Novo Nordisk 1Q14 – Diabetes Care grows 1% to $2.9 billion in challenging quarter; Tresiba revenue reported for first time (~$14 million); Tresiba CVOT ahead of schedule; oral insulin moving to phase 2 – May 2, 2014

Executive Highlights

  • Global Diabetes Care revenue grew a modest 1% as reported to DKK 16 billion (~$2.9 billion), driven by Levemir (up 21% as reported to DKK 3.1 billion [~$570 million]) and Victoza (up 9% as reported to DKK 2.9 billion [~$540 million]).
  • Novo Nordisk reported Tresiba sales for the first time, which were DKK 80 million (~$14 million) in 1Q14.
  • Tresiba’s CVOT, DEVOTE, enrolling more quickly than planned; US re-submission could take place ahead of schedule (now ~late-2015 instead of early-2016 to early-2017).
  • Novo Nordisk plans to advance an oral insulin to phase 2a for the first time in 2015: OI338GT (NN1953). The other two oral insulin candidates will be discontinued.

Novo Nordisk provided its 1Q14 financial update in a call led by CEO Mr. Lars Sørensen. Global Diabetes Care sales totaled DKK 16.0 billion (~$2.9 billion), up 1% as reported (6% in constant currencies) from the previous year. Novo Nordisk faced several headwinds in 1Q14 including a relatively challenging comparison to 1Q13, the loss of major US formulary contracts for Victoza and NovoLog (insulin aspart), the potential impact of SGLT-2 inhibitors delaying GLP-1 initiation, generic competition to Prandin, and inventory changes at the wholesaler level. Despite all of Victoza’s challenges, it remained one of the company’s growth drivers in 1Q14, growing 9% as reported to DKK 2.9 billion (~$540 million). Levemir was another driver of growth, rising 21% as reported to DKK 3.1 billion (~$570 million). The loss of Victoza and NovoLog contracts with the pharmacy benefit management organization Express Scripts (ESI) had a tangible impact on the company’s bottom line. Management took jabs at BMS/AZ and Lilly, who respectively won the GLP-1 and rapid acting insulin analog contracts with ESI, suggesting that the strategy of aggressively “buying your way” into a single-source formulary does not always reap rewards. Excitingly, the company reported sales of the ultra-rapid-acting insulin Tresiba (insulin degludec) for the first time, which totaled DKK 80 million (~$14 million) in 1Q14 – penetration is promising in markets where Tresiba receives reimbursement similar to Lantus, but continues to be hindered in countries like the UK and Denmark by a lack of reimbursement to cover the product’s relatively high premium. Finally, the company adjusted its full-year 2014 forecast downwards slightly.

On the pipeline side, on a positive note, Tresiba’s cardiovascular outcomes trial (CVOT), DEVOTE, is enrolling more quickly than expected, meaning that Tresiba may well be re-submitted in the US ahead of schedule (now ~late-2015 instead of early-2016 to early-2017). We aren’t sure what the implications are at this point for patients in the trial if interim results are released and don’t know the company’s views on this either. Also on an exciting note, Novo Nordisk also announced that it is advancing its first oral insulin candidate into phase 2a in 2015: OI338GT (NN1953), a basal oral insulin candidate using Merrion’s GIPET carrier system. It is discontinuing its two other oral insulin candidates in their current form (OI287GT and OI362GT). Encouragingly, another set of impressive phase 3 results was announced for IDegLira (which has now been named Xultophy – pronounced ZUL-tu-FIE. In DUAL III, the fourth of five phase 3 DUAL trials to report results for Xultophy, people with inadequate glycemic control on a GLP-1 agonist were randomized to Xultophy or to continue on existing GLP-1 therapy. Switching to Xultophy provided a mean 1.4% A1c reduction from a relatively low base of 7.8% (the control arm achieved a mean 0.4% A1c reduction). In addition, results for two new phase 3b trials for Tresiba and one for Ryzodeg were announced. The trials supported Tresiba as an add-on to Victoza, established Tresiba 200 U/ml’s comparability to Lantus (with better reductions in fasting plasma glucose and lower rates of hypoglycemia), and established Ryzodeg as statistically similar to full basal-bolus therapy (however, Ryzodeg did not meet the non-inferiority margin for the trial). Finally, the fourth of six phase 3a trials for once-weekly GLP-1 agonist semaglutide has begun (SUSTAIN 1). It will investigate semaglutide monotherapy compared to placebo in 400 drug-naïve people with type 2 diabetes.

Top Five Financial Highlights

Table 1: Sales of major diabetes products


1Q14 Revenue (Billions)

Reported / Operational Growth from 1Q13

Reported Growth from 4Q13

Total Diabetes Care

DKK 16.0 ($2.9)

1% / 6%


Total Modern Insulins

DKK 9.4 ($1.7)

4% / 10%



DKK 3.9 ($0.72)

-3% / 2%



DKK 2.4 ($0.43)

-2% / 5%



DKK 3.1 ($0.57)

21% / 27%


Human Insulins

DKK 2.6 ($0.47)

-9% / -5%


“New Generation Insulins” (Tresiba)

DKK 0.080 ($0.014)




DKK 2.9 ($0.54)

9% / 13%


Conversion from DKK to USD assumes an exchange rate of 0.1836.

1. Total Diabetes Care sales rose a relatively modest 1% as reported in 1Q14 (6% in constant currencies) to DKK 16.0 billion (~$2.9 billion). By product, the GLP-1 agonist Victoza (liraglutide) and the basal insulin Levemir (insulin detemir) were the main growth drivers of 1Q14. By geography, North America, International Operations, and China were the biggest drivers of growth, responsible for 45%, 27%, and 24% of total Diabetes Care growth, respectively (growth in each region was 7%, 12%, and 18%, respectively). Overall, this was a challenging quarter for the company, and the hardest year-over-year comparison the company will face in 2014 (sales grew 13% in 1Q13 compared to 11% in 2Q13, 3% in 3Q13, and 4% in 4Q13). Additionally, there were a number of other headwinds, including loss of positioning on some key formularies (see bullet #4 below), generic competition to Prandin, and inventory changes at the wholesaler level – collectively, according to the press release, these factors had a negative impact of ~5% on total Diabetes Care growth.

