true, i.e., that PIONEER 6 that was too small to show everything that you wanted in that trial, but it is seconded clearly by SUSTAIN 6, whether the FDA accepts that for a full CV indication on the current premises remains to be seen.

But we must have some belief in it, otherwise we would not have submitted.

So January 20 will be very exciting.

Yes.

And then on the gearing up on the capabilities for the oral space.

It's correct that we have looked at our infrastructure, of course, in the U.S. But it is so that already today, we actually cover 80% of the potential that relates to, also, the oral space in terms of the doctors that we are already calling on.

So now what we, of course, need to be -- to look at is how we optimize the call frequency because we are already actually seeing those people.

And -- but of course, in our -- also in our back-office functions, we have, of course, manned up with people who have experience from this field as part of preparing for the launch, both in our headquarters but also in our U.S. affiliates.

So we feel ready to execute on this.

Yes.

And then finally, on the gross margin.

So it's important to note that the products we sell next year, we sell a little this year, but predominantly next year.

First, we sell that they've been produced (inaudible) at Kalundborg and (inaudible).

So what we charge to the P&L next year in terms of the tablets has been produced in the Danish infrastructure.

So the U.S. API facility in Clayton, North Carolina, and now there's a factory in place and we have all the equipment built in.

So now it's -- we're moving closer and closer to qualification of the factory.

Of course, there will be an impact on our cost of goods as that factory is going into production.

So it would be a staggered impact over the coming years, but net-net, there will be a negative impact on our cost of goods.

That's kind of the classic when you do step up on a big kind of supply chain scalability project.

So there will be some negative impact, but it will be phased over time and of course, go into the total equation of productivity improvements, product mix and U.S. pricing pressure.

And then there was something on bariatric surgery.

Well, just in terms of whether you might get obesity carry the bariatric surgery and how sustainable it is.

You get people down to weight, they stop taking it.

How quickly do they (inaudible)?

Yes.

So actually, we follow that in the scale.

Obesity, a prediabetes study, where there was treatment ongoing for 3 years followed by a washout period.

And there was sustainability of the prevention of onset of type 2 diabetes to the extent that 12 weeks out of -- after discontinuation, there was still like an 80% reduction in the onset of diabetes compared to the non-Saxenda group.

But in terms of the weight rebound phenomenon, this is unfortunately extremely well known that as you lose weight while you are on pharmacotherapy or diet and exercise, your basal metabolic rate is tuned down a little bit to compensate to try to regain the weight for -- this is an evolutionary thing, as you know, and that is actually the case.

So it will mean that if you stop, for instance, on semaglutide, you've lost 15% body weight, then the likelihood is that over the next year, you are likely to regain a substantial part of the 15%, which is the reason why FDA asked us to do chronic studies.

We were actually discussing with the agency, at some point could we do cyclical therapy?

Could we look at body weight maintenance?

Actually, they said, "This is chronic disease.

It's chronic therapy just like using statins, antihypertensives and antidiabetics." We have to start thinking about obesity as a little bit the same as all the other chronic conditions.

And then the bariatric, yes, there are investigator-initiated studies ongoing to investigate GLP-1 agonists, those from (inaudible) as a precursor stage to bariatric surgery.

Okay.

Yes.

Way in the back.

Mads, could you give us your updated thoughts on Lilly's GIP/GLP (inaudible) just what advantage do you think that your Ozempic would have, given sort of the [AIC] reduction in weight loss that we've seen?

Yes.

So very briefly, we have recruited the patient for the SUSTAIN FORTE trial, which is the 2-milligram semaglutide study.

Our medical modeling tells us that, that will be able to match a GIP/GLP agonist, at least the one tirzepatide we know of from our friends at Eli Lilly.

But it does so -- again, we need to prove that with a nice and tolerable profile because we have treated people with 2.4 and 2.8 in the obesity setting.

And even though they have been a bit more on the heavy side than those with diabetes, then the exposure levels are reminiscent of what had been seen in those trials and that has been highly tolerable at dropouts being modest.

