Seeking Alpha
We cover over 5K calls/quarter
Profile| Send Message|
( followers)

Novo Nordisk A/S (NYSE:NVO)

Investor and Analyst Conference

September 24, 2013 12:00 pm ET

Executives

Mads Krogsgaard Thomsen - Chief Science Officer, Executive Vice President and Member of the Senior Management Board

Analysts

Michael Novod - Nordea Markets, Research Division

Lars Hevreng - SEB Enskilda, Research Division

Carsten Lønborg Madsen - Carnegie Investment Bank AB, Research Division

Mads Krogsgaard Thomsen

Welcome to this event. It's small and slightly different than we often have done, even though we have this kind of approach here before, as you'll see in a minute. I should make the quick information available that this is being webcast, as usual, and the plan is that I'll walk through essentially some of the more exciting presentation, some of the many exciting presentations that Novo Nordisk has over these days here in Barcelona. Some are already being made today, some are coming over the next few days, and I selected just a few of them, basically, either all presentations or posters that I think are worthy of discussion. That is going to take like 30 minutes or so, and then we have time for discussion, and you can ask questions about any of the other studies we have as -- once we have shown these forward-looking statements. We have quite a lot of presentations, a total of around 50, and they cover, apart from the modern insulins, it's all the new products, IDegLira Victoza, Ryzodeg, Tresiba, and a lot of research. So I'm actually only going to show you very, very selected sneak preview of some of these.

Now the first one, I think we have many people, both in the company but definitely also outside of Novo Nordisk, in the physicians community, that are really, really upbeat about the prospects for the fixed-ratio combination product called IDegLira, which, as you know, can give up to 1.8 mg Victoza and the 50 units of insulin degludec when you dial up the pen. And what you see up here is data that you have probably heard about before -- or I know you've heard about before, from this DUAL I study, which was a huge study in 1,660-around patients with all antidiabetic treatment, they were injection naive, and where we compare IDegLira being the curve out here, to the 2 individual components, insulin degludec and liraglutide 1.8 milligram. And in reality, what we saw was a totally unprecedented glucose lowering potential of this product, i.e., we took patients from a baseline HbA1c value of around 8.3% down to below the biochemical cutoff for type 2 diabetes, that is 6.5% A1c. And essentially, they stayed there, as you can see, throughout the trial and also in the follow-up period for the next half year of extension.

So what I would like to show you here is something very, very interesting. Maybe the notion that you are able to realize this HbA1c improvement, that is relatively unheard of, in the face of the same fasting plasma glucoses, i.e., the fasting blood sugar control, was similar between degludec -- degludec is a highly efficient fasting glucose regulator. It's a -- as you know, even compared to glargine, we have systematically seen numerically and sometimes significantly fasting glucose improvements for degludec compared to both glargine, compared to Levemir, and so on and so forth. So we actually matched that with this one, yet the HbA1c came out clearly superior, as you can see up here. And that can then only be explained by one thing, namely improvements in postprandial glucose because, as you know, HbA1c is a composite of the mealtime glucose excursions and the elevations in the fasting glucose level that we see at night time and between the meals.

So what I would try to show you now is I think something very exciting that we have basically and are announcing at this conference. Here, you can really just see that, at baseline, we have 9-point self-monitored blood glucose profiles. And you can that at week 26, they are all coming down because Victoza and Degludec are effective molecules. But in the combination treatment, you can already sense that there's something special going and that some of these meals, you have the notion that the excursion around the big meals after this once-daily injection is further improved compared to [indiscernible] liraglutide. So what we have done, and this should lead not into risk but interistitial glucose, which is essentially what you monitor when you do your continuous glucose monitoring. So basically, we both measured during meal tests with a standup liquid meal test and measured the glucose excursion as the predefined 90-minute cutoff after the meal and shown significant improvements at all 3 meals, breakfast lunch and dinner, in favor of IDegLira compared to degludec. But importantly, as we can see here, where down here we have the baseline interistitial glucose, so glucose values, and what we have here is the effect after 26 weeks. And for instance, if you compare, well, actually, across the meals and look at the CGM measurements, the one with the grayish curve here being IDegLira, is distinctly improved compared to both degludec and Victoza. So we have a situation where the PPG improvements that in the standard meals has but also in routine measurements used for 72 hours in the CGM setting favors IDegLira. And essentially, these postprandial improvements, they are, you can argue, enough to account for the greater than 10.5% improvement that we saw in HbA1c. So what I will do is elaborate a little bit on about how this came to be reality.

So on the next presentation, which is the result from a standardized meal test looking at IDegLira and the postprandial glycaemic elements control but also what happens in terms of insulin secretion in terms of beta-cell function, you will see some pretty exciting findings. Now here, we essentially are looking at the mealtime insulin profiles at this standard meal, which is around 60% carbohydrates, around 30% fat and around 10% protein. Such is a pretty reminiscent of a standard kind of meal in any setting. And what you see is baseline value is over here. But importantly, if we take a look here at the right panel, where we have normalized for the baseline values or the values at the onset of the meal, we can actually see that in the case of IDegLira and liraglutide, there is a distinct and significant improvement in the insulin secretion profile at the meal compared to insulin degludec. Now this is all the more impressive when you bear in mind that in the case of IDegLira, these patients are already much lower in A1c. So this is done at week 26. That means that these patients are already at A1c of 6.4, and that means that the insulin resistance, the glucose toxicity, the need for insulin is much less because a lower A1c, the burden of insulin secretion at meals is much reduced compared with high levels, which is hardly surprising because the glucose load that you have to cope with and distress up the beta cell would dictate that much less can do the trick. In spite of that, we are actually seeing a distinct improvement. So had we done this in the phase of an A1c of 7%, like in the case of degludec and Tresiba, as shown up here, we would probably have seen a curve all the way up here somewhere.