  • Following a re-evaluation, the company decided to revise its financial outlook for the full-year 2014 downward. The new guidance for sales growth in local currencies is 7-10%, down from 8-10%. The new guidance for sales growth in reported currencies is about 4.5 percentage points lower than local currency growth, compared to earlier forecasts for reported growth that is 3.5 percentage points lower. This means the new guidance window for reported growth is roughly 2.5-5.5%, a range that falls short of the 8% reported growth that Diabetes Care achieved in the full-year 2013. We can only wonder what the figures for 1Q14 and the forecast for the year would have looked like if Tresiba (insulin degludec) had been approved in the US last year. Management commented during Q&A on the factors leading to the low guidance target, which include:
    • A ~2% negative impact of Express Scripts formulary losses,
    • A ~1.5% negative impact of a reduction in wholesaler inventory levels, 
    • Lower than expected GLP-1 agonist growth in the US and Europe,
    • Changes in the US rebate environment, namely the flow of patients from managed care to Medicare Part D (which requires the drug manufacturer to help cover the “donut hole”).

2. Novo Nordisk’s modern insulins (NovoLog/NovoRapid, NovoMix, and Levemir) collectively grew 4% as reported (10% in constant currencies) in 1Q14 to DKK 9.4 billion (~$1.7 million) – notably, Levemir was the primary growth driver. Sequentially, the modern insulins declined 8% in 1Q14 against a challenging comparison (4Q13 modern insulin sales rose 8% sequentially). Novo Nordisk’s share of the total insulin market fell slightly from 48% in February 2013 to 47% in February 2014, although its share of the modern and next-generation insulin market held steady at 46% (see table 2 below). Within the modern insulins, growth was stronger in North America (7% YOY as reported) than ex-North America (2% YOY as reported). Notably, during Q&A, management commented on the volume-price mix for the modern insulins in 1Q14, sharing that more than two-thirds of the growth was due to pricing and less than one-third was due to volume (most of the volume growth was due to Levemir). This was a somewhat worrying comment, as it means that the fairly modest growth this quarter is not built on a wholly sustainable foundation. We wonder if the increase in pricing was intended to pre-empt the loss in volume from the loss of formulary contracts in 2014, or if it was more class-wide; some speculate the insulin pricing will continue to increase until biosimilars are available, at which point, pricing will fall back down – and the companies will look better broadly speaking in putting in place price reductions (which they ostensibly would not want to do from the lower pricing bases of 2012 and 2013. We do hear more patient backlash on pricing as patients are being asked to take on higher co-pays and higher deductibles. We are at the start of trying to understand what is accessible to US patients under the ACA and will be reporting more on that in the future. .

  • The basal insulin Levemir (insulin detemir) grew a striking 21% year-over-year (YOY) as reported (27% in constant currencies) to DKK 3.1 billion ($570 million). This represents Levemir’s largest reported YOY growth since 4Q12 when it grew 27% - we were a bit surprised to see Levemir grow so much and also continue to be surprised about the price increases though maybe it is done to match Sanofi’s Lantus price increases. Sequentially, Levemir sales fell 2% in 1Q14, the lowest sequential decline among the company’s major Diabetes Care products. We do imagine Levemir’s strong performance may also be due, in part, to increased marketing resources that the company had built up in anticipation of Tresiba’s US launch (which was postponed with the FDA Complete Response Letter) going towards marketing Levemir instead. The company attributed much of Levemir’s strong 1Q14 showing to robust market penetration. An increase in share for Levemir would have likely come at the expense of Sanofi’s class-leading Lantus (insulin glargine), but the impact was not clear from Sanofi’s 1Q14 results. After correcting for the impact of stocking, Lantus’ 12% growth in 1Q14 was consistent with previous quarters. However, a pricing increase for Lantus could have somewhat masked any effect. Additionally, Levemir is obviously growing from a small base, given that the Lantus market (~$2.0 billion quarterly) is much bigger.
    • Growth for Levemir was strongest in North America (36% as reported, 41% in constant currencies to DKK 2.0 billion [~$360 million]). The product also performed well in China, growing 56% as reported; however, the company’s sales base (DKK 81 million [~$15 million]) is relatively small in that market. The product performed relatively poorly in Europe (down 1% as reported YOY) and Japan & Korea (down 15% as reported YOY).
    • We imagine that the introduction of a biosimilar insulin glargine would have the potential to hurt Levemir significantly as well as Lantus. Currently, Lilly/BI have a biosimilar insulin glargine under review (halted in the US by a Sanofi lawsuit; see our Lilly 1Q14 Report), Merck and Samsung Bioepis have a biosimilar glargine entering phase 3 (read our report), and Biocon/Mylan have a biosimilar insulin glargine on the market in over 10 countries (mostly in emerging markets) and is beginning a global phase 3 program (read our Biocon F4Q14 Report). 
  • NovoLog (NovoRapid; insulin aspart) sales fell 3% as reported and rose 2% in constant currencies to DKK 3.9 billion (~$720 million). The year-over-year comparison was challenging, as NovoLog sales grew 14% as reported in 1Q13. Sequentially, sales fell 12% as reported, also against a moderately challenging comparison (sales were up 9% sequentially in 4Q13). The sluggish performance of Novo Nordisk’s flagship rapid-acting insulin was due in part to some tough formulary exclusions in the US this year – Express Scripts (the largest pharmacy benefit management group in the US) was the highest-profile case, but it was not the only one (read highlight #4 below). Management shared during Q&A that the company saw a noticeable drop in NovoLog sales volume in early 1Q14 due to the Express Scripts decision, but has since seen the effect stabilize. That trend was clearly visible in a graph included in the presentation slide deck, showing that the percentage of US patients with unrestricted access to NovoLog fell sharply at the beginning of 2014 by 15-20% (value estimated from graph) and subsequently stabilized at that level.
    • Outside the US: NovoLog was weakest in International Operations (down 11% as reported to DKK 360 million [~$65 million]), in the US (down 6% as reported to DKK 2.2 billion [~$410 million], for the reasons discussed above), and in Japan & Korea (down 3% as reported to DKK 220 billion [$40 million]).
  • NovoMix sales were also hurt by the Express Script formulary issue – global sales fell 2% as reported and rose 5% in constant currencies in 1Q14 to DKK 2.4 billion (~$430 million). This result was due in part to a challenging comparison, as NovoMix sales grew 12% as reported in 1Q13 (the best quarter for the product in 2013). Sequentially, sales fell 7% as reported. During the presentation, management highlighted the sharp decline in the premix insulin segment in Europe, where NovoMix sales fell 8% as reported in 1Q14 to DKK 550 million (~$100 million). NovoMix sales fell 12% in the US to DKK 580 million (~$110 million).
  • Novo Nordisk’s human insulin portfolio saw sales shrink 9% as reported (5% in constant currencies) in 1Q14 to DKK 2.6 billion (~$470 million). Sequentially, sales fell 5% against a somewhat challenging comparison (human insulins grew 5% sequentially in 4Q13, the year’s only quarter of positive sequential growth). Novo Nordisk’s human insulins appear to be slowing even in China, a market that used to provide stronger human insulin growth (by comparison, modern insulins seem to be doing increasingly well from a growth perspective in China, although from a much smaller base – we think that China will continue to show strength because such a relatively smaller percent of patients there are on insulin).

