Novo Nordisk 2Q15 – Diabetes/obesity portfolio up 8% operationally, driven by Levemir and Victoza; modest results for liraglutide in type 1 diabetes – August 6, 2015

Novo Nordisk provided its 2Q15 financial update this morning in a call led by CEO Mr. Lars Sørensen. Our top ten highlights (top five financial highlights and top five pipeline highlights) from the call are included below. 

Top Five Financial Highlights

1.  Novo Nordisk’s diabetes and obesity portfolio posted sales of DKK 21.3 billion (~$3.2 billion) in 2Q15, up 8% year over year (YOY) in constant currencies and 25% as reported, driven largely by Levemir (insulin detemir) and Victoza (liraglutide).

2. Novo Nordisk’s modern insulin portfolio (NovoLog/NovoRapid, NovoMix, and Levemir) collectively grew 5% YOY in constant currencies (22% as reported) to DKK 12.6 billion (~$1.9 billion) in 2Q15, driven largely by Levemir (up 7% operationally and 25% as reported to DKK 4.5 billion [~$670 million]).

3. Victoza (liraglutide) sales rose 25% in constant currencies (47% as reported) YOY to DKK 4.5 billion (~$665 million) in 2Q15. It has lost some share to Lilly’s Trulicity (dulaglutide) and GSK’s Tanzeum (albiglutide) but has been buoyed by an accelerating comeback for the full class – we think the class will continue to expand based on better and better applications and the emergence of basal/GLP-1 combinations.

4. Sales of the Tresiba (insulin degludec) franchise reached DKK 330 million (~$49 million) in 2Q15, up from DKK 271 million (~$41 million) in 1Q15, with sales heavily dependent on reimbursement. The German IQWiG recently handed down a negative assessment for GLP-1/basal combination Xultophy (insulin degludec/liraglutide).

5. Saxenda’s post-launch script trajectory continues to be modest, though that may change with recent low-tier formulary wins with ESI and CVS.

Top Five Pipeline Highlights

1. Novo Nordisk announced positive but modest topline results from the ADJUNCT TWO study of liraglutide in type 1 diabetes. The more robust ADJUNCT ONE study is scheduled to report within three months.

2. The timeline for DEVOTE (the CVOT for Tresiba) has been moved up yet again, with the trial now scheduled to complete in mid-2016 rather than the second half of 2016.

3. During Q&A, Chief Scientific Officer Dr. Mads Thomsen expressed cautious optimism that the ongoing LEADER CVOT for Victoza could plausibly demonstrate superiority.

4. Novo Nordisk initiated a phase 2a trial of its primary oral insulin candidate OI338GT (NN1953) and a phase 1 trial of a new oral insulin OI320GT (NN1957) last month.

 5. Novo Nordisk highlighted the positive results from the SUSTAIN 1 phase 3a trial of its once-weekly GLP-1 agonist semaglutide.

Top Five Financial Highlights

1. Novo Nordisk’s diabetes and obesity portfolio posted sales of DKK 21.3 billion (~$3.2 billion) in 2Q15, up 8% year over year (YOY) in constant currencies and 25% as reported. Sequentially, sales rose 7% as reported. The operational results are in line with those seen in 1Q15 (9% growth operationally and 24% as reported) and with the company’s 4Q14 forecast of 6%-9% growth in 2015.

• By region, growth was driven primarily by North America and International Operations, with more weakness in Europe and unusually sluggish performance in China (there are concerns about reimbursement there). Management and analysts on the call during Q&A spent some time discussing the aberrant results in China, which were attributable to timing of shipments, lower overall diabetes market growth, and increased competition. Management noted that the company’s insulin portfolio has been getting “slapped from both sides” by Sanofi’s Lantus (insulin glargine) on the branded end and a locally made insulin glargine biosimilar and human insulins on the other end – we could see this spreading to other countries with the broader introduction of insulin glargine biosimilars.
• By product, Levemir (insulin detemir) and Victoza (liraglutide) remained the main drivers of growth, in line with what we’ve seen in recent quarters. Each product was responsible for around 45% of growth in diabetes and obesity (operationally) in 2Q15. New ultra-long-acting basal insulin Tresiba (insulin degludec) accounted for a roughly 12% share of operational diabetes portfolio growth in 2Q15 – a meaningful percentage though certainly short of what the company might have initially hope for two years post-launch (the main drag of course being the lack of US approval so far – though we aren’t sure what access will look like in the US).
• According to our model, combined sales for Novo Nordisk’s basal insulin analog portfolio represented 26% of the total ~$2.6 billion branded market in 2Q15. Combined Levemir (insulin detemir) and Tresiba (insulin degludec) sales came in at DKK 4.9 billion (~$700 million). This share has been creeping up over the couple years. Sanofi’s Lantus (insulin glargine) and Toujeo (U300 insulin glargine) netted €1.7 billion (~$1.9 billion) this quarter, for a total market share of 74%. These calculations include only Novo Nordisk and Sanofi and do not include any products available only in the developing world such as biosimilars.
• Novo Nordisk held 51% of the total ~$1.5 billion branded rapid-acting insulin analog market in 2Q15 with ~$775 million in NovoLog/NovoRapid (insulin aspart) sales. Lilly’s Humalog (insulin lispro) held 43% of the market with $654 million in sales and Sanofi’s Apidra (insulin glulisine) and Afrezza (inhaled human insulin) came in at 7% of the market with ~$105 million in sales. These calculations include sales from Novo Nordisk, Lilly, and Sanofi only.
• Novo Nordisk’s GLP-1 agonist, Victoza (liraglutide), held 70% of the total ~$960 million GLP-1 agonist market this quarter. AstraZeneca’s Bydureon (exenatide once weekly) and Byetta (exenatide twice daily) held a combined 23% of the market – coming in second – with sales totaling ~$222 million. Newcomer Trulicity (Lilly’s dulaglutide) held 5% of the market with $44 million in sales, Sanofi's Lyxumia (lixisenatide) held 1.2% of the market with ~$11 million in sales, and GSK’s Tanzeum (albiglutide) rounded out the market with its 1.4% share and $14 million in sales according to our model. Trulicity and Tanzeum are essentially neck and neck in terms of market share by volume according to IMS Health data included in Novo Nordisk’s slide deck, with the disparity in sales due to Tanzeum’s lower price point.

Table 1: Financial results for Novo Nordisk’s major diabetes products  

Table 2: Whole-company sales and growth by geography

2. Novo Nordisk’s modern insulin portfolio (NovoLog/NovoRapid, NovoMix, and Levemir) collectively grew 5% YOY in constant currencies (22% as reported) to DKK 12.6 billion (~$1.9 billion) in 2Q15, driven largely by Levemir (up 7% operationally and 25% as reported to DKK 4.5 billion [~$670 million]). Levemir and NovoLog (insulin aspart) both saw slight gains in market share in the US. IMS data included in the presentation slides showed Levemir reaching 30% new to brand prescription (NBRx) share and close to 30% total prescription (TRx) share by volume – given what we’ve heard about rebating with Levemir we imagine share by value may be slightly lower. Novo Nordisk’s past few updates have suggested that Levemir has been gradually but consistently gaining ground in the basal insulin market over the past year or so. However, Sanofi noted in its 2Q15 update that its total market share (Lantus [insulin glargine U100] and Toujeo [insulin glargine U300] combined) measured by new prescriptions (NRx) had remained fairly stable. On the rapid-acting insulin side, we imagine that the slight gains for NovoLog may have been due to formulary gains vs. Lilly’s Humalog (insulin lispro), as alluded to in Lilly’s 2015 guidance call in January, though the effect does not appear to have been as dramatic as that of the 2014 Express Scripts contract awarded to Humalog at NovoLog’s expense (the impact of which was all too visible in the market share data included in the slide deck – quite a drop at the beginning of 2014). During Q&A, management noted that there had been a slight negative pricing impact for the insulin portfolio in 1H15 and stated that the company expects flat pricing in 2016.

