Novo Nordisk 3Q15 – Diabetes/obesity portfolio up 9% operationally to ~$3.2 billion, driven by Levemir and Victoza; Continued optimism for cardioprotection in LEADER – November 3, 2015

Novo Nordisk provided its 3Q15 update this morning in a call led by CEO Mr. Lars Sørensen. The company’s diabetes and obesity portfolio posted sales of DKK 21.6 billion (~$3.2 billion) in 3Q15, up 9% year over year (YOY) in constant currencies and 22% as reported. As in past quarters, growth was driven by Levemir (insulin detemir) and Victoza (liraglutide). Levemir grew 10% YOY in constant currencies (25% as reported) to DKK 4.7 billion (~$695 million) and Victoza grew 20% YOY in constant currencies (36% as reported) to DKK 4.7 billion (~$697 million). Levemir continued to gain market share in the US but will likely face a more challenging year in 2016 due to the loss of a significant formulary contract and a shift in focus to the launch of Tresiba (insulin degludec). Victoza has lost some market share throughout the year to new entrants in the class but appears to have stabilized at 59% share at least in the last few months. New-generation insulins (Tresiba, Xultophy, and Ryzodeg) posted DKK 376 million (~$56 million) in sales, up 14% sequentially. Management emphasized the high clinical value for Tresiba and Xultophy and reiterated that the company does not intend to sacrifice price for immediate broad access for either product.

On the R&D front, CSO Dr. Mads Thomsen reiterated his cautious optimism that the LEADER CVOT for Victoza may be able to demonstrate a cardiovascular benefit, though the company stressed that the trial is only powered for non-inferiority. The company also confirmed plans to submit its faster-acting insulin aspart formulation (Faster aspart) by the end of the year and highlighted two new phase 2 trials of once-daily semaglutide in type 2 diabetes and obesity. Earlier in the pipeline, Novo Nordisk has completed the last phase 1 trial for once-weekly basal insulin LAI287 and plans to further investigate side effects before initiating additional trials. Excitingly, the company has advanced a formulation of the gut hormone PYY (NN9747) into phase 1 for obesity – further evidence of Novo Nordisk’s broad commitment to early-stage obesity research. Read on below for an in-depth discussion of these and other highlights from the call, followed by Q&A.

Table 1: Financial Results for Novo Nordisk’s Major Diabetes Products

Financial Highlights

• Novo Nordisk’s diabetes and obesity portfolio posted sales of DKK 21.6 billion (~$3.2 billion) in 3Q15, up 9% year over year (YOY) in constant currencies and 22% as reported. The portfolio grew 1% sequentially as reported. The operational results are consistent with the growth rates in 1Q15 and 2Q15 and with the company’s 4Q14 forecast of 6%-9% growth in 2015.

Figure 1: Total Diabetes/Obesity Sales (1Q12-3Q15)

• By region, North America remains the main contributor to growth, aided by growth in the GLP-1 agonist market and volume growth and market share gains for Levemir (insulin detemir). The region accounted for 56% of the company’s growth in the first nine months of 2015. International Operations has also seen fairly strong growth throughout the year, driven by human insulin with some contribution from the rollout of modern insulins and Tresiba (insulin degludec). That segment accounted for 27% of growth in the first nine months of the year. Growth in Europe has been somewhat weaker, boosted by the performance of Victoza (liraglutide) and Tresiba but hurt by declining human and premixed insulin sales and pricing reforms. This could foreshadow similar challenges in the US in the coming years if the current public frustration with rising pharmaceutical prices translates into policy changes. Novo Nordisk continues to face challenges in China due to competition, a decline in the diabetes market, and cost-containing measures. Japan and Korea experienced moderate growth, driven by Victoza and strong uptake of Tresiba but offset by a decline in insulin volumes.
• By product, modern insulins and Victoza remain the main drivers of growth, collectively accounting for 68% of the portfolio’s growth in the first nine months of 2015. The insulin degludec portfolio (Tresiba, Xultophy, and Ryzodeg) accounted for 10% of growth in the same time period.

Table 2: Whole-Company Sales and Growth by Geography (First nine months of 2015)

Insulins

• Novo Nordisk’s modern insulin portfolio (NovoLog/NovoRapid, NovoMix, and Levemir) collectively grew 5% YOY in constant currencies (18% as reported) to DKK 12.5 billion (~$1.9 billion) in 3Q15. Sequentially, modern insulins declined 1%.

Figure 2: Modern Insulin Sales (1Q12-3Q15)

• Novo Nordisk’s share of the global insulin market held steady at 47% as of August 2015. The company’s share of the modern and new-generation insulin market rose slightly to 46% from 45% in August 2014. Modern and new-generation insulins now account for 82% of Novo Nordisk’s insulin sales.