So this, really, for the first time, very elegantly, I think, demonstrates that once-daily IDegLira not only cadence for fasting glucose to the level of bringing people down into the target range of below 6, actually, but also is able to take people's postprandial glucose excursions into the normal range. It is essentially so that we end up in a situation where the PPG increment at 90 minutes is down below the 7.8 minimal cutoff for normality, as defined by the International Diabetes Federation.

So if we look a little bit more specifically at what is known as the insulin secretion ratio, you can actually see, again, here that there's a significant improvement in IDegLira, that the beta cells simply secrete more insulin in less time, so to speak. And when we then measure that as the so-called static index of beta-cell function, this is also evidenced by a high significant improvement of IDegLira compared to insulin degludec.

So what is really happening is that IDegLira utilizes the beautiful glucose lowering in the fasting state of degludec, supplemented with some fasting glucose lowering associated with the liraglutide component, and the reason being that these guys got less, and ladies, they got less insulin. The dose of insulin in the IDegLira arm was only around 38 units compared to more than 50 units in the degludec arm, yet the fasting glucose was maintained at the same level, showing that liraglutide contributed to the fasting glucose control. But liraglutide was solely responsible for the prandial glucose control that happened at all 3 meals throughout the day. So in reality, in one injection, you are now normalizing the glycaemic picture both at meals, between meals and in the fasting state at nighttime. So I think these data are explaining some of the A1C data that you've known about for the last half year or so.

But there's more to it than just the ability to combine insulin degludec with a molecule such as liraglutide. And the way -- just to explain, the way degludec has been designed is in such a way that you can actually combine it with just about anything. And very briefly, to use 30 seconds on that, insulin degludec, unlike all other insulins, comes together as dihexamers so that when you have a formulation of Tresiba or insulin degludec, in that formulation, you'll have 2 hexamers bonding up so that the open ends, so to speak, the ones that are amenable to kind of interacting and interfering and you can say having a blunting of the pharmacokinetic profile due to mixing with the other molecules that you might put into the formulation, those ends, they are closed by bonding up as a dihexamer in the formulation, really allowing Tresiba to exist in the formation without interfering in any other molecule that you want to put into that formulation. So in IDegLira, we put liraglutide into formulation. And in Ryzodeg, we'll put insulin aspart into the formulation. So the whole concept here is that we've made the world's first and only, to-date, true fixed-ratio combination product between the most-used prandial analogue, namely insulin aspart, and the to-be, I hope most use, ultra-long acting fasting analogue or insulin, namely insulin degludec.

So what do you do in your Phase III program? Well, you compare up against different kinds of paradigms. We've compared it against basal-bolus and shown that twice-daily Ryzodeg can actually be supplemented with a lunchtime NovoRapid insulin, given a total of 3 injections per day can actually beat 4 to 5 injections of basal-bolus insulin therapy. We've done studies like that.

But the most logical study of all is actually to say who is today -- which product is today the biggest one in the premixed insulin segment, and that is, as you are aware I think, NovoMix 30. NovoMix 30 is 5 basic insulin aspart, where it's essentially liquid, soluble, ready-to-absorb insulin aspart, and then crystals of insulin aspart that have been dissolved over a period after 12 to 16 hours. So we are going head on, on a twice-daily basis IDegAsp being Ryzodeg up against a twice-daily NovoMix 30.

And essentially, just to make you understand, that the hangup or the drawback that there was to today's premixes is that NovoRapid does have this profile here as you would like it to see. But first of all -- and that's not so evidently. But first of all, that tends to be a little bit of a carryover effect because of the blunting between the crystals and the liquid component, so that you tend to have a little bit of a shoulder effect over here. But more importantly, in the latter part of the dosing interval, when you give it in the morning and evening, you don't have a steady release of the crystals. It is as if in many patients, after 12, 16, 20 hours, you have the waning and disappearance of the long-acting component. Unlike in Ryzodeg, were these are data that actually have been produced by Tim Heise, who's leading the collab center down in Germany, and he has basically shown the glucose infusion rates basically for Ryzodeg via the insulin degludec component are maintained completely flattish for a full 24-hour-plus period, whereas you preserve the nice NovoRapid insulin aspart effect around the meal time.

So this should give some clinical benefits. And what should those clinical benefits be? Well, on the one hand, and this is unfortunately not so evident here, but in reality, we know from many collab studies that all mixes, NovoMix, Humalog Mix, human mix at 30, they all have this -- they showed that they have a little bit too much insulin effect a few hours after the application, and that typically gives you some hypoglycemia after the meals. So what we call late postprandial hypoglycemia, you see, quite a bit up, unfortunately, with the premixes, that's the expense of having the convenience of injecting a premix, that is this hypoglycemia problem that some patients experience. The other thing is that you would expect that out here because of this relatively intermediate duration profile that you will get some creeping up of the fasting glucose. So you would actually expect that before you inject your next dose, fasting glucose has come up because there's no longer efficacy left in the preparation.

And let's see how that panned out. First of all, this was a treat-to-target study done multicenter in Europe, in Asia and in all kinds of countries, not the U.S. though. U.S. is not so much a premix market. But in many other countries, this was done and presented by Greg Fulcher here as poster. And as a treat-to-target concept, we ended up nicely around the 7% level, 7.1% HbA1c to be precise, which is a good news because it shows that we have treated to the target as desired. But the interesting thing already here is that the fasting glucose levels, they're below 6%. They are below 6% for twice-daily Ryzodeg, whereas they are close to 7% for NovoMix 30 given twice daily. So already here, you can actually see that this difference is essentially accounted for by the lack of insulin out here in the, you can say, pre-injection part of the dosing period. So in the late after-dosing period, you still have the efficacy of Ryzodeg but not so of NovoMix, at least not the full effect.