Table 3: Novo Nordisk Insulin Market Share

• Levemir (insulin detemir) continued to post strong results in 1Q15: sales grew 7% YOY in constant currencies and 25% as reported to DKK 4.5 billion (~$670 million). Levemir has posted consistently strong sales results in the last two years, with total 2014 growth averaging 23% as reported. Sequential reported growth in 2Q15 was also strong at 11%, despite a challenging comparison with sales rising 10% sequentially in 1Q15. 1Q15 was Levemir’s strongest quarter in the past few years, with sales growing 31% as reported to DKK 4.1 billion (~$617 million). Management noted that Levemir was the largest driver of growth within modern insulins, which as a category accounted for 34% of whole-company growth for 1H15.
  
  • As in past quarters, North America was the primary driver of Levemir growth. North American sales grew 9% in constant currencies (15% YOY as reported) to DKK 3.2 billion (~$480 million) in 2Q15. IMS data included in the presentation slides showed Levemir reaching around 30% new to brand prescription (NBRx) share and ~28% total prescription (TRx) share by volume in the US. Novo Nordisk’s past few updates have suggested that Levemir has been gradually but consistently gaining ground in the basal insulin market over the past year or so. However, Sanofi noted in its 2Q15 update that its total market share (Lantus [insulin glargine U100] and Toujeo [insulin glargine U300] combined) measured by new prescriptions (NRx) had remained fairly stable. If and when Novo Nordisk’s newer ultra-long-acting basal insulin Tresiba (insulin degludec) is approved in the US, Novo Nordisk plans to fully shift promotional focus to it from Levemir.
  • Ex-US Levemir sales were less strong but still grew 5% as reported to DKK 1.2 billion (~$190 million). Sales in China were particularly disappointing given strong sales in the region in previous quarters: Levemir fell 8% operationally (rose 30% as reported) to DKK 95 million (~$14 million) from DKK 83 million (~$15 million) in 2Q14. Management noted that growth in China was driven by increased market penetration of all three modern insulins, but that increased competition has had a “slight negative” influence on insulin prices more broadly. During Q&A, management cited intense competition from both Sanofi’s Lantus (insulin glargine) and a local producer of biosimilar insulin glargine in China in the provincial competitive bidding process and a larger cooling off of the healthcare market in China as contributing factors to the deceleration of Novo Nordisk’s modern insulin growth there. Management also pointed out that the central government in China has implemented new policies to control costs. The company repeatedly emphasized that the growth potential is still there in China, but results will likely not come as easily as in recent quarters.
  • Compared to last quarter’s call, there were some signs that Novo Nordisk considers Lilly’s biosimilar insulin glargine a significant competitor with Levemir. Management referred to the complexity of the basal insulin market during Q&A, noting that it will soon consist of Levemir, Sanofi’s Lantus (insulin glargine), Lilly’s Basaglar (biosimilar insulin glargine), Sanofi’s Toujeo (U300 insulin glargine), and Novo Nordisk’s Tresiba (insulin degludec). Last quarter, management speculated that biosimilar insulin glargine will compete most directly with Lantus, which is likely correct, but we imagine that it could potentially draw some share from Levemir as well. The relative pricing, reimbursement, and patient support programs for Novo Nordisk’s and Sanofi’s respective next-generation basal insulins (Tresiba and Toujeo) will likely go a long way in determining how the companies will fare. As a reminder, we learned during Sanofi’s 2Q15 update that Lilly’s biosimilar glargine (Abasaglar in Europe) is priced at a 15-20% discount in the three countries where it has been launched. This falls on the lower end of the range that has been speculated over the past few months, though the price cut may well be steeper in the US given the higher baseline prices for basal insulin analogs. We also think Lilly won’t necessarily be the “bellweather” as far as biosimilar insulin pricing goes since it has an incentive (Humalog) to keep insulin analog pricing high.  
• NovoLog (NovoRapid; insulin aspart) sales rose 6% YOY in constant currencies and 23% as reported to DKK 5.2 billion (~$775 million) in 2Q15. Sequentially, sales rose 12% as reported against an easy comparison (sales fell 3% sequentially in 1Q15). Management referred to NovoLog as a “key growth driver” and noted continued underlying volume growth and market share gains. IMS data included in the presentation slides showed NovoLog’s NBRx and TRx share in the US stabilizing at around 50%-55%, after a dramatic drop in sales following the 2014 Express Scripts contract awarded to Lilly’s Humalog at NovoLog’s expense. We imagine that the slight gains for NovoLog may have been due to formulary gains vs. Humalog, as alluded to in Lilly’s 2015 guidance call in January, but the effect does not appear to have been as dramatic as that of the 2014 Express Scripts contract.
  
  • By region, NovoLog performed best in International Operations (markets other than North America, Europe, China, Japan, or Korea) – sales rose 27% in constant currencies and 18% as reported to DKK 596 million (~$88 million). Sales in China and in Japan & Korea were also strong: sales in China rose 16% in constant currencies and 20% as reported to DKK 201 million (~$30 million) while in Japan and Korea sales rose 16% percent in constant currencies (-20% as reported) to DKK 227 million (~$34 million). Reported growth was also positive in North America (up 2% [3% operationally] to DKK 3.1 billion [~$467 million]) and Europe (up 3% [4% operationally] to DKK 1.1 billion [~$156 million]).
• NovoMix sales rose 15% as reported and stayed flat in constant currencies at DKK 2.9 billion (~$423 million). Sequentially, sales rose 4% as reported. During Q&A, management noted that NovoMix continues to enjoy double-digit growth in China – reported growth in China was 15% though operational growth was only 4%. Management also highlighted NovoMix as a major driver of growth in International Operations (where it was up 8% as reported and 13% in constant currencies). These results are largely in line with expectations given the wider market for premixed insulin in these regions. Growth was sluggish or negative in North America (8% as reported, -5% in constant currencies), Europe (-3% as reported, -8% in constant currencies), and Japan/Korea (-20% as reported, 1% in constant currencies). We imagine that the rollout of Ryzodeg (insulin degludec/insulin aspart) will cut into NovoMix sales in the coming quarters, and with better product combinations on their way to deal with both fasting and postprandial glucose (especially basal insulin/GLP-1 agonist combinations) we are not hugely optimistic about the long-term future of the premixed insulin category – we don’t see it going away, just growth slowing.
• Sales of Novo Nordisk’s human insulins rose 13% as reported and fell 1% in constant currencies to DKK 2.8 billion (~$423 million). Sequentially, sales fell 4%. In constant currencies, human insulin sales continued to grow in International Operations (up 38% [-3% as reported]), but were down in all other markets.