With tirzepatide, to what extent the GIP and the GLP components contribute to efficacy and tolerability remains unknown.

The ratio in the Lilly compound is about 1:1 on GIP and GLP receptor binding.

But it seems as if there is a, in my view, disproportionate amount of diarrhea and sometimes also discontinuations despite the slower titration that Lilly has used in the recently communicated trials brings us at the ASD meeting in Barcelona, but also at the ADA in San Francisco.

So even though we should know ourselves because we've done a lot of GIP work, our human database, which is only in its infancy or emerging over the time to come, does not really tell us to what extent diarrhea is aggravated by the GIP component as compared to GLP-1.

But there seems to be some extra contributions on the tolerability side from GIP, and that has to be borne in mind also when you think about the ratios between the 2 components.

And you need to find the sweet spot where tolerability is not compromised while efficacy is enhanced.

And if that exists, we'll be seeing in the time to come, both from us and from Lilly.

Where are you with your (inaudible)?

Well, so the triple agonist is a GIP/GLP glucagon coagonist that is in multiple dosing Phase Ib.

The coagonist that was -- sorry, that's not a GIP.

That's a glucagon.

And then we are actually having ones with the GIP molecules that we can titrate up against semaglutide, but that is not yet in the clinic.

Okay.

Thanks, Mads.

Stephen?

3 questions, please.

First on R&D spend.

You referred to it as a valley on the call as you're through transitioning.

So how quickly do we think it gets back to the 12%, 13% range?

Is that next year or a bit further out?

Second question just again on (inaudible) payers.

(inaudible) step edits coming up in the debate, and if you could remind me what your stance on accepting step edits would be?

Obviously, there's a variety that could be in that, which is your stance on step edits.

And for Mads, Lilly (inaudible) getting a bit more vocal on their Phase I oral GLP-1.

It's just -- I know you discussed it much before, but just your perspective and the data you're seeing on their profile, respectively, relative to [all else].

Yes.

So on R&D ratio, so we are basically on our way out of the valley.

So already in Q4, you should expect to see a high R&D burn taking the full year to around 12% R&D ratio for this year.

We don't guide for this year, but you shouldn't expect it to be lower than 12% next year.

So yes.

Yes.

Yes.

On Rybelsus, as you rightly said, we cannot comment on a lot of details on it.

What I can say, though, is that, of course, Ozempic is reimbursed at a similar list price, a level without any step edits and available to use after metformin.

So here, you have an oral GLP-1 in a tablet at a similar list price.

And so, of course, ideally, you would have the same conditions as that.

And then on the Lilly.

So Lilly, as you know, said that they're pursuing the (inaudible) compound, which is a medicinal chemistry base.

It's a small molecule receptor agonist, (inaudible) agonist.

That's what our approach is.

Historically, we know that it's difficult to make full agonist on GLP coupled receptors for peptide simply because you cannot get orthosteric modulation because you have too many contact sites to trigger -- to mimic the biologic.

But you get more typically allosteric modulation, which basically increases the sensitivity of the receptor to endogenous stimulation.

So you get, typically, partial agonism.

But you may, in this case, get closer to full agonism based on what I've seen, whether that translates into access into the compartments of the brain that signals satiety to the extent that semaglutide would surprise me from the perspective that no other GLP-1 receptor agonist that we have tested has been able to do so.

So I think the question is how efficacious will it be, both in A1c but also on weight loss.

And to safeguard ourselves against any risk that the (inaudible) compound came out close to Rybelsus, we could also, of course, over time, consider how we develop the oral GLP-1 franchise in terms of dosing, et cetera.

The other approach they have is combining a GIP and a GLP in a peptide that is sitting for Phase I by the [oral group], as I understand it.

Here, we are into a little bit the same boat that we need either to guesstimate what ratio we need and/or to really work out the ratios that are desirable for such a product.

And I think we will take an approach where we have the benefit that sema is on the market.

So we can take like once [we get] GIP, and then we can basically titrate up just like we do with AM833 today.