Now this is all fine and good, but what then happens to what you really worry about for these preparations being hypoglycemia? And I have to say this was key [ph] in the face of more than 10% lower insulin doses. So NovoMix 30, when you inject today's premixes, you will witness -- and also, that actually also goes for Lantus is not crystals in the preparation, but when you inject Lantus, it actually precipitates under the skin. That's the whole concept of protraction of insulin glargine. And when you either have crystals in your suspension or when you have formation of crystals or microprecipitates under the skin, you essentially are creating a situation where the body has to redissolve these things and absorb them. And in that process, you lose quite a bit. So bioavailability for crystal suspensions all for that matter for precipitating solutions, such as insulin glargine, will always be below 100%. Now in this case, this is solution. You don't have to rock and roll and shake the bottle or the cartridge with Ryzodeg because it's a liquid NovoRapid, and it's a liquid degludec. So essentially, that's 100% bioavailability, unlike NovoMix and unlike insulin glargine. So this means, actually, that this was achieved in doses that were more than 10% lower for Ryzodeg compared to NovoMix. And we've seen the same in Lantus comparisons that we essentially have a lower dose of Tresiba to achieve the better fasting glucose control as compared to insulin glargine.

So let's look at hypoglycemia. Well, and here it really becomes exciting because, overall, we have a 32% lower rate with Ryzodeg, significant in this period. And if we look specifically at confirmed nocturnal hypos, it's a 73% stunning lower rate of Ryzodeg compared to NovoMix 30. These reductions are highly unusual. And if you look at the amount of events, these were patients who had quite a bit of hypoglycemia, that have diabetes for like 13 years. They had BMI that's like 29-ish. They are patients with a beta cell that is burned out. So their ability to buffer against swings and fluctuations in the insulin levels after administration is miniscule at this. And this means that we have levels -- even though these are type 2 diabetics, we have levels that are approaching 10 episodes per year. So when you see this kind of reduction and this kind of reduction, the number needed to treat is lower than 1 to save a hypoglycemic event. It's lower than 1. You can treat half a patient, and you already have saved the hypo episode.

So interestingly, when we go into the maintenance period, that means after the titration phase, once they are in steady state in terms of dosing someone, these numbers become even more significant and higher. So we have more than 40% overall reduction. We have close to 80% nocturnal reduction. And when you look at severe hypoglycemia, that's 89% significant reduction in severe hypoglycemia, 89% in the maintenance period in these patients. So this, I think, in terms of value for money -- well, we're not discussing costs right now, but in terms of, you can say, things that help economically can make a mark in the societal cost, such as hypos, severe hypos, et cetera, we actually have a very strong case. Ryzodeg, today, is, as you may know, becoming available in the first country, the country of origin, Denmark, such that we are able to gradually roll out Ryzodeg in a prudent way and consistent with the fact that Tresiba has to come first to market, so people can familiarize with insulin degludec molecule. And then once they are familiarized with that in Japan and so on, so forth, then Ryzodeg can be launched in -- particularly in these premix markets like, for instance, Japan, they used to be at least, and make a significant difference for the patients.

Now one question is, how can you reconcile that you have the same A1c but you have such an improvement in fasting glucose? Because I just told you that HbA1c is a composite of postprandial and fasting. Did these patients then have worse postprandial control for Ryzodeg compared to NovoMix? That's got to be the logical conclusion, otherwise, HbA1c can't be the same. So I just checked that this afternoon, and that's not the case. Even though Greg Fulcher is not going to present those data, I can comfort you, it is so that at breakfast, there's slightly better control for Ryzodeg; at lunch, it's the same; and at dinner, it's slightly less good. So overall, it's a wash. If anything, there's a trend to what's an overall slightly better postprandial glucose control on Ryzodeg compared to Novomix, but it's not significant.

So what is it? This is it. When you have hypoglycemia occurring to this extent, and these are only the ones that the patients are measuring, if you did CGM on these patients and say everything below 3.1 is a hypoglycemia in a CGM setting, it's not 8 episodes, it's maybe 20, 30 episodes per year these patients are having. And when you are in hypoglycemia, which is a dangerous bad condition that can kill you, of course, brain damage or traffic accident, in that situation, hypoglycemia is actually pulling down the HbA1c because during the hypoglycemia, of course, you are not bombarding the red blood cells with glucose because glucose levels are lower. So it's an unhealthy way of getting the HbA1c down, that is experiencing frequent hypoglycemic episodes. So in reality, you can say, yes, we have better fasting glucose control, we have the same postprandial glucose control, but the reason why the A1c ends up being the same is because these guys and ladies are having quite a bit of hypoglycemia spread over the day and the night. And that actually gives you the same A1c but in a very unhealthy way. Same argument goes for some of the insulin glargine comparisons, but we actually are seeing fasting glucose improvements in the case of a degludec compared to insulin glargine, yet the A1c was the same. And postprandial, I mean, it must have been the same also because they only got a basal insulin and nothing else, also explained by the ability of -- or the propensity to give more hypoglycemia in the case of glargine that can account for an HbA1c that ends up being the same in spite of poor fasting glucose control.

So now, and actually, we are pressing ahead, this is -- well, only slightly faster than anticipated. The last presentation is a very exciting one, even though it's not real data yet, it's just baseline demographics and characteristics of the patients in the LEADER trial. The LEADER, as you know, is our post-marketing commitment to the FDA and also to European regulators to study for a time and event-driven period the cardiovascular performance of this agent. It is so that in the U.S., it's primarily driven as a CV safety trial, whereas in Europe, the Europeans, they're also concerned about CV aspects. They really are compared to the U.S. At least, they look at the data, and if it looks good, then they're not concerned. Whereas the U.S. have this cardiovascular guideline that we have adhered to, where products must show a performance like the one you'll see in just a minute.