3. Victoza (liraglutide) sales rose 25% in constant currencies (47% as reported) YOY to DKK 4.5 billion (~$665 million) in 2Q15. Sequentially, sales rose 13% as reported. The product appears to be continuing its rebound from a somewhat sluggish 1H14, with 2Q15 being its fourth quarter in a row of consistently increasing double-digit sales growth (and that from a rising base). Management highlighted Victoza as a strong driver of growth overall and particularly in North America, Europe, and Japan.

• The rise in sales was driven by growth for the overall GLP-1 agonist class, which appears to be picking up some real steam. IMS Health data included in the presentation slides showed volume growth for the class rebounding to slightly over 15% as of June in the US after dropping sharply to ~8% at the beginning of 2015 following a period of 20%-25% growth in early 2013. Moreover, the trajectory of class sales looks to be going nowhere but up.  We think this will only continue with better and better applications over time as well as the emergence of basal/GLP-1 combinations as well as basal/SGLT-2 combinations.
• Though Victoza’s sales tallies in 2Q15 were encouraging, new competitors are making inroads on its market share as expected. According to a chart in the presentation slides, Victoza held a 59% share of the US GLP-1 agonist market (based on total prescriptions) as of June, compared to 28% collectively for AZ’s exenatide franchise (weekly Bydureon and twice daily Byetta) and 6% for both Lilly’s Trulicity (dulaglutide) and GSK’s Tanzeum (albiglutide). The new entrants Trulicity and Tanzeum appear to have taken comparable bites out of Victoza’s and the exenatide franchise’s shares. Regardless, management characterized Victoza’s market share loss as “modest.” Overall, the trends remain consistent with our thinking that the new launches would serve to both expand the class and draw some share from Victoza. However, we imagine that future launches of Xultophy (insulin degludec/liraglutide) and the once-weekly semaglutide should help Novo Nordisk reclaim lost ground.  

Table 4: GLP-1 Agonist Market Share

4. Sales of the Tresiba (insulin degludec) franchise reached DKK 330 million (~$49 million) in 2Q15, up from DKK 271 million (~$41 million) in 1Q15. These figures include sales for Ryzodeg (insulin degludec/insulin aspart) and Xultophy in addition to standalone Tresiba. This represents a 22% sequential climb, though the magnitude of sales remains constrained by low reimbursement levels in some countries. Management noted that the Tresiba franchise accounted for 10% of overall growth in the first half of the year. Tresiba has now been launched in 30 countries and achieved 30% market share in Japan, and penetration remains strong in other countries where it is reimbursed at a similar level to Sanofi’s Lantus (insulin glargine). As a reminder, Novo Nordisk recently announced its decision to withdraw Tresiba from the German market (Europe’s largest) following a series of negative pricing decisions. Though not unprecedented in diabetes, it is disappointing both from patient and sales perspectives given that Tresiba had achieved a not-insignificant 8% value share of the basal insulin analog market in Germany.

• During Q&A, management offered some hints regarding Tresiba’s expected positioning on US formularies for 2016. Management stressed that the company cannot enter into negotiations with payers until Tresiba is approved and that Tresiba’s label will determine how strong the company is in negotiations. However, the company expressed some hesitation regarding Sanofi’s strategy of compromising on pricing for Lantus in order to gain access for Toujeo, commenting “they seemed to be in a hurry and got good access, but also at a price as we can see from the numbers.” Management further expressed confidence that Novo Nordisk will remain a major player in the diabetes field over the next 20 years and that it is in the company’s best interest to “retain value” in the marketplace. Despite the impact on Lantus sales, we do think that Sanofi ultimately made the right move in prioritizing access over price for Lantus, especially given that good access for Lantus builds a foundation for Toujeo.  However, we imagine that some of the reimbursement challenges for Tresiba in Europe resulted from Novo Nordisk’s decision to price the product at a significant premium that it believes is consistent with its value, and we expect that the company could encounter some friction with US payers as well if it takes a similar approach here. Management noted that the company’s long-term growth outlook does depend on Tresiba approval, followed by Xultophy approval, in the US. As a reminder, Novo Nordisk resubmitted Tresiba and Ryzodeg (insulin degludec/insulin aspart) for FDA approval in April, with a decision expected in October.
• Very surprisingly, IQWiG very recently handed down a “no additional benefit” ruling for the transformative basal insulin/GLP-1 agonist combination Xultophy (insulin degludec/liraglutide). The decision is pretty astounding given the clear benefits across the board on glucose, weight, and hypoglycemia with Xultophy. Management frankly acknowledged that Xultophy could meet a similar fate (withdrawal) in Germany as Tresiba if the final G-BA ruling matches IQWiG’s assessment. To be honest, out of all the IQWiG decisions, this one leaves us the most incredulous. Xultophy has now been launched in Switzerland, Germany, and the UK. Management stated that market penetration of Xultophy in Switzerland measured in value of the long-acting insulin market is comparable to that of Tresiba.
• Ryzodeg (insulin degludec / insulin aspart), which we expect to have the greatest success in emerging markets, has now been launched in Mexico, India, and Bangladesh. Management noted that feedback from those countries has been positive so far.

5. Saxenda continues to make measured progress, with recent coverage wins at CVS and ESI. Securing coverage is especially impressive given Saxenda’s higher list price of ~$1,068/month. These coverage wins come in the nick of time, as a chart showing TRx data showed that Saxenda is tracking well behind the post-launch trajectories of Orexigen/Takeda’s Contrave (naltrexone/bupropion) and Eisai/Arena’s Belviq (lorcaserin) – Saxenda’s script volume of 1,420 at week 15 compares with 6,929 for Contrave, 3,882 for Belviq, and 1,204 for Vivus’ Qsymia (phentermine/topiramate), though the latter had a stronger initial burst followed by a period of slowdown. However, when examining script share statistics it is important to keep in mind that Saxenda has a higher price than its competitors. Management indicated that the early feedback from providers and patients has been positive – based on past guidance, to maximize the benefit/cost profile for the product it is targeting it to a subset of particularly high risk patients within the broader population of patients with obesity.

• Expanding formulary access is the major priority for Novo Nordisk at this point, and the company had some positive news to share on this front: Saxenda has achieved coverage with CVS and ESI (Express Scripts) for the remainder of 2015 and will be on the lowest branded copay tier on CVS’ national formulary in 2016. However, management acknowledged that even these gains are unlikely to have a measurable impact on sales given how immature the obesity pharmacotherapy market remains. To that end, we’ve seen Novo Nordisk make major pushes at recent scientific meetings to change the way providers think about obesity, including a non-branded initiative.
• Management noted that Saxenda has obtained 3% volume market share in total anti-obesity prescriptions. Management was positive about the initial uptake, calling it “encouraging,” and emphasized that the anti-obesity medication market in the US is challenging in general, due to the lack of Medicare Part D coverage. Novo Nordisk views the CVS coverage win as a means to level the playing field for Saxenda vs. other anti-obesity medications on the market, but does not believe that the deal on its own will have a huge impact on Saxenda sales. Management emphasized that the company will have to make progress on multiple developments in coming years in the face of a very “immature” market. We agree that anti-obesity drugs face an uphill battle in the US – recent quarters have certainly been challenging for obesity companies, though Orexigen did report fairly encouraging results for Contrave (naltrexone/bupropion; partnered with Takeda) in its recent 2Q15 update.
• We noted that Novo Nordisk did not break out Saxenda sales numbers. Saxenda sales were grouped with NovoNorm and needles under “other diabetes and obesity care.” As a whole, the group grew 4% as reported to DKK 1.1 billion (~$159 million). From the script numbers included in the presentation we are not surprised that Saxenda is not yet meaningfully contributing to whole-company sales.