Table 3: Novo Nordisk Insulin Market Share

• Growth for the insulin portfolio was driven by North America and International Operations. Growth in North America was driven by continued market share gains for Levemir (insulin detemir) and NovoLog (insulin aspart) and underlying volume growth, partly offset by a decline in premix insulin. The growth for International Operations was driven by human insulin, NovoRapid (insulin aspart), NovoMix (biphasic insulin aspart/insulin detemir), and Tresiba (insulin degludec). In Europe, growth was driven by Tresiba and NovoRapid with some contribution from Xultophy (insulin degludec/liraglutide). It was offset by declines in premix and human insulins and the implementation of pricing reforms in several countries. Growth in China was helped by continued penetration of modern insulins but offset by a decline in the overall diabetes market due to cost-containment measures and increased local competition. Sales in Japan were helped by continued strong uptake of Tresiba and partly offset by declining insulin volume overall.
• Pooled worldwide basal insulin sales were ~$2.5 billion in 3Q15, down 9% YOY and 3% sequentially. We imagine the decline stems primarily from increased US payer pressure on the category, particularly for Lantus, which holds 68% market share by revenue and 66% by volume as noted above. This pressure is likely only to increase with biosimilars on the horizon. Pooled sales for rapid-acting insulin totaled ~$1.5 billion this quarter, down 6% YOY and up 3% sequentially. In Lilly’s 3Q15 update, management commented on challenges facing that market due in part to the impact of GLP-1 agonist and SGLT-2 inhibitor popularity. Price increases in the US have likely buoyed the segment to some extent, but we do not see those as very sustainable given the current outcry over rising insulin costs.
• Novo Nordisk offered relatively optimistic commentary on the future of the insulin market though did acknowledged a high degree of uncertainty, which was not surprising. During Q&A, management noted that the company expects continued volume growth in the insulin market and suggested that Novo Nordisk should continue to gain share. However, management also acknowledged that past expectations for price increases may need to be adjusted due to competitive pressure and pricing reforms. The company also predicted a “tremendous transition” in the US basal insulin market in 2016 with the arrival of Tresiba, Lilly/BI’s Basaglar (biosimilar insulin glargine), and Sanofi’s LixiLan (lixisenatide/insulin glargine). While all insulin manufacturers are clearly facing challenges and uncertainty going into 2016, the overall tone of this commentary was much more positive compared to Sanofi’s fairly gloomy predictions on its future in the US insulin market. We certainly believe far more patients would benefit from being on insulin and would love the field broadly to focus on how to help patients get to goal using therapy and education appropriately.

Levemir

• Levemir (insulin detemir) sales were up 10% YOY in constant currencies (25% as reported) to DKK 4.7 billion (~$695 million).

Figure 3: Levemir Sales (1Q12-3Q15)

• Levemir’s strong performance in North America was the main driver of its success. The product grew 14% YOY in constant currencies (36% as reported) to DKK 3.5 billion (~$515 million). Novo Nordisk shared IMS data showing continued market share gains for Levemir in the US despite the introduction of Sanofi’s Toujeo (insulin glargine U300). Levemir held a 24% share of the US basal insulin market by volume as of August 2015, compared to 66% for Sanofi’s Lantus (insulin glargine), 10% for NPH insulin, and 0.5% for Toujeo. Novo Nordisk did not share new-to-brand prescription (NBRx) share numbers for Levemir as it did last quarter, and one analyst noted during Q&A that Sanofi had made the greater gains for that metric. According to Sanofi’s 3Q15 update, NBRx share was 14% for Toujeo (up from under 10% in 2Q15) and 26.8% for Levemir (down from ~30% in 2Q15) as of October 9, 2015.
• Ex-US sales for Levemir were weaker, though the product did see double-digit growth in China from a small base. Sales in China rose 11% in constant currencies (29% as reported) to DKK 107 million (~$16 million). Sales declined 2% in constant currencies (flat as reported) in Europe to DKK 733 million (~$109 million) and rose 7% in constant currencies (2% as reported) in International Operations to DKK 319 million (~$48 million). Levemir declined 10% in constant currencies (8% as reported) in Japan/Korea to DKK 47 million (~$7 million), likely due to the strong uptake of Tresiba in Japan.
• Levemir has performed well throughout 2015 but will likely face a more challenging year in 2016. Novo Nordisk acknowledged during Q&A that it opted out of a significant 2016 US formulary contract for Levemir rather than accept aggressive price discounts (the opposite of Sanofi’s approach with Lantus). Management also admitted that next year’s US launch of Tresiba will likely cannibalize some sales from Levemir, though the company believes most switches will come from Lantus. On the other hand, Novo Nordisk does not believe the launch of Lilly/BI’s Basaglar (biosimilar insulin glargine) will have a significant direct impact on Levemir. We agree that the biosimilar will likely draw patients primarily from Lantus and wonder if it could have an indirect effect of driving prices down for the entire basal insulin market.

Tresiba

• Sales of new-generation insulins reached DKK 376 million (~$56 million) in 3Q15, up 14% sequentially from DKK 330 million (~$49 million) in 2Q15. These figures include sales for Tresiba (insulin degludec), Ryzodeg (insulin degludec/insulin aspart) and Xultophy.