Table 2: Novo Nordisk’s share of the insulin market, by region


Novo Nordisk’s share of the total insulin market

Novo Nordisk’s share of the modern and new-generation insulin market

Feb 2014

Feb 2013

Feb 2014

Feb 2013

Global Total















International operations















3. Victoza (liraglutide), Novo Nordisk’s once-daily GLP-1 agonist, grew 9% as reported (13% in constant currencies) in 1Q14 to DKK 2.9 billion (~$540 million). This result was particularly impressive given that 1Q14 had a fairly challenging comparison: sales grew 35% as reported in 1Q13 compared to 26% in 2Q13, 14% in 3Q13, and 19% in 4Q13. As is evident from those historical figures, Victoza sales underwent a slump over the course of 2013, due in large part to the incretin-pancreatitis safety controversy (read our Incretins-Pancreatitis Primer for a full chronicle of the issue). From 1Q14’s results alone, it appears that the slump might not yet be totally over. Management noted that the growth of the GLP-1 class in the US and Europe has not met the company’s initial expectations for the year, impacted, we believe, by the entry of SGLT-2 inhibitors, although it does look like a rebound is beginning. There was certainly worry coming into 1Q14 on how the Express Scripts formulary loss (see highlight #4 below) to AZ’s Byetta and Bydureon (exenatide) would impact Victoza. However, the press release notes that Victoza sales in the US were buoyed by an increase in pricing.

  • However, despite the uncertainty regarding the class’ growth prospects, management was optimistic on the outlook for Victoza’s share of the market, despite the Express Scripts loss. Management shared that Victoza’s market share saw a “limited” drop of market share in early January, but that the company is now seeing a slow increase in total Rx share in the US. Management suggested that these results indicate that patients and doctors are fairly unwilling to switch off of Victoza (we have also heard in the past that Express Scripts patients can be grandfathered into continuing to take Victoza). In a graph included in the presentation slide deck, we saw that there was a nearly 20% reduction in the number of US patients with unrestricted access to Victoza at the beginning of the year, followed by a very slight (2-3%) rebound (values estimated from a graph). It is uncertain whether the class can resume the growth seen pre-2012, but in any case (as management pointed out), the class should hopefully return to double-digit growth.
  • To help explain the less-than-robust rebound of the GLP-1 agonist class, CEO Mr. Lars Sørensen himself emphasized the entry of SGLT-2 inhibitors. He directly suggested that, had SGLT-2 inhibitors not been a factor, the GLP-1 inhibitor market rebound would likely have been stronger. A question was posed during Q&A on how SGLT-2 inhibitor/DPP-4 inhibitor fixed-dose combinations might impact the GLP-1 agonist class (since these FDCs could offer a glycemic and weight benefit on par with that of GLP-1 agonists, but without the injection burden or nausea). Management was a tad bit evasive, noting that Victoza is generally used after patients fail co-therapy on multiple oral agents (patients who might already be taking both a DPP-4 inhibitor and an SGLT-2 inhibitor and are still not at target), suggesting that it did not think that SGLT-2/DPP-4 FDCs would significantly impact GLP-1 agnoist initiation. Management did acknowledge that an FDC would have allure from a patient convenience angle – one is a pill, the other is an injection. We agree that the convenience factor is key, given the need to improve adherence. If SGLT-2/DPP-4 inhibitor combinations can achieve efficacy on par with GLP-1 agonists (which is not a sure thing, but it is fairly likely from what we understand), then they become a very enticing option for providers and patients, given convenience along with the lack of nausea with the oral combination.
  • During Q&A, management discussed how Victoza might stack up against the once-weekly newcomers to the GLP-1 agonist field, namely Lilly’s dulaglutide (under review in the US and EU) and GSK’s Tanzeum/Eperzan (albiglutide; recently approved in the US and EU). Mr. Sørensen suggested that dulaglutide will be comparable to Victoza on blood glucose lowering, and hazarded a guess that other parameters might be similar as well (dulaglutide did prove non-inferiority to Victoza in the AWARD-6 trial – read our report). Given the likely level of similarity (although once-daily vs. once-weekly administration is a big difference), Mr. Sørenson suggested that Lilly should not focus on undercutting Novo Nordisk, but rather focus on growing the GLP-1 agonist class. He brushed off GSK’s Tanzeum, suggesting that its lower efficacy would make it a less meaningful competitor; we are curious to see how GSK markets this compound.

Table 3: GLP-1 agonist market share


GLP-1 agonist share of total diabetes care market (by value)

Novo Nordisk’s share of the GLP-1 agonist market (by value)

Feb 2014

Feb 2013

Feb 2014

Feb 2013

Global Total















International Operations















4. Management suggested that pricing warfare to secure single-source formularies like Express Scripts ultimately does not help the contract winnerin managements’ words, “buying [your] way into an exclusive contract” does not necessarily yield stronger numbers the following quarter. In the case of both of Novo Nordisk’s lost contracts (one lost to AZ’s exenatide franchise, and another lost to Lilly’s Humalog), management suggested that the victor did not come away with much benefit in terms of sales growth, given the discounting required to secure those contracts. The point that the Express Scripts saga was a no-win situation (except, perhaps, for Express Scripts) would seem to point to market expansion as the better avenue for long-term growth, although management did acknowledge that companies will generally take advantage of the weakness of a competitor in the short term.   