Top Five Pipeline Highlights

1. Novo Nordisk announced topline results from the ADJUNCT TWO study of liraglutide in type 1 diabetes. The trial (ClinicalTrials.gov Identifier: NCT02098395) evaluated three doses of liraglutide (0.6 mg, 1.2 mg, and 1.8 mg) vs. placebo as an add-on to insulin in 835 patients with type 1 diabetes. The results were positive but fairly modest: a 0.2-0.3% A1c reduction in the context of a lower insulin dose and 2-5 kg weight loss (baseline 84 kg) with liraglutide vs. no change with placebo after 26 weeks (baseline A1c not disclosed). A greater percentage of the liraglutide groups achieved an A1c <7%, though the exact numbers were not disclosed. Disappointingly, there was no difference in severe hypoglycemia or nocturnal symptomatic hypoglycemia for any of the doses; there was actually a higher rate of symptomatic hypoglycemia in the liraglutide 1.2 mg group compared to placebo. There was no mention of glycemic variability during the call, but we later learned from the company that some participants did wear CGM and that there were likely some measures of variability or time in zone – we hope to learn more on this front when the full results are presented. Results from ADJUNCT ONE, a similarly designed but more robust (at least according to management) trial (1,398 patients followed for one year), are expected within three months. During Q&A, Chief Scientific Officer Dr. Mads Thomsen indicated that this trial will function as a pivotal study that would be gating for a regulatory submission in 1H16.

• At this year’s ADA we saw other results on liraglutide in type 1 diabetes that also didn’t knock our socks off. The LIRA-1 trial run by the Steno Diabetes Center found no significant difference in A1c after 26 weeks with liraglutide in type 1 diabetes patients taking insulin vs. the insulin-only comparator group, with a substantial 6 kg (~14 lbs) weight benefit, less insulin, and a non-significant trend towards less hypoglycemia. Though we are very eagerly awaiting the results from ADJUNCT TWO, at this point our expectations are tempered somewhat. It may be that SGLT inhibitors are the more efficacious type 2 diabetes drug class for type 1 diabetes due to their insulin-independent mechanism, though we’d ideally like to see both options in patients’ and providers’ toolbox since some patients cannot take or tolerate SGLT inhibitors and GLP-1 agonists are great when it comes to weight loss.
• Dr. Thomsen suggested during Q&A that other GLP-1 agonists are unlikely to equal liraglutide’s effects in patients with type 1 diabetes. He argued that shorter-acting agents like Sanofi’s Lyxumia (lixisenatide) and AZ’s Byetta (twice daily exenatide) would not be ideal as they would interfere with the action of prandial insulin. On the flip side, we imagine that they could also help enhance the effect of mealtime insulin and allow for lower doses, though titration could certainly become complicated if the effects are not consistent. Dr. Thomsen also suggested that the inferior weight loss typically seen with other long-acting GLP-1 agonists compared to liraglutide would make them less appealing options. We assume the ADJUNCT ONE results will be key in determining whether other companies pursue type 1 diabetes indications for their GLP-1 agonists – if they are similarly modest, we imagine that many may conclude it is not worth the investment, though we do think that the weight benefits are very meaningful. 
  
  • There are several ongoing trials of other GLP-1 agonists in type 1 diabetes in addition to the liraglutide studies. Yeshiva University is conducting a trial (n=20) evaluating the effect of Byetta on postprandial hyperglycemia in new onset type 1 diabetes and Yale University is conducting a study (n=120) evaluating Bydureon in patients with more established type 1 diabetes. Both are currently recruiting participants; the Byetta trial is expected to complete in December 2015 and the Bydureon trial in September 2016. GSK is also currently recruiting participants for a trial (n=68) investigating Tanzeum in type 1 diabetes that is expected to complete in December 2016.  
• Management suggested that Xultophy could also be a candidate for a type 1 diabetes indication but that the rationale is “less obvious” than in type 2 diabetes. As the main benefit would be fewer injections compared to adding a standalone GLP-1 agonist, we imagine that Novo Nordisk would only pursue such an indication if Victoza gains a significant foothold in this population. 

2. The timeline for the full completion of DEVOTE (the CVOT for Tresiba) has been moved up yet again, from 2H16 to mid-2016. This timeline has been accelerated several times due to higher than expected event rates. As a reminder, Novo Nordisk resubmitted Tresiba to the FDA in March based on interim data from DEVOTE made available only to a small firewalled group within the company. The company still expects a decision on Tresiba by October 1, and we assume that a Xultophy submission could occur shortly thereafter (boy, Novo Nordisk’s regulatory affairs team must be incredibly busy right now). We have assumed that the FDA will not require a second Advisory Committee meeting for Tresiba as that would require unblinding the interim data from DEVOTE; however, Novo Nordisk suggested in 1Q15 that the unblinded team would be prepared to hold some sort of AdComm behind closed doors if the FDA requested it.

• Novo Nordisk fielded a question during Q&A about whether the FDA might want to wait for the final completion of the trial before approving Tresiba. While acknowledging that speculating on potential FDA actions is always a bit hazardous (as the company knows all too well), management stated that Novo Nordisk does not expect this to be the case. In the worst-case scenario, our estimate is that approval would likely be delayed by about a year given the time required for submission and review of the DEVOTE data. However, we agree that this is very unlikely given the extensive efforts by Novo Nordisk and the FDA to ensure a smooth road to approval based on interim data.
• The broader issue of interim data disclosure from CVOTs remains unresolved a year after a dedicated FDA hearing on the subject. The need for clearer FDA guidance on how interim data should be used in the regulatory process was the main point of agreement that emerged from that meeting, but we have not seen any sign that a standard set of guidelines is forthcoming. In the absence of such standardized guidance, Novo Nordisk’s approach with Tresiba could potentially set a precedent for other companies that wish to submit products based on interim data, assuming everything continues to go smoothly. However, we do not believe this represents anything close to an ideal solution, as it involves a great deal of uncertainty on the part of manufacturers. If the Tresiba resubmission represents the greatest success story concerning interim CVOT data over the past year, then Orexigen’s experience with Contrave (naltrexone/bupropion) represents the greatest cautionary tale: the company is just now rebounding from the fallout over the termination of the Light Study following interim data disclosure in March.

3. Dr. Mads Thomsen expressed cautious optimism that the ongoing LEADER CVOT for Victoza could plausibly demonstrate CV superiority. During Q&A, Dr. Thomsen noted that LEADER’s greater individual patient exposure (a mandated minimum exposure of 3.5 years per patient and total exposure of over 30,000 patient-years) relative to other CVOTs makes it more likely to be able to demonstrate a benefit. He noted that while the total number of events is likely to be comparable to that seen in TECOS (the CVOT for Merck’s Januvia [sitagliptin], which recently reported neutral results), LEADER involves fewer patients with longer mean exposure time, which should help reveal any effect on CV outcomes. However, Dr. Thomsen was careful to stress that the trial is powered for non-inferiority, not superiority, and our sense is that even this longer duration may not be sufficient to reveal a subtle benefit, particularly in a high-risk population. That said, we agree that this trial is one of the most likely of the existing CVOTs to show superiority and are waiting with bated breath for the results, which are expected in 1H16.