Figure 4: New-Generation Insulin Sales (1Q14-3Q15)

• Tresiba has now been launched in 36 countries, and penetration remains heavily dependent on the level of reimbursement. Penetration is highest in Japan at 31% of the basal insulin segment by value. It is lowest in the UK and Denmark at 3%. Novo Nordisk’s decision to withdraw the product from Germany (Europe’s largest market) last quarter following a series of negative pricing decisions is likely also placing some limits on sales. In Sanofi’s 3Q15 update, management highlighted the steeper early launch trajectory in Germany for Toujeo vs. Tresiba. We see this as a bit of an unfair comparison, as Toujeo did not have to go through the IQWiG review process because it contains the same active ingredient as Lantus.  
• Management offered fairly extensive commentary on the company’s plans for the US launch of Tresiba (approved by the FDA last month), which will begin in 1Q16. This launch will clearly be the highest commercial priority for Novo Nordisk next year, a reflection of the company’s belief that Tresiba is a highly innovative product with enormous clinical value. The current plan is to price Tresiba at a 10% premium over Levemir in the US, though management left the door open for adjustments if the pricing environment is tougher than expected (which may well be the case given recent trends). As in the 2Q15 update, management emphasized that Novo Nordisk does not plan to sacrifice value to gain immediate broad access for Tresiba, meaning initial uptake will likely be gradual. The company even suggested that price increases could be possible in the future if the SWITCH studies demonstrate a hypoglycemia benefit with Tresiba. While such a result would certainly help the company in payer negotiations, we are generally skeptical about the possibility of any further price increases in such a cost-conscious environment. In terms of the patient base for Tresiba, Novo Nordisk expects approximately half the patients to be new to insulin and half to switch from other products. The company anticipates that the majority of switches will come from Lantus but acknowledged that there will likely be some cannibalization of Levemir sales as well.
• Ryzodeg has been launched in Mexico, India, and Bangladesh and was approved in the US along with Tresiba. Management did not comment specifically on Ryzodeg’s progress during the call, but the company’s financial report indicated that feedback has been encouraging. We continue to assume that most of the product’s sales will come from emerging markets and that penetration in the US will be fairly minimal.

Xultophy

• Management emphasized the high clinical value for Xultophy (insulin degludec/liraglutide), which continues to make progress in Europe. Xultophy has now been launched in Sweden in addition to Switzerland, Germany, and the UK. We learned during Q&A that it has also received a positive opinion in Scotland, and that Novo Nordisk is aiming for a price of ~DKK 35-40 per day ($5-6 per day; $2,000 per year). Management stressed that the company believes Xultophy has “exceptionally high clinical value” and that it is in a strong position in pricing negotiations. However, management also acknowledged that the challenges continue in Germany following a very surprising “no additional benefit” ruling from IQWiG last quarter. As with Levemir and Tresiba, Novo Nordisk appears to be willing to sacrifice some initial access for Xultophy in order to achieve a price it believes is appropriate for the product’s value. In terms of positioning, the company expects that Xultophy will initially be targeted to patients on basal insulin but could eventually be used right after oral medications as well – we certainly believe it will be.
• Interestingly, Novo Nordisk expressed confidence that Xultophy has a “much better profile” compared to Sanofi’s LixiLan (lixisenatide/insulin glargine). While we have not seen as much detailed data on LixiLan, we have assumed the two products would be fairly comparable clinically.
• GLP-1 agonist/basal insulin combinations continue to generate plenty of buzz on the conference circuit. The class attracted a great deal of attention at EASD last month with the presentation of new positive topline phase 3 results for LixiLan, a new sub-analysis from the DUAL I trial for Xultophy, and glowing commentary from Dr. John Buse (University of North Carolina, Chapel Hill, NC) and Dr. Stephen Gough (University of Oxford, UK). We also heard a few points of criticism from Dr. Naveed Sattar (University of Glasgow, UK) and Dr. Leszek Czupryniak (Medical University of Lodz, Poland) who supported the option of first initiating a GLP-1 agonist and then titrating basal insulin separately. However, even those critics felt that the combinations would be appropriate for some patients and have a role among the many available treatment options for type 2 diabetes.

NovoLog/NovoRapid

• NovoLog (NovoRapid; insulin aspart) sales rose 3% YOY in constant currencies (15% as reported) to DKK 5.1 billion (~$763 million) in 3Q15. Sequentially, sales declined 2% as reported. While the call included little commentary on NovoLog, the company’s financial report indicated that the product has continued to gain market share in the US. We assume this is due to more favorable formulary positioning in 2015 following the loss of a major Express Scripts contract to Lilly’s Humalog (insulin lispro) in 2014. The formulary situation looks to remain fairly stable in 2016 barring any major changes still to be announced: formulary exclusion lists released in August showed that Express Scripts will continue to favor Lilly’s rapid-acting insulins and CVS Caremark will favor Novo Nordisk’s.

Figure 5: NovoLog/NovoRapid Sales (1Q12-3Q15)

• By region, growth was strongest in China, where NovoLog rose 24% in constant currencies (45% as reported) to DKK 222 million (~$33 million). The product also performed well in International Operations, rising 10% in constant currencies (8% as reported) to DKK 507 million (~$76 million). Sales rose 3% in constant currencies (5% as reported) in Europe to DKK 1.1 billion and were flat in constant currencies in both North America (19% reported growth; sales of DKK 3.1 billion [~$463 million]) and Japan/Korea (1% reported growth; sales of DKK 216 million [~$32 million]).