And of course we also have to assess, will the GIP component, and I forgot to mention that to the gentleman just a minute ago, with the GIP component, if you take the overall pros and cons for adding GIP to sema or any other GLP-1 or a coagonist or triple agonist, will the benefits outweigh the drawbacks in terms of tolerability and compare that up against what is the similar situation, for instance, for amylin adding back to GLP-1?

So I think we are playing on a number of peptides, and I think the jury is out.

My personal favorite at this point, I think, has not escaped your memory.

That is probably combining semaglutide with a once-weekly amylin because that truly seems to be additive while yet being tolerable.

But again, the jury is out because we only have animal data and 7-week data as monotherapy.

Now we are looking for the full monty.

So it seems like we're almost exhausting all our questions, yes?

Peter Welford, Jefferies.

I have 2 questions.

Firstly, just on the Rybelsus.

Wonder if you could just outline with that what you're doing differently, I guess, with your marketing campaign relative to Ozempic, particularly, obviously, with regards to the fact that this isn't something fast that you can just pick up the prescription and start giving the titration regimen, given as well the GI issues, where you have to take this daily rather than by having it in your blood after a week.

And then secondly, just coming back to the obesity point there.

Are you also looking at -- and I know -- I appreciate we're in the early days of this from a regulation point of view.

But sort of software stroke IT to the point of view of sticking patients with stickers, I mean it could be whether apps or whatever it may be.

What's Novo doing in this area?

Are you potentially looking at potentially trying to get into digital medicine?

What is the combination with therapy?

Yes.

Good question, Camilla.

Yes.

So on Rybelsus versus Ozempic on the markets impact, what is important is, of course, that with Rybelsus, we are likely to compete much more in all spaces.

The whole point of this also means that our co-pay cards actually have -- means that we are -- the patients, at a maximum, will pay $10 per prescription, whereas in the injectable segment before, with Ozempic, it would be up to $25.

So from that point of view we have adjusted to sort of what is the usual standards in the OAD space.

When it comes to Rybelsus being taken half an hour before the first meal of the day, we know, actually, from clinical trials that from a preference point of view, this compares to (inaudible).

There was no difference in this, so it didn't make a difference to patients on their preference.

But what we are anyway doing is developing a patient, you can say, support tool that can make them design their own day in a way where we support them not only about when to take Rybelsus but also how they organize their life with diabetes.

And actually, we know that when we ask people about diabetes, what is -- what they really say to us is, can you help us maybe forget about this chronic disease so I don't have to worry about it?

And so those are some of the types of programs we are trying to support patients with so they really don't have to bother too much about it.

And perhaps, Camilla, you should cover ex U.S. in terms of marketing tactics for Rybelsus versus Ozempic because I think that's perhaps where we have a bit slightly bigger difference.

Yes.

So ex U.S., of course, we still need to see the product approved, expected next year.

And then here we are, of course, looking at how we make sure that we also will be relevant in the all space, whereas we, in the U.S., as we just talked about before, seeing most of the customers already.

That is not the case in the rest of the world.

So here, of course, we're looking at how we prepare ourselves better for that.

And that -- here we're taking a market, you can say, fit approach because different circumstances in each market to make sure that we cover all of the relevant positions because, traditionally, for example, in Europe, we have covered the injectable segment very well.

So insulin and injectable GLP-1, but actually not the primary care segment to such a large extent.

So that's where we would, of course, change our approach.

Good.

And then on digital health.

In the interface between the 2 of you.

Yes.

So tell everyone.

Well, just briefly, Camilla's and my teams are cognizant of the fact that if there's anything you need to reinforce in the mind of a person with obesity, then it's successes.

And if there's anything you want to deemphasize, then it's failures and sense of guilt and this really unfortunate situation that people can really not -- I mean it's genetics, right?

So that is being reinforced by both, some of the work that Camilla is ongoing with the Noom company and others, where coaches even fittingly involved in actually helping the patients kind of with behavior modification, have a more sustained both data on the product and greater, greater outcomes.

But in general, this whole use of software as a medical device is something that we, as a company, are starting.

We also got regulatory experts on board to help us make sure this is a different group of regulatory approval and so on.