• Dr. Thomsen suggested in 1Q15 that it would also be plausible for Victoza or another GLP-1 agonist to have a beneficial effect on heart failure. In response to a question about the potential impact of the DPP-4 inhibitor/heart failure controversy on the GLP-1 agonist class, Dr. Thomsen said there would be no reason to expect GLP-1 agonists to have any negative impact on heart failure and noted that some human data suggests the potential for benefit. As with the primary endpoint, we continue to assume that neutrality will be the most likely result, but any finding of superiority would of course be a huge win.  

4. Novo Nordisk initiated a phase 2a trial of its primary oral insulin candidate OI338GT (NN1953) and a phase 1 trial of a new oral insulin OI320GT (NN1957) last month. The phase 2 trial (ClinicalTrials.gov Identifier: NCT02470039) is a randomized, double-blind, multiple dose study evaluating OI338GT vs. Lantus (both in addition to placebo) in 50 patients with type 2 diabetes. The primary endpoint is change in fasting plasma glucose after eight weeks; secondary endpoints include change in ten-point plasma glucose profile, serum insulin concentration area under the curve, and number of treatment-emergent hypoglycemic episodes. The phase 1 trial (ClinicalTrials.gov Identifier: NCT02479022) is a single escalating dose study (7 doses from 300-21,600 nmol) evaluating OI320GT vs. placebo (double-blind) and vs. Lantus (open-label) in 84 healthy volunteers. The study is primarily intended to assess safety, with several secondary endpoints related to pharmacokinetics (PK) and pharmacodynamics (PD). Both trials are expected to complete in January 2016. We are glad to see OI338GT progressing into phase 2, as we had not heard updates on its status since Novo Nordisk’s 1Q14 update. We are curious to learn more about the differences between the two candidates and whether the company plans to advance both all the way through the pipeline or whether the new formulation could eventually supersede the more advanced candidate. Oral insulin is one of the more challenging areas in diabetes drug development due to variable absorption from the gut and insulin’s narrow therapeutic window. However, Novo Nordisk certainly has the edge in terms of experience and resources. Other companies with oral insulins in clinical development include Biocon and Oramed, which both have candidates in phase 2.

• During Q&A, Novo Nordisk promised an update on the phase 3 go/no-go decision for oral semaglutide “as soon as possible.” Management confirmed in 1Q15 that a decision will take place later this year after consultations with regulatory authorities and a commercial assessment. In this update, management noted that the company has been in a dialogue with regulatory agencies and is considering a range of factors, including investments and the design of a phase 3 trial, before making a final announcement. As a reminder, Novo Nordisk announced positive topline phase 2 results for oral semaglutide in February showing dose-dependent A1c reductions of 0.7%-1.9% (baseline = 7.9%) with once-daily doses ranging from 2.5 mg to 40 mg, compared to a 1.9% reduction with 1 mg injectable semaglutide (once weekly) and 0.3% with placebo. Both injectable and higher oral semaglutide doses also produced significant ~5.5 kg (~12 lb) placebo-adjusted weight loss. We assume that much of the discussion at this point is focused on the best dose to advance into phase 3. While the A1c reductions and weight loss with the 40 mg dose were impressive, Novo Nordisk suggested in 1Q15 that this highest dose will likely not be carried forward, presumably due to GI side effects. Bioavailability is likely also an important consideration, as oral semaglutide required a ~280x dose to match the efficacy of injectable semaglutide, which could create challenges in terms of pricing.
• Novo Nordisk has two oral formulations of liraglutide in phase 1. The two candidates (OG987GT and OG987SC) each use a different carrier technology to protect the liraglutide molecule in the harsh environment of the digestive tract. We have not heard any recent updates on if or when these molecules might move into phase 2. Given that semaglutide appears to beat liraglutide in terms of efficacy based on clinical trials of their injectable formulations, it would be understandable if the company is focusing the majority of its oral GLP-1 efforts on semaglutide at this point.
• We have not heard updates on Novo Nordisk’s long-acting insulins LAI287 and LAI338 in the past few quarters; both remain in phase 1 according to the company’s pipeline. Phase 1 trials of both LAI287 and LAI338 have been completed according to ClinicalTrials.gov, and we hope to see results in the near future.

5. Novo Nordisk highlighted the positive results from the SUSTAIN 1 phase 3a trial of its once-weekly GLP-1 agonist semaglutide. As a reminder, the company announced positive topline results from this trial last month demonstrating significantly greater A1c reductions with both 0.5 mg and 1.0 mg doses of semaglutide (1.5% and 1.6%, respectively) vs. placebo. In addition, significantly more patients achieved an A1c <7% with semaglutide (74% of the 0.5 mg group and 73% of the 1.0 mg group) than placebo. Both doses of semaglutide also led to superior weight loss (3.8 kg with the 0.5 mg dose and 4.6 kg with the 1.0 mg dose) compared to placebo (1.0 kg weight loss) and had a positive safety/tolerability profile. These results are certainly encouraging, if not particularly surprising, and we believe that this formulation offers significant potential for Novo Nordisk to maintain its leadership in the GLP-1 agonist class going forward. SUSTAIN 1 is the first of six phase 3 trials for semaglutide (see the table below for a complete overview), and the company confirmed that results from the remaining trials are expected within the next nine months.

Table: SUSTAIN phase 3 trial program for semaglutide

• Management suggested in 1Q15 that phase 2 dose-ranging trials of semaglutide in obesity could begin around the time of an approval in diabetes. Dr. Thomsen noted in 3Q14 that brain scan findings showing greater accumulation of drug in the arcuate nucleus, a center of appetite regulation, support the possibility of even greater weight loss with semaglutide compared to liraglutide. If true, this could also place semaglutide ahead of most other existing obesity drugs in terms of efficacy, though parameters like side effects and the responder/non-responder split would also be important. Novo Nordisk also suggested in 1Q15 that any plans for semaglutide in obesity would be independent of Saxenda’s commercial success; this is consistent with previous statements from the company emphasizing an overall commitment to obesity rather than an exclusive focus on Saxenda.

Honorable Mention

• There was no mention of the other diabetes and obesity candidates in Novo Nordisk’s pipeline, which include faster-acting insulin aspart (Faster aspart) in phase 3, two phase 1 long-acting insulins, and two phase 1 obesity candidates. Novo Nordisk announced topline phase 3a results for Faster aspart in March demonstrating non-inferiority vs. NovoLog, and management confirmed in 1Q15 that the company plans to submit the product in the US and EU around the turn of the year. Overall, we see the improvements in postprandial glucose and potential for flexible dosing with Faster aspart as meaningful clinical benefits but do not consider it to be more than an incremental improvement over NovoLog – for that we’ll have to wait for future candidates or novel means of administration such as with Sanofi/MannKind’s inhalable Afrezza. A phase 1 trial of the company’s glucagon analog G30L (NN9030) in 160 overweight/obese males is currently recruiting and expected to complete in November 2015. Novo Nordisk ultimately intends to co-formulate the candidate with liraglutide. A phase 1 trial of the company’s long-acting amylin analog NN9838 in 56 overweight/obese males is also recruiting and expected to complete this month. Both trials are primarily intended to evaluate safety and have secondary endpoints related to pharmacokinetics.
• Below, we summarize  year-over-year and sequential growth for insulin analogs and GLP-1 agonists in 2Q15.
  