Other Insulins

• NovoMix sales rose 2% YOY in constant currencies (11% as reported) to DKK 2.7 billion (~$405 million) in 3Q15. Sequentially, sales declined 5% as reported. By region, NovoMix performed best in China, rising 11% in constant currencies (29% as reported) to DKK 726 million (~$108 million). Growth was also positive in International Operations, rising 6% in constant currencies (7% as reported) to DKK 574 million (~$86 million). NovoMix typically performs best in these regions, where there is a larger market for premix insulin, but these growth figures are somewhat lower than those seen in recent quarters. Operational growth was flat or negative in North America (flat; +18% as reported), Europe (-7%; -5% as reported), and Japan/Korea (-8%; -6% as reported).
• Human insulin sales rose 3% YOY in constant currencies (12% as reported) to DKK 2.8 billion (~$413 million) in 3Q15. Sales were flat sequentially. The portfolio saw positive operational growth in International Operations, China, and (surprisingly) North America, while sales declined in Europe and Japan/Korea.

Victoza

• Victoza (liraglutide) sales rose 20% YOY in constant currencies (36% as reported) to DKK 4.7 billion (~$697 million) in 3Q15. Sequentially, sales rose 4% as reported. This continues a streak of double-digit growth for the past year after a more sluggish period in 1H14. By region, North America remained the main driver of growth, with sales up 23% in constant currencies and 46% as reported to DKK 3.4 billion (~$511 million).

Figure 6: Victoza Sales (1Q12-3Q15)

• As in recent quarters, Victoza was boosted by growth for the overall GLP-1 agonist class, especially in the US. IMS Health data included in the presentation slides showed US volume growth for the class rebounding to close to 20%. This is approaching the growth rates seen in 2012-2013 and is a very strong recovery from the ~8% growth at the beginning of this year. By value, the GLP-1 agonist class accounted for 7.5% of the total global diabetes market as of August 2015, up from 6.9% in August 2014, driven by expansion in the US and Europe. During Q&A, Novo Nordisk attributed this growth primarily to the ebbing of concerns about pancreatic safety and the arrival of new products.
• Victoza has lost some market share within its class in the US this year to new products Lilly’s Trulicity (dulaglutide) and GSK’s Tanzeum (albiglutide). However, Victoza’s share by total prescriptions remained steady at 59% from June to August, while Trulicity and Tanzeum both gained share (rising from 6% to 7%) at the expense of AZ’s exenatide franchise (weekly Bydureon and twice daily Byetta, which declined from 28% to 27%). While the new entrants appear to have taken comparable bites out of Victoza’s and the exenatide franchise’s shares overall, the most recent numbers suggest that exenatide may feel more of an effect going forward.

Table 3: GLP-1 Agonist Market Share

• Chief Scientific Officer Dr. Mads Thomsen reiterated his cautious optimism that the LEADER CVOT for Victoza may be able to demonstrate cardioprotection, though the company stressed that the trial is only powered to show non-inferiority. During Q&A, an analyst suggested that if the benefits seen in EMPA-REG OUTCOME with Lilly/BI’s Jardiance (empagliflozin) were due to the drug’s diuretic effect, it might reduce the likelihood that a non-diuretic drug like Victoza could produce similar outcomes. Dr. Thomsen stated that he believes the EMPA-REG OUTCOME results were great news for patients and agreed that the diuretic effect was the most likely driver of benefit. However, he expressed even greater excitement about the potential for benefits with Victoza that more directly address the underlying pathology of CVD, though the company stressed that the trial is only powered to show non-inferiority and that neutrality is therefore the most likely outcome. He believes the drug could interfere with the pathological processes that occur in the vasculature in type 2 diabetes and thereby produce a legacy effect that persists even after the drug is discontinued. As he noted, such a result would also suggest that combining Jardiance and Victoza could yield even more impressive effects, as the two drugs would be improving outcomes via completely different mechanisms.
  
  • We agree that this trial is more likely than most CVOTs to demonstrate benefit due to its relatively greater individual patient exposure (mandated minimum exposure of 3.5 years per patient and total exposure of over 30,000 patient-years). That said, we are doing our best to keep expectations in check, as the trial may not be powered to reveal this sort of subtle long-term benefit, especially in a high-risk population.
• Management noted Novo Nordisk’s recent decision not to pursue a type 1 diabetes indication for Victoza following modest results from the phase 3 ADJUNCT trials. The ADJUNCT ONE (n=1,398; 52 weeks) and ADJUNCT TWO (n=835; 26 weeks) trials both demonstrated modest placebo-adjusted A1c reductions (~0.2%-0.3%) and 2-5 kg placebo-adjusted weight loss with 1.2 mg and 1.8 mg doses of liraglutide. In both trials, there was no difference between groups in rates of severe hypoglycemia, and rates of confirmed symptomatic hypoglycemia were actually significantly increased with the 1.2 mg and 1.8 mg doses in ADJUNCT ONE and with the 1.2 mg dose in ADJUNCT TWO. We have not seen glycemic variability results from either study. Essentially, the striking “composite” results seen with Victoza in type 2 diabetes (significantly lower A1c, substantial weight loss, and significantly less hypoglycemia) were not reproduced in this population. We do believe that these drugs provide clinically meaningful benefits for at least some patients in the “real world” but agree that the risk/benefit profile demonstrated in these trials would be unlikely to pass muster with the FDA and especially payers. It may be that SGLT-2 inhibitors will emerge as the more efficacious type 2 diabetes drug class for type 1 diabetes due to their insulin-independent mechanism, though the risk of DKA provides some cause for skepticism over that class’ future in this population as well.