So what we did was to do a multicenter more than 400 investigator sites all over the world. We -- it's multinational, including countries like China, U.S., India, Europe, all over the place. We needed to recruit slightly less than 9,000 patients. We actually ended up with 9,300 because they were coming in, as you'll see, very fast. And basically, it's a standard-of-care placebo or standard-of-care control, so we have daily blinded injections, either of a liquid called Victoza or a liquid called placebo. And they are blinded such that there's nowhere a difference. And these patients were then going to be treated for a minimum of 42 months, each of them, in that trial. So it's driven by, of course, the amount of events that we need to see to document the safety of the drug, but also the exposure period. We both need events and exposure so that the regulators can say, "Yes, Novo Nordisk, not only have you found this amount of events, but you've also convinced us that the exposure period is long enough for us to have seen any safety signal, if there was one." And this is relevant not only for the cardiovascular performance, it may be even more relevant for the pancreas safety considerations, the thyroid safety. The more and the longer exposure, the more likely you are to see a detrimental effect of the product, if there is one. So the non-inferiority criterion was basically for the MACE, and the MACE is the major adverse cardiovascular events composite endpoint, death, infarction and stroke. And in this one, we assume that we get slightly less than a 2% event rate. Actually, it has turned out to be higher, so we have been conservative deliberately -- well, not deliberately, we were conservative because there was a notion in the market that all these CV trials, they were starting to dry out all the high-risk patients, and we were running the risk that patients would come into trials going forward with lower event rates than in the good old days. It turned out not to be the case.

So we have then done the CV guideline-determined assessment that the FDA guidelines require to establish non-inferiority by showing that the upper 2-sided 95% confidence limit for the relative risk of the primary endpoint is below 1.3. And hierarchically speaking, if and when we show that, we will go on and test whether we have, as Steve Nissen said yesterday at the symposium, no drug has ever systematically been able to show any effect of the good -- well, many assume bad affects on the heart, but have shown good effects on the heart in the field of diabetes medicine. So we will have a chance to be the first. And if we show non-inferiority, the next step in the testing hierarchy is to look at superiority against standard of care, and that is by looking at the upper 2-sided 95% confidence limit for the relative risk of the MACE endpoint being below 1.0, i.e., we statistically can say with 95% certainty that the point estimate of the hazard ratio is significantly below 1. That would be a big day. And we needed 611 events, but 2 things have saved us that we're going to get more than that. 611 is to document non-inferiority. To document superiority, you need more than that. That's a statistical analysis of how you narrow down the upper and the lower bound of your confidence interval. But I can tell you 2 things: one is that we've recruited 600 patients more than required, as you can see up here; the other is that the event rate is not 1.8%, it's a lot higher than 1.8%. And that bodes well in terms of the amount of events that will come out of this trial.

So we're also looking at a number of key secondary endpoints. You can imagine that we're both looking at macro-vascular but also microvascular outcomes, both composite and individually. We're also looking at the time from randomizing the patient to the occurrence of other causes of death. And of course, MESIs, medical events of special interest, not Lionel Messi, but actually -- so that's the real Messi, but these ones are called the MESIs, the medical events of special interest. Those will also follow not only on the soccer pitch but also in the pharmacology level. So these MESIs, they are cancer, pancreatitis, gall stone disease, calcitonin increases, thyroid disease, hypoglycemia, immunogenicity and so on. So we have a long shopping list of things that will be extremely interesting for us to be able to document to EMA and FDA that we have safety on our side once we are completing the trial.

So the patients came in with an overweight of men. This is not so unusual because at this age, you will still have the propensity for more males than females to experience a cardiac event. And this is in an enriched population. We have enriched it such that 80% have had a previous history of cardiovascular disease in their own body, and the rest have risk factors for establishing cardiovascular disease. And that gives us a population of patients who are either on orals or insulin or orals and insulin. So it's a broad spectrum of people with a relatively long-standing disease of 13 years; they're aged 64; BMI is high, it's about 32.5; and the HbA1c is high, it's 8.7. These are not good lives. So if we then look at the execution of the LEADER trial, we decided that we would seek to recruit in a matter of around 18 months, and that's aggressive, to get 9,000 patients onboard in a highly controlled manner all over the world in 18 months. We actually exceeded by 1 month that performance. We really were almost proud about the quality of the recruitment and also the quantity of recruitment and the speed with which it took place. And we do have the steering committee, where leading authorities like Steve Nissen, like John Buse, like Bernie Zinman, like Michael Nauck, people who are either thought leaders in statistics, like Michael Pocock from U.K., so we have either thought leaders on cardiology, on statistics, on diabetology, on tier 2 [ph] and 1 [ph] medicine, stuff like that. They're all sitting there, and I think they like what they're seeing, and they like being part of. This.

So overall, you can see that we have lost less than 1% of the randomized patients to follow-up due to them withdrawing their informed consent. That is good news because the FDA's -- not overly, but appropriately concerned with very poor performance in historical cardiovascular outcome trials because historically, a lot of patients have been lost to follow-up. What happened to them? Did they die from an event? Did they die on placebo? Did they die on your drug? Why did they die? So we are able to maintain about 99% of the patients consenting, such that we will -- even if they drop off drug at some point in time, which hopefully they won't, we as you can see here, have low drop out rate at this point. But still, we can -- even those who drop out, follow them and say yes or no, are they alive? Are they well? Have they had a CV event, once this comes to an end. We have a higher-than-expected rate of events, as I mentioned and we have not, at this point, seen any other safety signals as estimated by the Data Monitoring Committee. I'm not part of that DMC, so I don't know what they're discussing but we haven't been notified of anything, a particular happening. So that was it. We are going to complete the LEADER trial and announce the results, that will be very exciting, by 2016, which is also what we have committed to do with the FDA and EMA. And that gives us time to talk.

Question-and-Answer Session

Michael Novod - Nordea Markets, Research Division

Michael Novod from Nordea Markets in Copenhagen. Just a few questions. Maybe, Mads, going back to IDegLira. Looking at the delay to Tresiba in the U.S. and other companies now saying that they are on track to start Phase IIIs in mid-'14, will this then be your primary offering? Should we expect a simultaneous filing of IDegLira together with Tresiba in the U.S. or how do you see that panning out? And then secondly, do you see or do you have any comments to [indiscernible] NDA for lixi in the U.S. Do you think that is changing the game rules in the U.S., i.e., the possibility of filing on interim analyses and does it change anything for you guys on Tresiba interim and potentially also, obesity for liraglutide where you might have to show some data from the LEADER trial?