  • Aggregate worldwide revenue for basal insulin reached ~$2.6 billion in 2Q15, falling 7% YOY and rising nearly 7% sequentially. These sales include Sanofi’s Lantus and Toujeo (together $1.9 billion) and Novo Nordisk’s Levemir and Tresiba (together $700 million). This is slightly lower than 4Q14’s record aggregate sales of ~$2.9 billion. We suspect that the YOY decline stems from increased payer pressure on the category, particularly in the US. The 7% sequential growth appears to be largely driven by 1Q15’s unusually low basal insulin sales. As we had forecast at the end of 2014, growth in worldwide basal insulin sales has slowed, likely due to increased competition, delayed introduction of insulin due to better oral treatments, and pricing pressure. With biosimilar insulins on the horizon, the pricing pressure will likely only increase moving forward. That said, from 2011 up until the start of 2015, the class has accounted for over 50% of insulin market growth since 2011 (excluding 3Q11), and still reached $11.1 billion in 2014, up from $9.6 billion in 2013.
    • Sanofi maintains its lead within the segment with Lantus, which still owns 72% of the worldwide market. However, this is in the context of declining sales following Sanofi’s acceptance of higher rebates in exchange for continued broad access in 3Q14 (Sanofi owned ~76% of the market a year ago). Meanwhile, Novo Nordisk had its strongest quarter yet in 2Q15 in terms of basal insulin sales, with Levemir’s global market share in the basal insulin market slightly up at ~26%. Regarding the newest basal insulin products, market shares for Novo Nordisk’s Tresiba and Sanofi’s Toujeo were 1.9% and 0.5%, respectively in 2Q15 – both should be significantly higher a few years from now.
    • The recently published Express Scripts 2014 Drug Trend Report indicates that payers and consumers are facing increased cost pressure with regard to insulin, despite the slowing of the overall basal insulin market. The report noted that Lantus unit cost increased by 34%, 32%, and 33% in the commercially-insured, Medicare, and Medicaid sectors, respectively. Levemir unit cost increased 48% in the Medicare sector. Express Scripts largely attributed the overall increase in diabetes spend in 2014 (the fourth year in a row that diabetes spend has gone up) to rising insulin costs. Thus, despite increased pricing pressure and stress on drug makers, payers and consumers are not seeing lower costs on the other end..
  • Rapid-acting insulin revenue globally totaled ~$1.5 billion in 2Q15, falling 3% YOY and rising 2% sequentially. This includes Lilly’s Humalog ($654 million), Novo Nordisk’s NovoLog/NovoRapid ($775 million), Sanofi’s Apidra ($103 million) and Sanofi/MannKind’s Afrezza ($2 million). Growth has stagnated in this segment, having fallen from its peak with worldwide rapid-acting insulin revenue staying flat at ~$1.7 billion from 4Q13 to 4Q14. Rapid-acting insulin has especially slowed in international markets, as revenue outside of North America fell 8% YOY and rose 2% sequentially. Lilly’s Humalog and Novo Nordisk’s NovoRapid continue to dominate the market with 46% and 51% of the market share, respectively. Market share for Sanofi’s Apidra has stayed at ~7% and Sanofi/MannKind’s newly introduced Afrezza had a small 0.1% market share in 2Q15. We believe that increasing uptake of safer drug classes addressing postprandial glucose such as GLP-1 agonists and SGLT-2 inhibitors may have partly contributed to the segment’s declining sales. As with basal insulin, the decline in revenue has occurred alongside increasing frustration among patients and providers about rising prices, making the current equilibrium a bit of a lose-lose situation. 
  • GLP-1 agonists continued to be a market to watch in 2Q15, with an aggregate worldwide revenue of $956 million, growing 24% YOY and 13% sequentially (from 1Q15’s $846 million). This class includes Novo Nordisk’s Victoza ($665 million), AZ’s Bydureon and Byetta (together $222 million), Lilly’s Trulicity ($44 million), GSK’s Tanzeum ($14 million), and Sanofi’s Lyxumia ($11 million). We are curious to see whether this class will be increasingly positioned as an alternative or an add-on to insulin (or whether the answer may vary by product). As we saw at ADA, GLP-1 agonist/basal insulin combinations were a hot topic this year, and the future for this class appears bright. In 2014, global GLP-1 agonist sales totaled over $3 billion, and we expect sales to continue to grow this year. Last year also featured the addition of three once-weekly GLP-1 agonists (new Bydureon pen, Trulicity, and Tanzeum/Eperzan), offering more options and additional convenience. Meanwhile, innovative delivery devices are likely to contribute to continued growth of this market. We predict that Lilly’s Trulicity, which has an easy-to-use injector pen and demonstrated non-inferiority vs. Novo Nordisk’s Victoza in the AWARD 6 trial, has especially significant potential to expand this class. Indeed, the drug appears to have driven a good portion of growth in 2Q15 during its second full quarter on the market, pulling in $44 million compared to $18 million in 1Q15. While we have been interested as to whether these newer additions would gain greater market shares or expand the class as a whole, the market seems to be leaning towards the latter with a sequential increase in world sales for five out of the six GLP-1 agonists – all but AZ’s Byetta.
    • In the medium term, the GLP-1 agonist market could be significantly disrupted by a successful oral formulation and/or Intarcia’s ITCA 650. As a reminder, Novo Nordisk just announced its decision to advance an oral formulation of its long-acting GLP-1 agonist semaglutide into phase 3 trials. For its part, Intarcia released positive topline results last week from the phase 3 FREEDOM-2 trial of ITCA 650 (implantable exenatide mini-pump) vs. Merck’s Januvia (sitagliptin). Both companies seem to be framing their products as second-line therapies after metformin, and both will likely have significant implications for existing GLP-1 agonists if successful.

Questions and Answers

Q: You have been quoted discussing what you saw in terms of pricing in 1H15 and what you expect for the rest of the year and into 2016. Could you clarify what exactly you were saying in terms of pricing, the price and mix effect, and the time period?

A: If we look entirely at insulins, we have had a slight negative influence on price in the first half of this year. If we look at the total portfolio, we have flat pricing, and that is of course a reflection of the strong performance of Victoza, where we have pricing power because we are the gold standard in that market. When we look at insulins going forward, we are looking at full-year expectations for flat to slightly positive pricing. As we now have entered into contract for the remainder of 2015 and into 2016 without any impacts on the remainder of the year, we know the price picture. We don’t know the channel mix, of course, but we know the price picture. So that will be our guidance for the full year and with the numbers we know today for 2016, it will also indicate to us flat pricing for our insulin portfolio next year. Depending on the extent to which Victoza continues to outperform, there might be a slight positive impact from Victoza if you add the mix.