Saxenda

• The story for Saxenda (liraglutide 3.0 mg) remains one of gradual uptake and limited access in the US. We assume this is largely due to price (~$1,068/month) as well as PCPs not having as high a comfort level at this stage to prescribe obesity medication. Novo Nordisk took an optimistic tone on the product overall during the call, noting that uptake has been encouraging and that temporary and contracted coverage had emerged earlier than expected with more than 50 million lives now covered. However, TRx data in the company’s presentation showed Saxenda still tracking well behind Orexigen/Takeda’s Contrave (naltrexone/bupropion) and Arena/Eisai’s Belviq (lorcaserin) at comparable points post-launch. Saxenda had 2,122 total prescriptions 24 weeks after its launch compared to 5,021 for Belviq and 10,206 for Contrave. It is tracking ahead of Vivus’ Qsymia (phentermine/topiramate), which had only 1,600 TRx at 24 weeks. Management also acknowledged that access for Saxenda remains restricted, particularly pointing to the fact that Medicare Part D does not cover any anti-obesity medications. While all new obesity medications are likely suffering from this lack of reimbursement, Saxenda may feel more of an impact relative to its competitors due to its higher list price.

Pipeline Highlights

Faster-Acting Insulin aspart

• Novo Nordisk still plans to submit its faster-acting insulin aspart formulation (Faster aspart) to regulatory authorities by the end of the year. The company has completed an additional 26-week treatment period for the phase 3 Onset 1 study (n=1,143 patients with type 1 diabetes) and stated that results confirmed the positive initial results reported in March. Those results, and those from the Onset 2 trial in type 2 diabetes, demonstrated non-inferior to slightly superior A1c reductions with Faster aspart vs. NovoLog (insulin aspart). Faster aspart also blunted one-hour postprandial glucose excursions significantly more than NovoLog and provided comparable A1c reductions to mealtime NovoLog when given post-meal. The overall rate of hypoglycemia was comparable between groups, but Faster aspart did cause more hypoglycemia in the postprandial period. Based on these results, we do not expect Faster aspart to be a completely transformative “once a generation” product, but the potential for dose flexibility should be a meaningful benefit for patients.

Semaglutide

• Novo Nordisk has initiated two phase 2 trials of its next-gen GLP-1 agonist semaglutide dosed once-daily, one in type 2 diabetes and one in obesity. The type 2 diabetes trial (n=704) is evaluating a variety of doses of semaglutide (0.05-0.3 mg/day) vs. liraglutide (0.3-1.8 mg/day) and placebo. The primary endpoint is change in A1c after 26 weeks. Secondary endpoints include change in fasting plasma glucose, body weight, and blood pressure. The trial is recruiting participants and expected to complete in October 2016. The obesity trial (n=935) is evaluating semaglutide (0.05-0.4 mg/day) vs. liraglutide (0.6-3.0 mg/day) and placebo in people without diabetes with a BMI ≥30 kg/m². Its primary endpoint is percent change in body weight after 52 weeks. Secondary endpoints include the proportion of participants achieving ≥5% and ≥10% weight loss, change in body weight, and change in A1c and fasting plasma glucose. The study is also recruiting participants and expected to complete in April 2017.
• Novo Nordisk clearly sees semaglutide as a very versatile product. The company is developing the drug primarily as a once-weekly agent (see below). However, Dr. Thomsen explained during Q&A that Novo Nordisk also believes semaglutide could be quite appealing as a once-daily product. The available data suggests that semaglutide has greater intrinsic efficacy than Victoza, and once-daily dosing could provide steadier concentrations and therefore offer better coverage and fewer peak-related side effects compared to once-weekly administration. Novo Nordisk believes there is a long-term market for multiple agents in the GLP-1 agonist class, including oral formulations and injectable agents with a variety of dosing regimens (once-daily, once-weekly, once-monthly). The company is clearly trying to cover all its bases in this area, and is likely also planning ahead for Victoza’s eventual loss of patent exclusivity. 
• Novo Nordisk highlighted the recent positive results from the phase 3a SUSTAIN 3 trial of once-weekly semaglutide vs. AZ’s Bydureon (once-weekly exenatide). The trial demonstrated significantly greater A1c reductions (1.5% vs. 0.9%; baseline = 8.4%) and weight loss (5.6 kg vs. 1.8 kg; baseline = 96 kg) with semaglutide vs. Bydureon, though with double the rate of nausea (22% vs. 11%). SUSTAIN 3 is the second of six phase 3 trials for once-weekly semaglutide to report results. SUSTAIN 1, which reported topline results in July, demonstrated superior ~1.5% A1c reductions with semaglutide vs. placebo. The next trial to report results should be SUSTAIN 4 against Sanofi’s Lantus (insulin glargine). The remaining trials should provide a good sense of semaglutide’s profile compared to other type 2 diabetes drug classes and its logical position in the treatment algorithm, though we are disappointed that there are no trials evaluating the drug against or in combination with an SGLT-2 inhibitor. If the current timeline holds, we would expect FDA approval in early 2017. See below for an overview of the trials and their expected completion dates.