  • Although the increasingly competitive diabetes pricing environment, aggressive pricing might not be a way to achieve growth, but it might be necessary to keep from falling behind. Examining the 1Q14 results of the three players in the Express Scripts decision shows that the contract-winning products outperformed the contract-losing products, even if the winning results weren’t always impressive. Lilly’s Humalog (a contract winner) grew 3% as reported in 1Q14, while NovoLog fell 3%. AZ’s Bydureon grew 47% as reported in 1Q14, while Victoza grew 9% (albeit from a ~5-6x larger base). Therefore, while buying placement on a formulary may not get you ahead, losing out on that contract will likely mean that you will be behind.
  • In a bigger-picture sense, the most worrying trend at play is the loss of choice for patients and HCPs. Back when the Express Scripts decision was announced, we were surprised that the pharmacy benefit manager would be dramatically reducing options within a class that is relatively heterogeneous. Options to slow the progression of this trend include increased patient advocacy before payers, and changing the regulatory-reimbursement paradigm so that clinical trials are more aligned with the outcomes payers are interested in (to improve the overall reimbursement climate)

5. Notably, the company broke out Tresiba (insulin degludec) sales for the first time: the ultra-long-acting basal insulin achieved sales of DKK 80 million (~$14 million) in 1Q14. This is compared to sales of DKK 9 million (~$2 million) in 1Q13, although the product was not on the market for all of 1Q13 (it received EU marketing authorization in late January). The company has now launched Tresiba in 12 countries – the most recent additions are Germany, Malta, Bangladesh, and Lebanon. Novo Nordisk plans for launches in 20 additional countries during 2014.

  • The presentation slide deck included a graph showing Tresiba’s market share in a set of seven countries. In countries where the product receives reimbursement similar to that of Sanofi’s Lantus (insulin glargine), Tresiba has steadily grown to take a 10-17% value share of the basal inuslin market. In countries where reimbursement is not on par with Lantus, penetration is more modest (1-3%). Out of the countries with comparable reimbursement, Switzerland, Mexico, and Japan have the best uptake of Tresiba (basal insulin market shares of 17%, 10%, and 10%, respectively), and we are pleased to see such a diverse range of countries in the top three, which reinforced Novo Nordisk’s strong and diverse management team.
  • Although $14 million might appear to be a modest total for a new-generation insulin that has been approved for over a year, Novo Nordisk has held to its decision to price the product at a substantial premium, accepting slower penetration as a result. CEO Mr. Lars Sørensen commented extensively on this decision during the company’s 3Q13 update, saying that it should not be up to the US to pay for innovation. The company firmly believes that Tresiba’s long duration of stable action, potential for flexible dosing, and reduction in hypoglycemia merit a premium. We aren’t sure that US payers will accept this but we are glad Novo Nordisk is being open about its strategy. We believe the vision is more acceptable in light of truly breakthrough products like Xultophy (IDegLira).

Top Five Pipeline Highlights

1. Tresiba’s cardiovascular outcomes trial (CVOT), DEVOTE, is enrolling more quickly than expected, meaning that Tresiba may be re-filed in the US ahead of schedule. DEVOTE began in October 2013, at which time management forecast that it could conduct a pre-specified interim analysis two-to-three years after initiation of the trial (late 2015-late 2016). Novo Nordisk now expects, under the conservative assumption that the event rate during the trial is around 2% or slightly higher each year, that enough events could accrue by mid-2015 to conduct the interim analysis. Preparing the re-submission package to the FDA would take at least two months, suggesting that the re-submission could occur sometime in late 2015. During Q&A, management suggested that since the trial is largely based in the US, the actual event rate might likely be higher, around 3%, which would serve to speed interim analysis even more. As a reminder, Novo Nordisk intends to use the interim data to re-submit Tresiba to the FDA to satisfy its requirement to preliminarily exclude a 1.8 hazard ratio from the upper bound of the 95% CI. The results of the completed trial will be used to satisfy the requirement to exclude a 1.3 hazard ratio. Trial completion is now expected three to five years from trial initiation (late 2016-late 2018), whereas it had initially been expected four to six years from trial initiation (late 2017-late 2019).

  • DEVOTE is a double-blind trial (n=7,500 people with type 2 diabetes ≥50 years old with existing or at high risk for CVD) using insulin glargine (Sanofi’s Lantus) as a comparator. The primary endpoint is MACE (CV death, non-fatal myocardial infarction, non-fatal stroke), and it is an event driven trial with the estimated primary completion date currently listed as November 2018 on (Identifier: NCT01959529).
  • In terms of other next generation basal insulins in development, Sanofi’s Toujeo (new U300 insulin glargine formulation) leads the pack. Sanofi expects to submit Toujeo in 2Q14 in the US and Europe. Timelines are a bit hazy for Lilly’s PEGylated lispro (peglispro) and Lilly/BI’s new insulin glargine formulation: Lilly/BI’s insulin glargine formulation has been set back by a lawsuit from Sanofi – while it was filed in the EU in mid-2013 and in the US in December 2013, Sanofi’s lawsuit triggers a hold on any regulatory action for as long as 30 months, which could push approval back until 2016. Topline phase 3 data for Lilly’s peglispro are expected in 2Q14. Lilly management has remarked that it would only launch peglispro if it felt it could significantly differentiate it from Lantus, and the phase 3 data readout should help show whether phase 2 results (weight loss instead of weight gain, less hypoglycemia) are borne out in further trials. Merck and Samsung Bioepis have a biosimilar glargine entering phase 3 (read our report), and Biocon/Mylan have a biosimilar insulin glargine on the market in over 10 countries (mostly in emerging markets) and is beginning a global phase 3 program (read our Biocon F4Q14 Report). 

2. Novo Nordisk plans to advance its first oral insulin candidate into phase 2a in 2015: OI338GT (NN1953). It is discontinuing its two other oral insulin candidates in their current form (OI287GT and OI362GT). OI338GT is a novel long-acting basal insulin analog that uses Merrion’s GIPET carrier system to carry it from the GI tract into circulation. Novo Nordisk announced that OI338GT completed three clinical pharmacology studies in a total of 118 healthy volunteers and people with type 2 diabetes, supporting its safety and dose-dependent glucodynamic effects. Novo Nordisk is planning for OI338GT’s phase 2a trial to be a proof-of-principle trial investigating glucose lowering and safety, including rates of hypoglycemia, after individual titration of OI338GT in people with type 2 diabetes. Based on results of this phase 2a trial, larger phase 2b proof-of-concept studies may be initiated.