Mads Krogsgaard Thomsen

Well, Michael, that was a lot of questions. I'll see if I can remember all of them. First of all, on -- you said other companies, I think you meant 1 other company, is trying to make a combo product and that's lixisenatide and glargine. As late as today, many of you will have seen data on the get go, where they do like a post-hoc analysis and all kinds of things. And the short version is that in our minds, based on these data, there's no doubt that there's a huge difference in the efficacy of these combination products, driven by the fact that we have long-acting -- very long-acting versions of prandial [ph] insulin and they have a half-life that is about half, in the case of glargine compared to Tresiba and about, you can say 1/5 in the case of lixisenatide compared to liraglutide. So as such, we're not afraid of competition but what we can see is that even companies that have loose combinations like albiglutide combined with basal insulin, like Byetta, combined with basal insulin Bydureon, combined with basal insulin -- there is no doubt, and I think you're hinting at that, that the market is excited by data such as the ones I showed you[indiscernible]. They may be less exciting for some drug combinations based on what I have seen, but they're still exciting. And that is something that when you can -- and I forgot to mention that we actually lose a bit of weight. These patients rather put on weight as you do with insulin, and when you add NovoRapid or some other prandial insulin to your basal insulin, not only do you increase your propensity to hypoglycemia but also, you on a couple of kilograms of weight. So we have presented another study here which actually documents a 4-kilogram benefit of the combination of Victoza and Tresiba, compared to the combination of Tresiba and NovoRapid. And a hypoglycemia risk reduction that is very distinct, while at the same time, getting more patients to target. So there's a lot of good stuff to be said for combination of [indiscernible] and insulin, and every physician who knows about this field will agree to that, I think, nowadays. So the issue then becomes what is the primary offering and what is not? I think we will be able to submit IDegLira because we have the dosage already before we submit it in Europe, and we will launch in Europe in the first territories next year. In the U.S., we can submit once we have the interim data from the Devote trial [ph], the CV outcome trial that is going to start later this year worldwide. And that, you would imagine, IDegLira will be given a 12-month review period because it's a new product, whereas, Tresiba will be given a 6-month review period because, essentially, we have the drug application in there and it's then being supplemented by this Devote data. So there will be a time lapse of maybe 6 months, all other things being equal, from we can launch Tresiba until we can launch IDegLira. I think IDegLira has very big potential in very many markets but what you call the principal product and whatnot, it's a long debate. I think the way we see this is that this is a unique opportunity for patients to come in early and come in late and come into control. And some of the hassle that people are witnessing, the physicians, they tend to cycle up and down, then they try metformin, then they try issue[ph] , then they CV, then they try DPP-4, then they try Anti-IL-20[ph]. So the patients are constantly in and out of different drugs. They change medicines quite a lot but they don't get into control, while they are still having hypos and weight gain. Here, you actually have something that if you start early, or even if you start after placebo, whenever you start, it seems as if you can maintain excellent control for very extended periods making it a highly exciting value proposition for patient and physician. In terms of the lixisenatide withdrawal from the U.S., it's highly confusing because there is no change to the notion that the FDA is fully accepting interim analysis. There's no change for that. We even heard Steven Nissen talk about it yesterday at the symposium. So the FDA has not changed the guidance on being able to do interim analysis and let them form the basis for a drug approval and then have, you can say, the end-of-trial results dictate the decisive verdict. And as it, I think, was also mentioned by Dr. Steven Nissen yesterday, that verdict would then be that you confirm the product stays on market or basically, if the decisive data, end-of-trial data were to show that you do not come below the 1.3 cut-off, as we have discussed for the LEADER trial, then you can run the risk of being pulled off the market. These are things that as you know, have been discussed for event and so on and so forth. So I don't quite understand what has happened here. I think you have to ask the folks at Sanofi that question. But it's not relevant to how we see neither liraglutide 3 milligram for obesity. It's not relevant for what we're discussing, neither with the LEADER nor with the Devote trial. Yes?

Lars Hevreng - SEB Enskilda, Research Division

Lars Hevreng from SEB. Can you speculate a bit about the event rates in the LEADER trial with benefits being higher than the trial was designed for because I guess, most of the trial today, we see the opposite. Event rates are lower, so trials are being expanded in terms of patient population. But could you speculate a bit about that?

Mads Krogsgaard Thomsen

Well, first of all, one thing is that the ethnic propensity to develop cardiovascular pool outcomes differs, so that some populations like some Asians, Indians tend to get higher event rates than Caucasians. So we may have a different split. We are a very global company in the way we operate our key operations[ph] so we have Indians, we have Chinese, we have Brazilians, we have -- and Brazilians, they are African-Americans, many of them and African-Americans also have a higher propensity to, unfortunately, get cardiovascular problems. So one thing can be the ethnic distribution. Another thing can be the inclusion criteria that we're using. And our inclusion criteria, I won't go through them, but that's a lot. Either you have a history of CVD and 80%, 81%, to be precise, actually had a history of CVD, personally when they took the trial, giving them the high risk. As we know, I think that was also mentioned yesterday, once you've had a CV episode, then the likelihood that you're going to get a new one is 7% per year. And 7% per year is the same risk that you get appendicitis. And everybody knows, everybody knows, family members and friends and maybe themselves, having the appendix removed because of appendicitis. So walking around every single year, running the same risk of a new attack as getting appendicitis is not a good situation. So something needs to be done about it, that's what we're trying to do. And that basically means that these many patients have that risk. So having it -- and reached the way we have, there will be the higher risk. And the other thing is that the inclusion criteria, in terms of those who didn't have CVD, they have to have a high propensity for development of CVD based on some risk factors that we put into the criteria.

Lars Hevreng - SEB Enskilda, Research Division

And can you just remind us, how many planned interim analyses do you have during the course of the trial? Is it 1 or 2?