It is clear on top of that when you look at the gross margin, there is a positive impact on our gross margin…coming from overall higher sales of Victoza and the modern insulin at the expense of older products like human insulin. That’s giving us a slight positive impact on gross margin from selling the higher value products. That trend is also expected to continue into 2H15 and potentially 2016.

Q: Your long-term growth rate expectations have been kind of seven plus three to get to ten. Is there any change to that given the lower pricing component from the US that is contributing to the three and the macro outlook that’s slowing down China, which was a significant contributor to the volume outlook?

A: Our long-term growth outlook does depend on a Tresiba approval and therefore our subsequent ability to get Xultophy approved and launch it in the US. But our long-term aspiration is still holding onto the seven plus three. This is all assuming that what we are seeing currently in China is an aberration and there’s more short-term deterioration of growth. So let’s see when we get to the end of the year whether we have a different perspective on China, but that would still be my aspiration: seven plus three, seven being volume and three being value, and value coming primarily from new products.

Q: On pricing into 2016, have you had any negotiations around Tresiba positioning on formularies ahead of approval? Is there any risk that as you head into those you might have to give up some price on Levemir to get Tresiba lifted as was the case with a key competitor?

A: We cannot go into a dialogue with the PBMs before the product is approved. And we can only speculate whether or not it will have any implications for Levemir. Obviously it had implications for Sanofi when they wanted to get access with Toujeo. They seemed to be in a hurry and got good access, but also at a price as we can see from the numbers. So all of that is deferred until we get an approval and we see the label. That will determine how strong we are in the negotiations and how strong we are in that competitive marketplace, which in the future will be comprised of Levemir, Lantus, Lantus biosimilar, and Toujeo and Tresiba. So it’s going to be a quite complex marketplace. But so far, we are seeing the price level that I indicated before for 2016 and the same access in 2016 as we’ve seen in 2015.

Q: As you think about the speed of access for Tresiba, are you more interested in getting quick access as a competitor seemed to or protecting pricing? Do you see the Tresiba negotiation process as a risk or an opportunity as you head into 2016?

A: If you take the perspective from what we have historically done, we are a player that is going to be in insulin and diabetes therapy 20 years from now, and we have an innovative portfolio. So it is at least in our individual interest to retain value in the marketplace and go for a long-term strategy as opposed to a short-term conversion strategy. Having said that, we will determine that at the time of the launch depending on the competitive situation and the label.

Q: With Saxenda entering the CVS Caremark preferred list, even though it’s with the highest branded copay, is there any way to try to calculate what kind of impact this should have? Do you still see resistance from doctors or patients?

A: The CVS listing is of course a positive in the sense that it’s a prerequisite for the continued development of Saxenda sales in the obesity market, that we see payers that are willing to make agreements. That puts us in a situation where we are largely on par with some of the other branded anti-obesity medications. But with that said, we’re still faced with a market that has a very limited universe of prescribers. So market access is only one of the multiple developments that we need to work on in the coming years. So I wouldn’t be as concrete as saying that specific deal would have a measurable impact. I think we’re still looking at a very immature market, and we’re still looking towards a long and steady development of this market.

Q: At ADA this year when ELIXA and TECOS were presented, it was highlighted that the duration of follow-up was pretty small, at 20 months to 30 months. Now with LEADER, you’ve got 9,000 patients being followed for at least two or three times longer. So given the A1c, weight, and blood pressure profile of Victoza, how are you thinking about the potential for LEADER to actually be a positive outcomes study, or is your upside expectation that it just shows no harm?

A: The LEADER study is, as per regulatory request, designed as a non-inferiority study. That means 611 strict MACEs, double-blinded, 9,300 patients, and so on. The difference from the existing trials is the individual patient exposure. We have now exceeded 30,000 patient-years of exposure and that is because in the trial there is a minimum exposure of 3.5 years per patient. This is because it’s a time- and event-driven trial. If it were only an event-driven trial we could have closed up shop after 611 MACEs, which we passed a long time ago. But the European regulators also had some very reasonable requests for long-term safety that were unrelated to the cardiovascular system, which is why it is event- and time-driven. That then means that we will end up with a total number of MACE events that is reminiscent of what you have in the superiority-designed trials such as TECOS. But the distribution of patients is different in that we have fewer patients but they have more exposure. And that should ideally be beneficial for the discrimination of any difference in CV outcomes.

As you will recall, when we look at VADT, ADVANCE, and in particular UKPDS and DCCT, the curves only start spreading out after several years of exposure, not after two years as we’ve seen in many of these trials. Therefore, it is powered for non-inferiority, but statistically speaking, the next step in the statistical hierarchy is to test for superiority, and with more than 1,000 MACE events, if there is a discernible difference, that becomes an option.

Q: With the final results of the DEVOTE trial getting closer and closer, is there a chance the FDA may want to wait for the final completion of the trial before approving Tresiba?

A: It is difficult to answer because speculating what the FDA might do is quite hazardous, as we’ve seen in the past. The FDA is a science-based agency. That means at the point when they saw the interim results that none of us know about, the agency officers and regulators have looked at that and decided whether to accept the interim submission or not. Then there’s what they see from the time they get those early data to the full submission, including all the analysis that we have co-submitted from the BEGIN program and from the extensions and so on. Until their action after six months, which we expect to be around October 1, they should not be influenced by things outside of the regulatory dossier. It should be influenced by whether they are comfortable with approving it as it stands or not. But to be honest, I cannot really speculate on regulatory bodies ad how they think. On a positive note, we just announced that we would be completing the trial faster in mid-year 2016. So should the worst happen, that’s the timeline you’re looking at. Then of course the data would need to be compiled and added as a different submission again at that point. But this is not our general expectation.

Q: On Xultophy, what is your scenario in Germany given the fact that you withdrew Tresiba a couple of months ago?

A: We are disappointed about the evaluation of Tresiba in Germany and we’re disappointed that we were forced to withdraw this product to the inconvenience of 40,000 people using this product. Could we be in the same situation with Xultophy? Maybe. We don’t know. We have gotten an initial assessment from IQWiG that is not positive. We will, of course, be putting our best foot forward in our compiling of our dossiers and our negotiations and then we shall see. I would say that if we cannot get a positive resolution with Xultophy with the German insurance agencies, then I don’t look very positively on the German market going forward because Xultophy is sine qua non the best type 2 diabetes product available on the market. Of course the question will be about the price. So we shall see, but the same thing might happen. I actually have to admit that. It’s very frustrating for us. It must be very frustrating for the patients and for the medical community because they have been very productive in assisting us in the clinical development of these new drugs. But that’s the situation we face. We have a bit of a similar situation in Denmark and the UK. That’s the nature of our business these days.

Q: On Victoza in type 1 diabetes, at what point can you file based on this data? Do you need both trials?

A: The ADJUNCT pivotal program consists of two trials, 1 and 2. What you heard today is the number 2 trial, and that is a regulatory trial where they mandate us to show and document two priorities. We have to prove statistically significant improved A1c control by capping insulin doses so it’s apples-to-apples and similar insulin dosing in both arms. In fact, we actually had an insulin dose lowering in the liraglutide arms. And we did meet the statistical hierarchy of testing in that trial. But ADJUNCT 1 is the big one. That’s a full-year trial, unlike this one that was a half-year trial. It’s also a trial with 1,400 patients, whereas this one was only with 800 patients. And it is a treat-to-target trial. So that is to be perceived as the pivotal trial and if that data comes out in the way that we hope, that would be gating for submission of the dossier in the first half of next year. We have five dossiers we hope to submit in the US this year and it’s simply not feasible to get the filing done before the first half of next year. But let’s hope we get there.