  • Oral insulin is notoriously difficult to develop given the very narrow therapeutic window for insulin coupled with the variability in drug absorption when taken orally and the susceptibility of proteins to degradation in the stomach.
  • Of the companies currently developing oral insulins (including Biocon, Oramed, Generex, Aphios and others), Novo Nordisk has the most experience and the most resources to devote to the endeavor (in October 2013, Novo Nordisk management estimated that $3 billion would be required to develop oral insulin and oral GLP-1 agents).
    • Generex recently received a $2.1 million capital investment, part of which will support its phase 3 buccal insulin spray (absorbed through the cheek), Oral-Lyn. Efficacy results for Oral-lyn have been modest, with <0.5% A1c reduction after 12 weeks in people with type 2 diabetes and demonstrated glycemic inferiority to injected human insulin in type 1 diabetes.
    • Oramed’s ORMD-0801 recently reported positive phase 2a topline results in 30 adult patients with type 2 diabetes. The study had been requested by the FDA to establish safety prior to moving into larger phase 2 studies. We heard some more detail at GTC Bio 2014 Day #2, and we recently learned that the company admitted a “formulation issue” that compromised the 24 mg dose of the drug – this seemed quite troubling and further called into question Oramed’s place in the healthcare world. .
    • Biocon’s prandial oral insulin, IN 105, is in phase 2 with results from US clinical trials expected to read out toward the end of its FY 15 (April 2015-March 2015).
    • Aphios’ APH-0907 has been studied in animals, and the company is looking for a partner to advance to clinical studies.
    • While not exactly an oral insulin, MannKind’s inhalable prandial insulin, Afrezza, received a positive FDA Advisory Committee vote in April. Its PDUFA date, which had previously been April 15, 2014, has been pushed back until July, 15 2014. We will be interested to see Afrezza’s market expanding impact on insulin and diabetes therapies in general.

3. As noted, IDegLira (Novo Nordisk’s great Tresiba/Victoza [basal insulin/GLP-1 agonist] fixed-ratio combination) now has a brand name – Xultophy; Novo Nordisk released impressive results of DUAL III, the fourth of five phase 3 DUAL trials for Xultophy to report results.  In DUAL III, 438 people with type 2 diabetes who had inadequate glycemic control on a GLP-1 agonist (Victoza or exenatide) were randomized 2:1 to either switch to Xultophy or to continue on the pre-existing GLP-1 receptor agonist therapy for 26 weeks. From a baseline A1c of 7.8%, the Xultophy group achieved a mean 1.4% A1c reduction to 6.4% compared to a mean 0.4% reduction to 7.4% in the control group. In the Xultophy arm, 75% of people achieved an A1c <7% and 63% achieved an A1c ≤6.5%. In the control arm, only 26% and 23%, respectively, achieved those goals. As would be expected, the rate of hypoglycemia was higher in the Xultophy arm, but was still characterized as “low.” Patients in the Xultophy arm gained an average of 2 kg (4.4 lb) compared to a loss of 0.8 kg (1.8 lb) in the control arm.

  • All phase 3 results presented for IDegLira have been highly, highly promisingvery striking. As background, DUAL I and DUAL II results were presented in 2013 at ADA and IDF, respectively; Novo Nordisk released DUAL IV data during its 4Q13 financial update.  DUAL I compared IDegLira to Tresiba and Victoza alone. In DUAL I IDegLira provided a 1.9% A1c reduction compared to a 1.4% reduction with Tresiba and 1.3% reduction with Victoza. The IDegLira arm experienced nearly one-third less hypoglycemia compared to insulin alone and also experienced slight 0.5 kg (1.1 lb) weight loss compared to a 3 kg (6.6 lb) weight loss on Victoza alone and a 1.5 kg (3.3 lb) weight gain on insulin alone. DUAL II demonstrated that the Victoza component of IDegLira contributed significantly to the combination’s efficacy on glycemic control, with the IDegLira arm experiencing 1.9% A1c reduction compared to 0.9% reduction on Tresiba alone. DUAL IV studied IDegLira as add-on to SFU with or without metformin; patients experienced a 0.9% placebo-adjusted A1c reduction and a 1.5 kg placebo-adjusted weight difference in favor of placebo.
  • We believe that the DUAL III study population (people with inadequate control on a GLP-1 agonist) represents a very relevant patient population for Xultophy. Xultophy offers an option for escalating from GLP-1 agonist therapy to insulin therapy without adding any extra injection burden. In addition, we imagine that most patients who are already taking a GLP-1 agonist are covered by insurance that would also ultimately provide access to Xultophy.
  • Novo Nordisk filed Xultophy for regulatory approval in Europe in May 2013, and its regulatory submission in the US depends on the fate of Tresiba. Sanofi is also working on a GLP-1 agonist/basal insulin fixed-ratio combination, LixiLan (Lyxumia/Lantus), which is currently in phase 3. Sanofi hopes to file in late 2015, which means that LixiLan will be behind Xultophy in Europe but likely ahead of Xultophy in the US.  

4. Novo Nordisk presented the results of three phase 3b trials for Tresiba (ultra-long acting insulin degludec) and Ryzodeg (pre-mixed Tresiba/Novolog):

  • The first trial studied Tresiba as an add-on to Victoza in people who with type 2 diabetes and inadequate glycemic control on Victoza (A1c >7%). This double-blind 26-week trial randomized 346 people to adding once-daily Tresiba or placebo to background Victoza treatment. From a baseline A1c of 7.6%, Tresiba patients experienced a mean 1.1% A1c reduction compared to a mean 0.1% reduction in the placebo group. In addition, 78% of people in the Tresiba group were able to achieve an A1c <7%, compared with 36% of people in the placebo group. Not surprisingly, hypoglycemia was statistically significantly higher in the Tresiba group, but still characterized as “low.” The Tresiba group experienced a mean weight increase of 2 kg (4.4 lb) compared to a weight loss of 1.5 kg (3.3 lb) in the placebo group.
    • Management also announced during the call that the European Commission has just approved the use of Victoza and Tresiba in combination in any sequence. This follows the company’s announcement in March of a positive CHMP opinion for these expanded indications.
  • The second trial demonstrated that a doubly concentrated formulation of Tresiba (Tresiba 200 U/ml) conferred similar glycemic control as insulin glargine with a statistically larger reduction of fasting plasma glucose and a statistically significantly 40% lower rate of confirmed hypoglycemia. This was a randomized, cross-over trial with two 16-week treatment periods for 145 people with type 2 diabetes requiring high-dose insulin treatment. After a 16-week run-in period with insulin glargine, patients had a mean baseline A1c of 8.2% and were randomized to Tresiba 200 U/ml or insulin glargine. Patients also reported a higher level of satisfaction with Tresiba’s FlexTouch pen compared to Sanofi’s SoloStar pen (the pen used for Lantus [insulin glargine] administration).
  • The third trial demonstrated that Ryzodeg twice daily (pre-mixed Tresiba/Novolog) provided a statistically similar A1c reduction as full basal-bolus therapy using Tresiba once daily and Novolog two-to-four times daily. However, the trial did not meet the non-inferiority margin (unspecified) for the primary endpoint. In the 26-week trial, 274 people with type 2 diabetes on background basal insulin treatment needing intensification to mealtime insulin were randomized to Ryzodeg or basal-bolus. From a baseline A1c of 8.3%, the Ryzodeg group achieved a mean A1c reduction of 1.3% to 7.0% compared to a 1.5% reduction in the basal-bolus group to 6.8%. There was no statistically significant difference between the two groups, but Ryzodeg did not meet the pre-specified non-inferiority margin. This is not necessarily surprising given that full basal-bolus allows for finer titration of insulin doses, but we imagine that the reduced injection burden associated with Ryzodeg will be a tradeoff that many patients are willing to make (and would likely promote better adherence and, therefore, better results in real life).