Mads Krogsgaard Thomsen

0.

Lars Hevreng - SEB Enskilda, Research Division

0. Okay. So there's no -- okay. And just final, on IDegLira. It seems -- I mean, 2 of -- both of the components when -- once launched, will have undergone CV outcome trials, do you think there is a good or will there be a reason to do an outcome trial with the combination eventually?

Mads Krogsgaard Thomsen

I would -- we have a lot to do on, because I think -- so what can happen here? This is scenario planning. If LEADER comes up positively, i.e., not only in non-inferiority, but we are the first drug in the market to have a superior profile in terms of cardiovascular outcomes, in that event, I will predict that physicians will think that IDegLira has that property because there's liraglutide in there. So no need to do it, you can say. In the event that LEADER shows a reduction in the point estimate, but it fall align achieves significance, and you can see that we were so close. So you get non-inferiority, you're close to superiority, you didn't quite make it, you can argue, this product, with its ability to maintain excellent control in the absence of weight gain, actually with a slight weight loss with few hypos -- hypoglycemia is a huge risk factor, by the way, for cardiovascular problems, as I think everybody is aware. You just have to look at Accor[ph] trial to document that. So then we could be tempted to do something even out of enthusiasm rather than regulatory requirement. But I think right now, we're doing Sustain 6 for semaglutide. We're doing LEADER for liraglutide. We are doing Devote for degludec. I don't think we're going to start a fourth one right now, but maybe someday.

Carsten Lønborg Madsen - Carnegie Investment Bank AB, Research Division

This is Carsten from Carnegie. Just 3 questions here. First on Devote, is that purely event-driven or is there also a treatment time constraint like in the LEADER trial? And then, just a conceptual question, so in a perfect world, without any cost containment or fear of needles or anything, what should patients start out on? Should they start out on, for example, on IDegLira or should they start out on Victoza or dulaglutide? What will happen? And finally, we have about 6 data coming up, are you afraid that dulaglutide will show superiority to Victoza in that trial?

Mads Krogsgaard Thomsen

Okay, Kasper will help me if I forget some of the questions. On the Devote outcome trial for degludec, that is an event-driven thing where we basically need a number of events to go below 1.8, a number of events to ultimately go below 1.3. And those numbers are consistent with the guidelines that I just showed you as part of the LEADER. So the reason why the LEADER trial is also time-driven is that the European said to us, we're convinced about the cardiovascular safety of Victoza, but we're not yet convinced about the pancreas safety and the thyroid safety, so we want time in there as an element. So LEADER is event-driven CV-wise, but time-driven safety-wise in both general perspectives. So for interim degludec, you don't have this thing looking about pancreas and thyroid and calcitonin and whatnot, so there, it's event-driven.

Carsten Lønborg Madsen - Carnegie Investment Bank AB, Research Division

And just a follow-up to that one, because looking at the enrollment times for both LEADER and also other CV trials, are you modeling this data readout point in time a little bit conservatively? It seems that the other companies have been able to do it a little bit quickly quicker. And you also are pretty fast with the LEADER even though you enrolled more patients than you expected.

Mads Krogsgaard Thomsen

It's not many companies that have enrolled faster than you just saw in the LEADER. There are several uncertainties. One is, will we repeat this high event rate performance that we've seen in the LEADER? Which is good for the trial and bad for the patients of course. Will we be able to basically have the same recruitment performance as in LEADER? There are things that -- LEADER is a very sexy trial because you know that you are having a -- or it's a high-profile trial, where you have a chance of getting access to Victoza, 1.8 milligram and you have a 2/3 chance -- no, 50% chance, I guess, but still, a good chance and that essentially means that many people were really wanting to go in there. Here, you can say it could be the same but at the same time, most people do have access to insulins like Lantus and Levemir. And the differential upgrade to Tresiba, from that perspective, may not be as -- seen as being as revolutionary as when Victoza came to market in 2010. But we hope to repeat that performance and if we do that, then you are right. If we have high event rates, we have good recruitment performance, then you can take a piece of paper and do your own on math and then, it's not going to take 3 years and it's not going to take any more than 2 years. So that's why we are guiding 2 to 3 years from onset to interim analysis. You had...

Unknown Analyst

[indiscernible]

Mads Krogsgaard Thomsen

Yes, so this is debated heavily among physicians because there are a number of doctors who say in type 2 diabetes, you are lacking insulin, that's the definition. You have a relative deficiency of insulin, otherwise, you don't get diabetes by definition. Because many people can become morbidly obese, yet they maintain hyperinsulinemia throughout their life. They never get diabetes. It's only those who fail to do so and have a relative deficiency that get the disease. So some would argue that the benefit of IDegLira is that you're taking 2 hormones, GLP-1 and insulin, that are both relatively in need of replenishment. Some of us heard that also at the incretin meeting today, we do know that there's a relative too-low level of GLP-1 in the body when you have a higher need for insulin because of obesity, should be higher not lower. So some are arguing physiologically that IDegLira is the ideal drug to start early and stay -- start and stay on IDegLira because you don't have to go to burden the physician's office and go up and down both in weight and hypoglycemia and new drugs all the time. Others are saying, yes, but there's insulin in there and after all, even though we can do it, very easy, you can just measure 1 glucose, titrate on a weekly basis, these kind of data are part of also the IIIb evaluation where we are making simpler and simpler and simpler to titrate. So yet, there seem to be in there -- so there's an element of titration. Whereas, for GLP-1 agonists, there is no titration at least for the dosage of 6, 12, 18, and it's only if you had some side effects that you dialed down to 12 or 09 or wherever that you can do with Victoza. So some think it's easier to start on Victoza, but everybody believe that IDegLira has the potential to maintain even better levels of control as shown in the DUAL I trial for very long periods, and they would advocate to start immediately. So I think the way we see it is probably, it makes sense for us to continue to have Victoza as the initial injectable medication. Once that fails, you should actually stay on either liraglutide, i.e., Victoza and then, add Tresiba or go to IDegLira directly and stay on that. For ages and once IDegLira, at some point, fails to control glycemia, then you add first 1, [indiscernible], at that time, we'll be [indiscernible] I guess or ultimately, 2 or even 3. So we have a stepwise process that I think the physicians are, when we have the advisory boards, we are having vivid debates about this topic because there's not unanimous agreement that you shouldn't start immediately on IDegLira, actually. On 6, well, I saw as you did, some of the other data today from the program, there's no doubt that in terms of glycaemic control, it looks fully efficacious at the level of 1.8 milligram Victoza, so I think there will be a matching of dula and Victoza, probably having the same level of A1c decrement. On the other hand, the data on body weight management or reduction are less compelling for dulaglutide. Some of you saw today that in the metformin trial, there was no difference in the weight profile between the metformin group and the dulaglutide group. I would say we have consistently, for all comparisons, with the Victoza, seen benefits even when we compared up against, for instance, Bydureon. There was a difference in favor of Victoza. So I think we'll end up in a situation, personal speculation, forward-looking statement, all that stuff, I think that we'll end up in a situation where dulaglutide will not take away the gold standard kind of image of Victoza, but it will be a valuable treatment because it controls glucose, may do slightly less well on weight but of course, it's a nice product based on a once-weekly administration principle for those who want the once-weekly administration principle. Others will prefer to have the flexibility of daily dosing up and down, depending on nausea and so on and so forth. So it will hopefully expand the market together with the existing products.