Table 4: SUSTAIN Phase 3 Program for Semaglutide

Oral GLP-1 Agonists and Insulins

• Novo Nordisk highlighted the recent decision to initiate phase 3 trials for oral semaglutide. The phase 3 PIONEER program, similar to the SUSTAIN program, will consist of seven trials and enroll approximately 8,000 patients. The first study, slated to begin in 1Q16, will evaluate three doses of oral semaglutide (3 mg, 7 mg, and 14 mg) vs. Merck’s DPP-4 inhibitor Januvia (sitagliptin). The remaining six trials will be initiated in 2016 and will include a CVOT. This product could have significant disruptive potential if the challenges around dosing and bioavailability are overcome. The possibility of future fixed-dose combinations with an SGLT-2 inhibitor is also very exciting.
• Novo Nordisk has two oral formulations of liraglutide in phase 1. We have not heard any recent updates on the plans for either candidate (OG987GT and OG987SC) and assume that the company is focusing the majority of its oral GLP-1 efforts on semaglutide at this point – understandable given the molecule’s superior efficacy as an injectable formulation.
• Novo Nordisk initiated a phase 2a trial of its primary oral insulin candidate OI338GT (NN1953) and a phase 1 trial of a new candidate (OI320GT/NN1957) in July. The trials are expected to complete in December 2015-January 2016. It is not yet clear whether the company plans to advance both candidates all the way through the pipeline or whether the new formulation could eventually supersede the more advanced candidate. We still consider Novo Nordisk the heavy favorite to develop a successful oral insulin formulation due to its significant resources and extensive experience developing peptides for diabetes. However, in Biocon’s F2Q16 update, management stated that the company is “extremely excited” about data from the first set of US trials for its oral insulin Insulin Tregopil (IN-105), and we are curious to see if these results change our assessment of the competitive landscape. Oramed also has an oral insulin candidate (ORMD-801) in phase 2. Evaluating that compound’s prospects has been difficult, as results from completed trials have been complicated by small sample sizes, variability in baseline insulin dose, and manufacturing malfunctions.
• Perfecting oral delivery of peptides is clearly a long-term priority for Novo Nordisk. Research in this area will be a major focus of the company’s recently announced collaboration with Dr. Robert Langer’s laboratory at MIT. This collaboration will focus on new early-stage compounds and will not affect Novo Nordisk’s existing oral GLP-1 agonist and insulin programs. Key research goals include avoiding premature degradation of peptides in the body, effectively transporting them over epithelial barriers, limiting variable absorption, and producing sufficient amounts of the peptide and delivery vehicle in a cost-effective manner. Novo Nordisk has identified ensuring bioavailability and reducing variability as the biggest challenges associated with oral peptide delivery, so we expect those will be high priorities for the team. 

Other Pipeline Candidates

• The company has completed the last phase 1 trial for once-weekly basal insulin LAI287 and plans to further investigate side effects before initiating additional trials. In a five-week multi-dose trial in type 2 diabetes, LAI287 demonstrated dose-dependent pharmacokinetics (PK) and pharmacodynamics (PD) in patients previously treated with once-daily basal insulin. The candidate also demonstrated low variability comparable to that of Tresiba and a half-life of 185 hours supporting once-weekly dosing. However, the trial also apparently revealed side effects that the company believes warrant further investigation before proceeding. Management declined to provide further details on those side effects, cautioning that the small sample size (n=49) limits the conclusions that can be drawn. We are glad to hear updates on this candidate (which has also completed a phase 1 trial in type 1 diabetes) after a fairly long streak of no news and believe the company’s cautious approach regarding adverse events is a smart one.
  
  • Novo Nordisk’s other long-acting insulin candidate LAI338 is no longer listed on the company’s pipeline. The compound completed a phase 1 trial in March according to ClinicalTrials.gov. We have not heard any updates on the reasons for discontinuation but assume they relate to safety signals or PK/PD that was not compatible with once-weekly dosing. We assume there may Other companies developing insulins with the potential for once-weekly dosing include PhaseBio (PE0139; phase 1), Hanmi (LAPSInsulin-115; phase 1), and AntriaBio (AB101; preclinical).
• Novo Nordisk advanced a formulation of the gut hormone PYY (NN9747) into phase 1 for obesity. The first clinical trial of the candidate is evaluating its safety, tolerability, and PK/PD vs. placebo in 118 otherwise healthy participants with a BMI of 25-34.9 kg/m² (part 1) or 27-39.9 kg/m² (part 2). It is currently recruiting participants and is expected to complete in October 2016. This news is further exciting evidence of Novo Nordisk’s broad commitment to early-stage obesity research. The company’s phase 1 obesity pipeline also includes the glucagon analog G530L (NN9030) and the long-acting amylin analog NN9838. We expect that more products will appear in the early-stage pipeline in the near future following Novo Nordisk’s recent acquisition of Calibrium and MB2, two companies founded by the renowned Dr. Richard DiMarchi and focused on peptides for diabetes and obesity.

Questions and Answers

Insulins – Big Picture

Q: Could you remind us of your expectations for the long-term insulin market growth to remain roughly at 7% volume and 3% price? Sanofi has cut its diabetes guidance today and Lilly has talked about slowing insulin market outlook. Has your outlook changed, and if not, why are you different from the competition?