Q: Will you be starting Xultophy trials to become a new standard of care for type 1 diabetes?

A: Xultophy could obviously be a candidate to make the multiple daily injections fewer. It’s less obvious to use Xultophy in type 1 than in type 2 diabetes because they use two MDI regimens anyhow. But of course it could be an option and we’ll look at that as so many other things.

Q: Can the benefits with Victoza in type 1 diabetes be replicated with other GLP-1 agonists?

A: I would say yes and no. The ones that are short-acting like lixisenatide and exenatide will interfere, in my view, with the prandial bolus element of basal-bolus and not be ideal. So it would have to be a long-acting GLP-1 agonist and they typically command less weight control than does liraglutide. So some of it could be replicated with some of them, but probably not all of it with any of them.

Q: On oral semaglutide, there was some expectation that you might already have completed dialogues now and be able to announce some plans. Could you comment on what kind of dialogue you are in with the authorities? What are the most notable outstanding issues or challenges to be able to give us an update later this year?

A: We have indeed been in a dialogue with the various regulatory agencies as you have to be between the end of phase 2 and the phase 3 go/no-go decision. But as you can imagine, there are numerous things coming into play here, including investments and getting the phase 3 pre-program totally planned and approved, etc. So we will get back to you as soon as possible. Unfortunately I won’t be more specific on that, but it will be as soon as we can.

Q: Could you add a bit of color on the increasing competition in China?

A: The question is specific to the more intense competition there from local producers, and the more competitive bidding around China in the region. So that has an impact. But I need to underline that that’s one of a number of effects that has influenced 2Q15, ranging from our own internal shifts in inventories and trends we’ve observed over longer periods of time, with some cooling off of the healthcare market in China. But we’re specifically talking about the aggressive bidding in the human insulin category in the provincial biddings.

Q: Sanofi is doing pretty well with Lantus in China, at the expense of mix. Could you talk about what you can do short-term around that?

A: It’s correct, we’re sort of getting slapped from two sides in the sense that Sanofi’s doing well with the gold standard so far in the basal market with Lantus. Then we have a local competitor with a generic version of Lantus, which with slightly more aggressive pricing has an impact in the marketplace. And we are getting slapped by the local manufacturers on human insulins because the bidding and use of human insulins are more extensive in the rural areas, and that is where the local manufacturers have more clout and they’re able to outcompete us. So that’s the current situation. And one of the longer-term remedies is getting Tresiba approved in China and supporting the mix market with Ryzodeg.

Q: Longer-term, where are you in terms of the timeline for the Ryzodeg launch in China?

A: Tresiba is involved in China as it is all over the world and we are making inroads into getting towards a filing. We did a Chinese phase 3 study that actually completed more than a year ago, but you need a lot of administrative things to  happen before you can submit. But we are making progress in that regard. And only then can you progress the combination product because the Chinese rules from the CFDA mandate that you first have the mono-components in the system and then you can file a two-component system like Ryzodeg. But we are making progress. It just takes longer than in the Western world. That is just like the US, by the way.

Q: On China, could you quantify what the impact from shipments was?

A: The impact has three more or less equal contributions, between the slowing of the market, the increased competition, and then the shipments.

Q: In China, you’re heavily reliant on the old portfolio. Reimbursement for the newer portfolio is years off as we’ve seen with Victoza, and sales and marketing efforts are being curbed by government interventions. How can you change those dynamics?

A: We still firmly believe in the growth case in China. With NovoMix 30, we’re still seeing more than double-digit growth, and the focus going forward will be to leverage the current portfolio. It is correct that the newer products come with more delay than we’ve seen in the past, and with slower uptake in terms of national reimbursement and so on. So growth will have to come from the current portfolio as we move ahead. The strategy remains the same and we have followed through on an expansion of the field force to go yet again into the smaller community hospitals, where the Chinese government intends to move some of the treatment of diabetes. With 100 million people with diabetes in China, we need to be firmly positioned in all areas of China, in all regions and all the institutions. That’s the strategy we’re pursuing. We just have to get used to the fact that the market will operate with growth rates that are lower than we’ve seen historically, but growth for sure. And it is a market that requires a full portfolio. We need to be available and get access with human insulin because that’s where we build the loyalty. And then people, once they get income, can upgrade and then we can capture the more affluent part of the market. So I think we’ve got the right approach to this and we still believe in the long-term growth, and we will of course continue to reassess this.

Q: In China, beyond the local competition and Sanofi, are there any fundamental changes in this market?

A: It depends on what level you want to look at it. We still believe in the fundamental growth case in China, but we have seen changes, including some we haven’t yet discussed here. Part of the efforts from the central government in China to control costs were, amongst others, to bend institutions’ ability to mark up with 15% and thereby derive an income from selling pharmaceutical products to patients. I wouldn’t call it a fundamental change, but it changes the dynamics in how we interact with institutions. Government has opened up the possibility that after the provincial biddings, there could potentially be a second round of negotiations with the institutions – not a fundamental change but still one that will further increase the competitiveness from a pricing point of view. And there are also some new rules around the whole way that prices are governed. So if you added it all up, we’ve come out of a decade of very high growth in China and I think we’re looking at a market now where it’s most similar to others where you have to work hard for the growth but growth is still going to be there. That fundamental part is not unchanged.

Q: On R&D, you’re running at a very low rate at the moment relative to peers, but you also have a very different longevity of your asset mix. Could you help us understand how R&D may have picked up in the short term and what you think a sustainable level is going forward?

A: It is a bit of an unusual picture for 1H15 because you have two impacting forces. One is the discontinuation of inflammation and on top of that you have a very substantial currency impact that especially hits our R&D line when you look at it in reported numbers, as the predominant part of our R&D cost is within the euro zone, including Denmark. What we really look at when we look at it as an underlying trend is more of a meaningful growth level, and we believe that we can successfully continue to invest more within the R&D area and especially within diabetes. I would presume we would at least see growth levels of around 10% in investments in that area. This will be slightly volatile and of course be very dependent on the level of phase 2 and especially phase 3 compounds we are dealing with.

I also would note that I see a trend towards us investing additional resources within programs like LATIN as a label expansion for Victoza but also more generally in the portfolio of our next-generation insulins and taking us to phase 4 trials enabling better market access. I see more requirements hitting us there, which will have indications for the R&D ratio. If you adjust for the inflammatory discontinuation, the growth in the first half of this year was 6% and that’s not deviating a lot from 10%. I see us being in that vicinity.

All the molecules you see right now in the pipeline are patented by our researchers mostly in Denmark, and that means that R&D to sales ratios sometimes can seem less than competition because we don’t have as many milestone payments and other more expensive elements associated with licensing deals. Not that we will not license, but so many of our fruits will be homegrown, and sometimes that can be cheaper if you’re decent at it.  

-- by Emily Regier, Helen Gao, Manu Venkat, and Kelly Close