Unknown Analyst

Pete August[ph. Just for clarification on the LEADER trial. Is the 42 months cut-off that will decide upon when you close the study or is there actually a maximum in terms of number of events that either -- you may assess that this is the maximum we can exit so to speak? And then, in terms of Ryzodeg, if we look at your pricing strategy on Tresiba, then at least we have some kind of a picture of it. It seems as if -- compared with NovoMix, then you have actually a more favorable product. What type of price strategy should we expect from a medical perspective?

Mads Krogsgaard Thomsen

Right. So the last one first. It is true that health economically, there is a lot going our way in terms of Ryzodeg and both glycaemic -- the 2 of the drivers of health care cost, at least for classic and for diabetic medications, is hypoglycemia, of course, severe hypoglycemia and it's glucose control because HbA1c improvement compared to comparators and even fasting glucose improvement, postprandial glucose improvement. When you document that for an agent then, long-term, you will see that the impact on diabetic-related complications become less and so on, so forth. And when you put all that together, this is a highly valuable asset. Problem is that today, NovoMix, even though it's the leader in its class, it's still quite a bit cheaper in many markets than Lantus and Levemir. So we have a situation where in many countries, the basal segment is actually priced higher than the premix. And since we have a basal in there, right? We have degludec as part of Ryzodeg, so we have to consider how to do the pricing. And of course, we'll do that in a way that in those very important premix markets like Japan, like emerging markets and so on, so forth, we will have a good trade-off between cost and, you can say, efficacy. What -- you had one more question?

Unknown Analyst

Yes, in terms of the 42 months call? Can I follow-up on the Ryzodeg, just a clarification. I mean, would you -- in Denmark, you achieved or you have set the same price as Tresiba. Is that a fair notion that I mean, if you can have equal prices on Tresiba and Ryzodeg, then you will be happy?

Mads Krogsgaard Thomsen

We haven't exactly established and communicated the price on a per market basis, but you can actually say in Japan, for instance, we have in the label for Ryzodeg -- I didn't show those data. We have a Japanese study were once-daily Ryzodeg is given against once-daily glargine. We documented clinically and statistically meaningful superiority on the primary endpoint being HbA1c, it's in the Japanese label. So you can argue, if you have price parity between Ryzodeg and Tresiba, and Tresiba is only priced 4% higher than Lantus in Japan, and you have a label that documents superiority, then you have a highly acceptable value proposition. But we haven't been more specific about the details on a market-by-market basis and we'll communicate more on that as time goes by. In terms of the 42 months, the 42 months, all patients are supposed to be treated for at least 42 months after randomization, for the sheer notion that, in particular, the Europeans were at that point, more so than today, I think, which is a little bit strange because of all the media reactions, they were more worried about thyroid safety, pancreas safety. Now we have more exposure, we have more epidemiological data. We have medical studies. We have many things that make us, as a company and as the owner of the product, less worried than when we got the approval. Yet, of course, we have to document that and one way of doing that is having 42 months at least exposure for every patient in the LEADER.

Michael Novod - Nordea Markets, Research Division

Michael from Nordea. Just a few follow-up. First of all, with the head-to-head, you said earlier that you had -- so you've been planning to do a head-to-head against lixisenatide. Have you skipped those plans since the delay in the U.S.? And then second to that also, what about your own head-to-head against dulaglutide, of course, depending on the petition outcome of the AWARD-6? And then secondly, we start to see the Phase III trials reading out during -- next year on SGLT2-DPP-4 combos, fixed combos. What is your view on that given the strong launch of, so far, the SGLT2s in both Europe and the U.S.?