5. Management announced that it initiated SUSTAIN 1 in February, the fourth of six trials in semaglutide’s phase 3a SUSTAIN program. Semaglutide is Novo Nordisk’s novel once-weekly GLP-1 agonist for type 2 diabetes. SUSTAIN 1 will investigate semaglutide monotherapy compared to placebo in 400 drug-naïve people with type 2 diabetes for 30 weeks.

  • SUSTAIN 6 (semaglutide’s CVOT), SUSTAIN 2 (semaglutide vs. Januvia), and SUSTAIN 3 (semaglutide vs. Bydureon) are the three other phase 3a trials for semaglutide that are currently underway. SUSTAIN 6 completed recruitment in just 10 months. During Novo Nordisk’s 4Q13 update, management remarked that phase 3 discontinuation rates have been much lower than in phase 2, suggesting that tolerability has been better in phase 3 than it was in phase 2 (when Novo Nordisk claims to have over-estimated the potency of the drug and used quite a high dose and too rapid of a titration scheme).
  • The remaining two SUSTAIN trials, SUSTAIN 4 (semaglutide vs. Lantus) and SUSTAIN 5 (semaglutide as add-on to basal insulin), are expected to begin in 2014.
  • The long-acting GLP-1 agonist market is starting to crowd, but Novo Nordisk believes that semaglutide will be able to sufficiently differentiate itself. In phase 2, semaglutide appeared at least as good, if not better than, Victoza, which is the current GLP-1 market leader.
    • AZ’s once-weekly Bydureon and GSK’s once-weekly Eperzan (albiglutide) are currently the only once-weekly GLP-1 agonists on the market (both are approved in the US and EU).
    • Lilly’s once-weekly dulaglutide was filed in the US and EU in October 2013. Novo Nordisk considers Lilly’s dulaglutide as the most meaningful competitor for its GLP-1 products. Dulaglutide is the only GLP-1 agonist that has demonstrated non-inferiority to Victoza.
    • Intarcia’s ITCA-650, the once-yearly exenatide osmotic mini-pump, is in phase 3 trials and reported striking interim results at JP Morgan 2014 showing >3% A1c reductions in a high baseline A1c population (>10%).

Honorable Mention: In March 2014, Novo Nordisk launched the “Cities Changing Diabetes” initiative – read our coverage of the announcement here, which includes an interview with Novo Nordisk USA President Mr. Jesper Høiland and our speculation on some considerations for a US city partnership. This initiative aims to map out urban diabetes and potentially identify best practices in development, dissemination, and coordination of solutions amongst multiple stakeholders. The program launched in Mexico City in partnership with the city government, University College London, and Steno Diabetes Center. Novo Nordisk expects to make future partnerships with cities in North America, Europe, and Asia.

Table 4: Novo Nordisk Diabetes and Obesity Drug Pipeline

Drug Name




Other Remarks

Tresiba (insulin degludec)

Ultra-long-acting basal insulin analog

Type 1 and type 2 diabetes

Approved OUS; CRL in the FDA

Re-submission to the FDA expected after interim analysis of DEVOTE CVOT occurs around mid-2015

Ryzodeg (insulin degludec/insulin aspart)

Pre-mixed basal/bolus insulin

Type 1 and type 2 diabetes

Approved OUS; CRL in the US

Re-submission to the FDA expected after interim analysis of DEVOTE CVOT occurs around mid-2015

Xultophy (formerly IDegLira; insulin degludec/liraglutide)

GLP-1 agonist/basal insulin fixed-ratio combination

Type 2 diabetes

Filed in the EU May 2013

Awaiting Tresiba decision in US

Liraglutide 3 mg

GLP-1 agonist


Filed in the US and EU December 2013


FIAsp (NN1218)

Ultra-rapid-acting insulin analog

Type 1 and type 2 diabetes

Phase 3

All four trials in the phase 3 “onset” program have been initiated. See Novo Nordisk 4Q13 report.


Once-weekly GLP-1 agonist

Type 2 diabetes

Phase 3

Four of six SUSTAIN trials have begun. SUSTAIN 1-5 expected to complete in late 2015 and SUSTAIN 6 (CVOT) in early 2016.

LATIN T1D (liraglutide for type 1 diabetes)

GLP-1 agonist

Type 1 diabetes

Phase 3

Phase 3 initiated in 4Q13. “Hoping” for a 2015 launch.

OG217SC (oral semaglutide; NN9928)

Oral GLP-1 agonist

Type 2 diabetes

Phase 2

First phase 2 oral GLP-1 ever. Phase 2 initiated in 4Q13. Novo Nordisk may be able to take small shortcuts wth regulatory proceedings sine it is already conducting a full program for injectable semaglutide. Uses Emisphere’s Eligen carrier technology.

OI338GT (NN1953)

Oral basal insulin analog


Phase 2 expected to begin in 2015


LAI287 (NN1436)

Ultra long-acting basal insulin (potential for once-weekly dosing)


Phase 1

Completed first phase 1 trial in July 2013, one month ahead of schedule.

OG987GT (oral liraglutide; NN9926)

Oral GLP-1 agonist


Phase 1

Uses Merrion’s GIPET oral delivery technology

OG987SC (oral liraglutide; NN9927)

Oral GLP-1 agonist


Phase 1

Uses Emisphere’s Eligen carrier technology

OG217GT (oral semaglutide; NN9928)

Oral GLP-1 agonist


Phase 1

Uses Merrion’s GIPET oral delivery technology

OI362GT (NN1954)

Oral basal insulin analog


Discontinued 1Q14

Uses Merrion’s GIPET oral delivery technology

OI287GT (NN1946)

Oral basal insulin analog


Discontinued 1Q14

Uses Merrion’s GIPET oral delivery technology

Questions and Answers

Q: One question on US pricing. Can you speak to how Express Scripts is really impacting the market? Does this mean payers have worked out how to break the oligopoly such as diabetes? Or do you think this is just a temporary bump where we return to normalization next year?