A: At the current level, our assumption for overall insulin market growth is still 5% in volume, and given our portfolio, we expect that we will be gaining share. So the overall model still applies. We are assuming we will get 7% volume lift. The value will very much be dependent on the pricing environment, the competitive pressure, and pricing reforms. We have previously said that well-penetrated, it should look like +3% coming from mix and price. This outlook is perhaps somewhat optimistic given the current conditions. So we’ll have to adjust as we go along, but our overall expectations for market growth are 5% in volume, Novo Nordisk beating the market growth with market share gains, and 7% is a pretty good guess in that regard.

Q: Is there potential for pricing to go negative?

A: I think we will be getting some mix effects because we have a new portfolio we are introducing. We will be replacing some of the old human insulins and the modern insulins we know of today, Levemir and NovoMix. But it’s a little bit too early to state. I’d prefer that we discuss this in more detail when we get to the long-term financial targets.

Q: Sanofi’s change to their guidance clearly implies something has changed. The one thing that may or may not be a driver of that is the arrival of a biosimilar in Europe and the rest of the world. We do see a healthy uptake of the biosimilar in Japan, for example. How do you see the environment changing at all in the basal space where that biosimilar is available?

A: It’s difficult for me to guide you on what’s included in Sanofi’s midterm outlook for the insulin market. The only thing one can speculate on is that Sanofi’s basal franchise is under pressure from two sides: the pressure from the launch of an innovative premium product, Tresiba, and the launch of a biosimilar, Basaglar. So obviously this is putting pressure on their pricing and their volume growth expectations going forward.

We talk about Sanofi being pressured on both sides. Part of it is coming from what Tresiba delivered. So when we look at our situation, Tresiba is a positive we can add in. As we’ve stated before, we would say that the biosimilar versions of glargine up against Lantus is more a dynamic within the glargine area and has limited direct impact on our current sales of Levemir. And where we are then in a position where we can promote Tresiba, we further withdraw ourselves from that dynamic. So not that you can say it has no impact, but it’s much, much less than when you sit and are getting squeezed from two side as is the case for others.

Q: Could you add some color on the contract you’re going to lose in the US in January? Was that a decision not to participate because the pricing went too low, or did you bid and just ended up losing the contract?

A: This was a deliberate decision on our part not to buy that contract at the price that we would have to accept, because we felt that it was a very aggressive pricing. This is a significant contract, but we do win some and we lose some every year. But this one will have some market share volume impact on us. On the other hand, it was at a very low price, so the profit impact will be less.

Q: We’ve seen two contrasting diabetes outlooks from you and your friends in Paris today. To date, in the US basal insulin market it’s all been about co-preferred contracting rather than competitors going for exclusivity. Do you think exclusivity becomes part of the game in US basal post-2016?

A: It is clear that the US market will undergo some tremendous transition next year with the advent of Tresiba now, with the advent late next year of Basaglar, and with the advent of LixiLan. We don’t know how Sanofi is going to position LixiLan, but I would imagine it will go up head-to-head against Xultophy, which will be launched at the same time, because we are certain we have a much, much better profile. So in a way, they have the option of positioning this as a better basal also. So there’s going to be a lot of traffic. Our pricing strategy will remain as it is until we see competitive actions. This means that we think we are pro-choice, and we think there are big parts of the market in the US, and especially that we are targeting initially, that are interested in choice. But it’s anybody’s guess what will happen post-2016 and into 2017 when the market gets crowded. And if some of the competitors are not successful getting market penetration and access, then anything can happen. Of course it’s clear on the payer side that some of them would like to build exclusive contracts to drive up rebates. Other insurance companies prefer to have different plans with choices that they market to employers that are willing to offer this to their employees. So it’s a very complex situation to forecast. Our pricing strategy remains. And the moment the market condition dictates us to make a change, we will inform you.

Bear in mind that the exclusive positions some of the prandial insulins have in the US market is because they’re basically the same. But you cannot claim that Tresiba and detemir and glargines are the same. These are truly different choices.

Q: If we look at the total IMS data, we do see Novo still gaining some share vs. Sanofi. But if we look at the new-to-brand data in IMS, there seems to be the reverse situation where certainly since the launch of Tresiba the new-to-brand seems to be moving in Sanofi’s favor. Is this just the usual inconsistencies of IMS and the fact that they’re missing some important contracts, so it’s not giving us the right story?

A: When we compare NBRx up against total prescription shares, they’re both highly correlated. Whether you can always make the math work out completely is another question. That’s sometimes not possible. When you have a launch of a new product you will see the NBRx be impacted initially, so short-term you will see a large movement and it will take awhile before that plays into the actual market share. This is what we’ve seen in many other launches and what we expect. We will still expect the impact to be modest. Of course there will be one but it will not be as dramatic as you might read into the first weeks or months of NBRx data. And to confuse matters, when we launch Xultophy in certain markets there we’ll also see a dampening of the NBRx in the basal segment because some clinicians use Xultophy as a better basal. In a way you should look at Tresiba plus Xultophy as one.

Q: You see continued problems in China. You also guide down on 2016. Could you provide a bit more color on what should change this? Is it only market dynamics or is it also a broader use of locally produced products?