Mads Krogsgaard Thomsen

Right. In terms of lixisenatide, we have a directed-on comparison that is at the stage of execution where it wouldn't make sense to reconsider that one. I think everybody expects what the outcome will be at the trial but it's nicer to have data than to speculate. So for us, it's like almost a no-brainer, it make sense to do head-on because as you recall, Sanofi did do a very short one. They only looked at 4 weeks and the primary endpoint was postprandial glucose control like at 1 meal, not at 3 meals, yet there was already a significant difference on A1c levels in favor of Victoza in that trial after only 4 weeks of treatment. So for us, it's an expectation and typically something we do, do, i.e., head-on comparisons in fair and square ways so that the physician can make informed decisions. In terms of dulaglutide, dula, Lilly is doing the study, the AWARD-6 or whatever trial. And in that regard, you can say, if it comes out as I just made a forward-looking statement about, i.e., that we come out on parity on A1c and we are maybe better on blood pressure, on the body-weight, on local solubility or whatever, then, I don't think there's a need to repeat a study because duration 6 and how many, 7 studies, they were just as helpful to Novo Nordisk. Well, they were actually only helpful to Novo Nordisk, not to the companies that sponsored them. So it's actually very nice to have other companies. Then, of course, if the study comes out and due to study design flaws or whatever, we see something that we think is not reflecting reality, then you might consider to go back to the drawing board and do set it right. But I think Lilly are a very reasonable bunch of R&D people, so I don't foresee that they have done this state-of-the-art as Novo Nordisk would've done. And then, their combo. Yes, SGLT2 and DPP-4, I think, Michael, you will see, INVOKANA or canagliflozin has actually done quite well, as you are seeing. We -- I think that combo products for reasons that may be a little bit defy rationality, you would argue that you should make pure pills and combo pills and whatnot and use a lot of them to make life easier for the patients. Very often, the physicians still prefer to be able to both titrate and individualize these different remedies. But for sure, that is one of the more exciting options to combine is SGLT2 and DPP-4 inhibitors because SGLT2 seem to be complementary to the different drugs. There a number of things that you have to bear in mind, I didn't show that, but in the LEADER trial, you actually have 20% of the patients who already have a kidney function that shows that they have moderate, chronic renal insufficiency. So many of the patients who have been treated with such agents, they actually have compromised kidney functions and do bear in mind that both for SGLT2s, but also for DPP-4 inhibitors, there are concerns about kidney function in the case of SGLT2, you'll have less effects simply because when people stop having impaired glomerular filtration rate, at that point, they will also have impaired glucose excretion rate and hence, the whole thing about turning a diabetic symptom into a treatment, which is essentially what has been done, will work this well and so on, and the DPP-4 because most of them, apart from linagliptin, TRAJENTA from Boehringer, they also have issues about kidney function and so on. So I don't think -- we don't do comparisons against all kinds of different combo products, but we do comparisons against the new class agents such as DPP-4 inhibitors, such as GCD even, and so on and so forth. We'll do that as relevant also with SGLT2 but, do that on a compound by compound basis and then take it from there. And then, physicians can make their own decisions. Yes?

Unknown Analyst

Carl Hampton from DNB[ph] . To follow up on that one, if you look maybe 10 years ahead, on the oral space, you might have an oral GLP-1. Can you kind of speculate ahead what the oral space will look like, oral GLP-1, DPP-4, SGLT2 combo? And so on...

Mads Krogsgaard Thomsen

So I think 10 years from now, you're absolutely right. If oral semaglutide, as the first in the whole class, is actually the only one as a representative of a GLP-1 peptide and a very long-acting one, if that makes it all the way, I would argue that, that will be the first time where a GLP-1 based therapy, unless if the LEADER trial shows superiority, I foresee that physicians will start much more than today using its parity in the treatment cascade because we all know that even though the patients in this trial, they've had CVD, cardiovascular disease for quite some time. We also know that the appropriate time for Novo Nordisk to intervene in that process is not 13 years after the diagnosis, it will be a lot earlier because once you have a rusty pipe, it takes a lot of work to get the rust out of the pipe, if you can do that at all. And the same goes with the arteries. So I actually think that unless LEADER proves superiority, then the fact that we're injecting these peptides will restrict them from early introduction in most cases, unlike with the oral semaglutide, where this could have a pull position. If you assume that you have the weight loss and the glucose lowering and potentially even the cardio protection of a tablet, the way that, that could be transformational, then I think we are talking very early positioning. Novo Nordisk is also developing oral insulins and these ones are also our attempt to make it less invasive so that people don't feel stigmatized and hampered in their lifestyle, the way they do today. So it is introducing an earlier stage when the pancreas can still counter-regulate because I think when you are low down in 5% beta-cell function, the ability to buffer the ups and downs, the swings in the prevailing insulin levels are probably inappropriate for oral therapy because the day-to-day variability will, after all, be higher than for Tresiba, given that's a once-daily injection because that's very low. So I think you could foresee that oral insulin will make early inroad into the diabetes arm interim and so will definitely oral GLP-1 and then, all the injection-based products, they will still exist but they would exist at later stages, maybe not IDegLira because IDegLira probably will also find a way into earlier treatment because of the things we've discussed so, it will be exciting. And I think this will be the last question. We can go out and discuss other things.

Unknown Analyst

And kind of -- the LEADER trial, it's not only this LEADER trial, it's more. But what happened to the beta-cell preservation debate? I mean, it's still a fundamental problem. Why has that kind of -- why not looking at that one anymore when it comes to liraglutide?

Mads Krogsgaard Thomsen

I think that what we have come to realize is that the turnover, that the beta-cell turnover in humans is not counted in weeks like it is in mice. So we can -- if we inhibit apoptosis and if we increased replication of [indiscernible] to even mature beta cells, some claim inroads, we can do that very efficiently and actually grow the functional beta-cell mass which we've met in particular in animals that are hypoglycemic, not so in normal glycaemic animals. That has been difficult to verify in humans. It simply seems as if the turnover time for beta cells in humans is so long that even a drug like a GLP-1 agonist like liraglutide to influence that in a way that you can see a different outcome in terms of more beta cells will: a, take many years to document; and the likelihood that will occur because of these reasons is probably somewhat less than what we had hoped a few years back. So there's not so much beta-cell preservation debate ongoing. So in the scale obesity program, where we have great data on prediabetes and diabetes actually also, where we really reduced, in a very meaningful way, those you can say numbers, it's more driven I think by pharmacological effect and the weight loss. The weight loss and the beta cell improvement rather than the fact that they are actually getting more beta cells.

Okay. Thank you very much. There's a little hors d'oeuvre and drink, I guess, out there. So we can just proceed for the next 30 minutes or so.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!

This Transcript
All Transcripts