A: As you know, we lost the Express Scripts contract for NovoLog and NovoLog Mix, and then we lost the Express Scripts contract on Victoza, up against the exenatide family; you can speculate on how that changes the pricing power of suppliers on different diabetes products, injectable diabetes products in the US. If I should give you a brief rundown, I think we can all see the weekly market shares, you will be able to see that there is a negative impact on the Victoza market that has been significant and recent. It’s clear to see that it happens right after January 1. However, it’s a limited drop in market share we’re seeing, and right now, we’re seeing actually a slow increase in the total prescription market share in the US.

I would say there has been a limited willingness from patients and doctors to switch from Victoza over to the exenatide family; consequently, that’s probably also a very limited gain for the net sales of Bydureon and Byetta as a consequence of getting higher rebates but not really gaining any significant volume. From a supplier point of view, it has not been a very beneficial change in the contract we set up for the competitor.

In the space of fast-acting insulin, it’s slightly different. We see a slightly higher impact from the change in reimbursement status for NovoLog with Express Scripts – we see it leveling off. If we look at the public data from Lilly on their sales of Humalog in the US, then that is an indication that its net gain has been very limited; of course, the company paid a higher rebate to get a slightly higher market share, but it doesn’t look like in its first quarter that there is significant value gain on that part. Overall, I would say the jury is still out on whether this is something that we’ll see continued in the future. It at least doesn’t seem to drive significant profits for suppliers who go into these kinds of arrangements.

Perhaps we would also note that the methodologies used by Express Scripts and the blocks have been quite effective, more effective than we have seen in similar contracts. Therefore, we also felt the impact faster than we anticipated.

Q: If you have mid-2015 data for the DEVOTE trial that is positive, how fast can you add that to the filing data and file with the FDA? What is the best-case scenario there?

A: The best-case scenario on such data sets is about two months. That is not based on the total analysis of all data, but rather primary analysis of the mass data. So two months.

Q: Could you update us on the LEADER trial event rate for cardiovascular events? When is that expected to report? Have you taken the event rate that you’re seeing into account for the new timelines for the DEVOTE trial?

A: When we planned the LEADER trial back in the post-approval days of 2010, our assumption was ~2-2.1% annual events of MACE based on historical performance in such trials. You’re probably also aware that since we have updated the market, the factual event rate is more to the tune of 3% or more in LEADER. That’s based on very substantial evidence surround that particular disease population.

When we designed the DEVOTE trial, you first have to realize that geographies may pan out slightly differently for DEVOTE than for LEADER, and you can see that differences do play a role as to how you imaging what the MACE event rates will be. That has also meant that we have decided to have a precautionary approach planning for an event rate in the ballpark of 2% or more on an annual basis; however, of course, there’s the opportunity that this population is at least as sick as the population we had in the LEADER trial – then we are in the situation where we cannot make predictions because the planned event rate of 2% may seem on the shy side. On the other hand, the trial is too early to make any more firm predictions. Additionally, this is more US-based while LEADER has a more global distribution in terms of geographies.

With that said, the estimated time completion for the LEADER trial in the first half of 2016. At this time, we will have provided a great deal of MACE evidence to analyze both the non-inferiority and the hierarchical structure of the statistic analysis plan thereafter pending data.

In the case of DEVOTE, what we have is an assumption that something can happen with the planned MACE rates in mid-2015, per the protocol. It’s simply not driven by any assumption about the event rate, but rather the fact that the recruitment has panned out quite favorably and much ahead of schedule.

Q: If the event rate in DEVOTE is actually 3% rather than 2%, when could that study read out? Do you have any visibility on the geographic variation that led you to plan slightly differently?

A: Obviously right now what has been the primary determinate for our assessment of when we’ll have data ready for interim analysis has truly been a question of having exposure to patients. At this point, we have not even guesstimated, as I mentioned, what the exact MACE event rate we’re going to see. However, you are of course correct in the sense that if it is a higher rate than the protocol specifies, things will happen earlier. At this point, we are speaking just a few months at this point, but it will have a bigger impact on the end of the trial. The biggest impact is obviously when you’ve fully recruited the trial, and then there is a huge difference as to whether or not you have 1%, 2%, 3%, or 4% annual event rates. That goes without saying, with a swing factor of two-fold or more.

Geographically-wise, you did ask the question about this as a vial-based, blinded vial-based trial, and that means that there are many geographies we are not operating in to the same extent we were in the LEADER trial. This is because, basically, vials are not as population in, for instance, Western European countries. When that is set, I cannot go out and claim to you that we are seeing clearly dissimilar MACE event rate, or at least we don’t know the data because they’re blinded by this. It is not my interpretation that there are very distinct differences at this point, and they are just lagging because the geography is different in this trial. However, it is too early to tell whether we will get to 3% MACE events.

Q: I appreciate your comments on the market slowdown for GLP-1 agonists, but when we look at the volume growth that IMS gives us, we’ve seen a nice recovery off the GLP-1 market – within that, Victoza does quite well. What particularly surprised you outside the faster-than-expected forced switches that Express Scripts has been driving?

A: We are very positive in terms of the outlook for the Victoza market share in the GLP-1 segment. We saw a small decline in the US due to EBITDA in the first quarter, but we’re seeing it stabilize and start to grow again now. Worldwide, we see good development of the market share for Victoza. We are still very positive on that. We still see growth, just like what you referred to in IMS. The comment is more to the growth level came down last year in the second half of 2013, and we have really seen the rebound of the growth level, which is what we were hoping for. We are very positive about the segment in general, but we just don’t know if it will actually rebound to the sort of growth levels we saw in 2012 or early 2013, or whether we will stay at the current level. In all cases, we will still see nice double-digit growth of Victoza sales.

This is where the SGLT-2 inhibitor comes in – had we not had SGLT-2 inhibitors coming into the marketplace, the rebound for the GLP-1 agonists might have been stronger given the clearances we have seen with both the FDA and the EMA. However, now the SGLT-2 inhibitors offer a new treatment choice and so, consequently, they snap up costs that are intended for either DPP-4 inhibitors or GLP-1 agonists.

Q: Could you provide some color on whether you expect a benefit to the GLP-1 market with dulaglutide and albiglutide launches? Or, given the slightly lower growth that you’re seeing, what’s the risk that the new players GlaxoSmithKline (GSK) and Lilly are more aggressive on share gains than the market expansion that we’ve been expecting?