So my team and Camilla's teams are working extremely closely together.

Yes.

And it is important to see a chronic disease.

It is more than just a product that people are benefiting from, but there are a lot of other areas around their whole everyday life that they are even more concerned about because normally, the product works very well.

But they are more concerned about other types of their daily life, and we can support them with that, with data tools, behavior modification tools.

And it goes in obesity, but it also goes in diabetes where we, for example, with connected pens, can find ways to give patient and doctor more transparency about when is it -- why is it that if they do the same things, they feel they do the same things on an everyday basis, why is it that their HbA1c fluctuates still?

And we can help create more transparency with digital tools.

So it goes all the way from supporting behavior modification to actually getting more information about what is it in your particular life that gives rise to daily changes in HbA1c.

So luckily, we stand to gain from that also in the future.

Great.

And then [Dan]?

[Daniel] from RBC Capital Asset Management.

Without front running your capital markets, I was thinking over the long term, now lots of moving parts on where you are.

What does success look like for maybe R&D and commercially?

I mean is it going to be a lot more digital?

I guess in that sense, it's not going to be sort of business as usual of what we've seen in our last 5, 10 years.

So I'm just thinking about what are some early thoughts about that?

What is exactly looking like for you maybe 5 years but just thinking about just longer-term horizon?

You talked about this until a little bit here about cities, R&D changing, the commercial models are changing.

What is success?

Yes.

Perhaps, Camilla, you should start with the [first] question, and we'll work from there.

Yes, I can.

So if we take a really longer-term perspective, success for us would really be to drive a sustainable business on a couple of dimensions.

So one is, of course, on the business, business path, where we talk a lot about some of the drivers today.

But we also should expect that diabetes and obesity are becoming so big issues for societies that it's not enough for Novo Nordisk just to make sure that we can address sort of the symptoms of those chronic diseases in terms of HbA1c or weight loss.

We also have to find ways to make sure that those patients who are either not covered or somehow, as we say in Danish, fall between tiers, that we find ways to make sure that they can get access to lifesaving medicines.

So that's another point we need to address to drive a sustainable business so that access and affordability, and that looks like different country by country, because of the different circumstances.

But then finally, you should also expect that there is an element of prevention in a more sustainable business going forward that relates to how can we help societies manage some of these big problems that diabetes and obesity will be on -- not just on health care budgets but also on related comorbidities in the future.

And are there ways that we, with our insights, can try to work on actually preventing this problem becoming as big as it could become?

So where our primary offerings is, of course, the innovations.

And it will continue to be, and we just talked about a lot of that today.

You should expect to see that a sustainable business also requires that we can do this in a way where we can address those people that are most vulnerable but also address prevention.

And then finally, do the whole thing in an environmentally sustainable way as well.

And Mads, long-term vision for R&D?

Very briefly because that can take hours.

In the existing businesses where we are faced with injection-based medicine and have leadership position, of course, we want to solidify and expand that.

In the new areas, whether it be cardiovascular, NASH, potentially kidney disease, it's important for us to build a platform there and a business and longer-term strive for leadership.

In the more platform technologies like stem cell-based therapies, regenerative medicines, our plan is to go for like 1 IND in a new area per year.

They will not all succeed, but we would like to see a regenerative medicines platform built over the next decade, and I think we are on track to doing that.

And that is through unmet needs because we're only going through various serious diseases with very limited treatment offerings, as you are aware of, with that platform.

And then, of course, longer term, we have today access to both MIT collaborations on the oral biologics delivery with the Soma device, the (inaudible) device.

We have the driving platform from the (inaudible) company.

We have the SNAC molecule, et cetera.

I believe whenever we can long term replace injections with all biologics, we have a leadership position there.

We need to expand that going forward.

So those are just some thoughts about what success could be like 10 years from now.

Great.

I think this was a very excellent way to end this lunch meeting.

So thank you for attending our Q2 London lunch meeting.

And any further questions -- Q3 lunch meeting.

If any further questions, please reach out to Investor Relations, and you get the numbers right.

And thanks for attending, and have a great rest of today.