A: I would say that dulaglutide and albiglutide is, in a way, similar to Victoza and exenatide in that our expectations are that dulaglutide will be comparable to Victoza on blood glucose lowering. We don't know what the other part of the profile will look like. It's anybody's guess, but it will be similar. Let's just assume it will be similar. I believe that Lilly, like us, would see an interest in expanding the market segment.

If GSK is a case study in a learning from the Express Scripts experience of BMS/AstraZeneca on what they gained from buying their way into that exclusive contract, it them actually not gaining anything. I would tend to believe that the players will look to expand the segment if they have long-term interest in this field. I mean, your guess is as good as mine. Weakness on part of competitors often leads people to make short-term move.

Q: Given that you’ve cited SGLT-2 inhibitors, how incrementally concerned should we be about the SGLT-2 inhibitor/DPP-4 inhibitor fixed-dose combination? Those combinations launch late next year, and have a potential A1c and weight loss profile not materially dissimilar to Victoza.

A: First of all, my view is that Victoza even though we are advocating the use of Victoza as a great first line or second line therapy after metformin, that's what we've been saying ever since 2009. We also know that in reality Victoza is typically best chosen after, for instance, DPP-4 inhibitors, where the stay time typically is short, but because of the more modest efficacy. Obviously when you have a combination test it's between, for instance, a DPP-4 inhibitor and an SGLT-2 inhibitor, and you do get the added benefit of more efficacy and some weight loss, which is driven mostly by the SGLT-2 inhibitor component. However, you have to bear in mind that Victoza is typically used today when people have been on two or even three OADs in advance. In that scenario, it doesn't make a big difference. Of course, the difference lies in the convenience of using the combo rather than two individuals pills before injection-based therapy.

Q: Ahead of actually all the phase 3 data read-outs next year, could you just update us on your thoughts regarding the opportunity to win back the Express Scripts contract, or more broadly your willingness to offer more rebates?

A: I would say I don’t think there is a fundamental change to the competitive situation. The situation is mainly determined by the demand side that developed stable in terms of demand for injectable diabetes therapy and there are no major launches from the supply side. We have the same competitors as we normally have, and we have the same number of competitors as normal. With regard to rebating, we constantly optimize our rebate strategy with the aim of securing long-term growth and profitability of our business. You will see us continue to do that, and that’s done on a case-by-case basis. We think we’re relatively good at it, and that’s what we intend to do going forward, also, so optimization on a long-term profit level.

Q: Could you give us a simple price X volume dynamic that you're seeing for US modern insulin business at the moment?

A: You could say that price is slightly more than two-thirds and market share and volume increase is less than one-third, the volume increase primarily coming from, of course, Levemir, in the US.

Q: Clearly the capital expenditure investment has come a little bit too late for the hemophilia franchise to come earlier, I suppose. When would you consider doing capital expenditure investment for the oral GLP-1 and oral insulin products in order to avoid these issues?

A: This is an interesting point, and as you note we’ve run into a situation where the progress of our hemophilia pipeline has been more successful than we anticipated. It’s interesting. Looking at it historically, we had a surplus capacity a number of years because we actually expected to be able to expand the indications for NovoSeven. Now since we have to manufacture several different types of clotting factors, the complexity of doing so influences our expanding capacity.

What are the implications going forward for a successful plan in the oral space? We are already moving oral insulin forward. We already have decided to move oral GLP-1 forward. I can tell you that we already have internally started the more theoretical discussions, but this will also entail a different kind of investment in an active ingredient capacity expansion. This would also include active ingredient capacity expansion. The full value chain, the new active ingredient manufacturing with purification and then different downstream formulation and tableting since these are oral products.

It’s our current best guess that you should start to see us including such investments in 2016 and onwards, and they will drive our capital expenditure up about 1% of sales in the years, and to the extent that we are successful. Eventually, that’ll be dependent on the clinical results, but, much like we did with NovoSeven, we have to invest a little bit ahead of the curve. Otherwise, we won’t be ready for market, and, of course, we would really like to be able to serve the market with products in our key therapeutics areas – namely, this is diabetes for oral GLP-1 agonists and oral insulin.

Q: I notice that you have been launching Tresiba in Germany, but do you have any kind of pricing with the German authorities? How are we supposed to consider this launch? Is it full, or it is part of a shadow market in Germany?

A: There is no shadow boxing. We are all in, so to speak, on the German launch. It’s a full-blown launch of a uniquely good long-acting insulin, and we are very encouraged about the strong feedback we’ve seen in Japan, Switzerland, and other markets. We are following the standard German pathway, and we will have the different interactions with IQWiG [Institute for Quality and Efficiency in Health Care – provides evidence-based reports] and G-BA [Federal Joint Committee in Germany for reimbursement] during the 12 months.

Q: 2014 is kind of a tail of two halves. If I look at Q1, you had 7% underlined currency adjusted growth. In the press release you called out the 5% impact from Express Scripts and Prandin. If I net that off, that’s 12%; you also point out that tails off the headwind in the second half, yet your lower end of the guidance is 7% implies that there is some sort of headwind that could actually linger into 2H14. I’m wondering what could drive that headwind.

A: If you look at the 5%, there is a 1.5% element of that which only hits the first two quarters and has a limited effect; it also hits in Q3, and from then on it has a benefit effect. You could say Express Scripts effect – which we would estimate around 2% of points of sales – would have a similar effect throughout all of the quarters. The final one which we are noting is an impact around 1.5% points, which is what we see in a lowering of the inventory levels, and we see that predominately related to the first half of the year. What we’re seeing in the US compared to our original expectations is that everything else on a lower GLP-1 overall market growth. We’re also seeing a lower market growth in Europe, and the can have an impact.

On top of that, I would argue that the US rebate environment still provides some uncertainty for us. We are seeing that there is a gradual flow of patients moving from an under-managed care set option moving to a Medicare Part D set-up. That is actually providing rebate costs in terms of us covering a part of a donut hole under Medicare Part D. That continues to account for some of the effect. Then volatility in the international operations will always also fluctuate between the years. Everything else being equal, we will probably have a slightly lower part of tender orders in the second half of 2014 compared to what we had in the second half of 2013. Those would be the factors that would drive us down toward the 7% mark in the worst-outcome scenario.

Q: On China, you call out 90% growth, but I just wanted to know what the underlying growth is in China when you exclude the distributor inventory levels being raised?

A: I would estimate probably ~5% is the split between 4Q13 being shifted into first quarter of this year. So, I would guesstimate ~2-5% on China growth.

-- by Jessica Dong, Manu Venkat, and Kelly Close