A: We are impacted by increasing local competition and we are impacted by a segment shift. We have historically experienced the same in Japan where we have a strong position in the premix market. China used to be a premix market. Now the basal market is going and the competition has a gold standard and the local competition has a copy of the gold standard, and that means we get fewer patients when the patients shift from mix to basal. And so we are being hit by that. The only real solution to this is that we get Tresiba into the Chinese markets. That is a couple of years out, so I think we will be facing relatively tough market conditions in China for a couple of years.

We have filed Tresiba for approval in China, so it’s on its way and the volume growth is still good, albeit a little bit in the segments where we’re not so strong. We do believe at some point the pricing impact will normalize and the volume effects of the vast under-treatment of the Chinese diabetes population will be a positive effect again. So we expect China to come back, but here in the immediate future, we expect lower growth.

Tresiba

Q: On Tresiba pricing, the wording has changed to a small premium to your own products vs. a small premium to the broader basal market. Sanofi’s guidance clearly implies price declines over time, which suggests that a price premium for Tresiba at launch could increase with time. How do you think about that on a three to four year view?

A: We have picked a price that we know, which is Levemir, and then we have added a premium. We have added a 10% premium to Levemir. In the future, our pricing strategies and our rebating strategies will be dependent on where competition will be moving with their prices. So we’ll have to take into consideration that in reality, what the payers are looking at is the net price. That’s list price minus rebates. And we will have to adjust accordingly, should the current pricing environment become more tough.

Q: Could you give us some idea of the dynamics we should play through on the Tresiba ramp in the US? Specifically, how are you thinking about a 180-day block for Medicare Part D? Do you think that applies and how can you potentially negotiate that in a faster way?

A: You should expect a relatively slow ramp-up of Tresiba in the United States. We are not currently going to pursue an aggressive access strategy for Part D. We’re primarily going to go after the commercial market in insulin. We think Tresiba offers value to the patients and the healthcare professionals and therefore we want to price that to the extent that we can. We are not in a panic. We are going to be in diabetes many years into the future, and it is befitting and beneficial for us to try to retain the segment value as opposed to short-term cutting price.

So very gradual penetration much like you have seen in the past. What is different is that whereas in the past we would normally guide that for new products we’d get a part of the new starts on insulin, in this case we think the value of the product is such that we are expecting to get half the patients coming from new starts and the other half coming from already insulin-treated patients. Therefore there will be some cannibalization of our own Levemir patients, but not to the extent that we expect to see Levemir decline. But the growth of Levemir will be stunted as we penetrate the market. But if Lantus has the lion’s share of the basal market in the United States, our anticipation is that as far as switches go it will be primarily Lantus patients that would be switching to Tresiba.

Q: Does the pricing strategy you elucidated over the last few quarters on Tresiba remain valid if the SWITCH study fails to show a hypo benefit?

A: Our current pricing strategy is based on the current label. And this means if SWITCH is positive, that’s also an opportunity to raise the price. It’s not only going in one direction, we should remember. But let’s have the data and then we can chat about that.

Xultophy

Q: On Xultophy, you’ve received a positive opinion in Scotland. Do you expect to see that the average price of Xultophy is going to be in that ballpark, around DKK 38-40 per day and that you will see a general positive reimbursement environment for Xultophy in Europe compared to Tresiba where it has been very negative?

A: That is correct, we have positive opinions out of the Scottish medical consortium and at a price point in the high DKK 30s per daily treatment. Whether that will be accepted widely is another question. We are also faced with dynamics, for instance, in Germany where we are struggling to have them even look at the data because they do not accept the comparators. My view is that Xultophy is such a good product. If we can get an acceptance at that price point, we have a possibility in the near to medium term to get Xultophy into a number of European markets and accepted. It is exceptionally high value from a clinical point of view, so we’re in a good position to hopefully come to an agreement on what prices should be reimbursed. But it will be small incremental steps here in the beginning. We believe in the value so we’re not willing to not have the full value of innovation of Xultophy acknowledged.

Q: Where do you think Xultophy will be positioned first? Do you think this is the sort of drug people will take when they’ve failed orals before they move to insulin? Or would you expect it to be reserved to people who’ve tried insulin and are going to a combination therapy? That could then perhaps justify an even greater premium price.

A: It is clearly positioned as an intensification of basal patients because they’re already on an injectable, on an insulin. They’re measuring their blood glucose. Transitioning into Xultophy will give them a better A1c and weight lowering without additional injections and with a very low risk of hypoglycemia relative to being on pure insulin. So it’s a very, very strong value proposition there. Over time it is likely we will see some clinicians starting patients off after orals on Xultophy because they believe an interim step on insulin will not be worth it and they can go straight to Xultophy. But initially it’s going to be from the pool of uncontrolled patients on basal insulin. And that’s also our positioning. That is bound to change over the next five years, but this is where we start.

Victoza

Q: Can you talk about the thought process in terms of a daily semaglutide vs. Victoza in a relatively competitive sub-segment of the GLP-1 space? Is this biosimilar protection or just a better product?

A: The idea about once-daily semaglutide arose from two sides of the equation. One is that we know Victoza is today’s best product in the market and that’s witnessed by how we perform. But with that said, the data we’ve seen on semaglutide so far justify the hypothesis that semaglutide possesses even more intrinsic efficacy. Now since we do have some peak-to-valley fluctuation after once-weekly dosing, we can go down to a peak-to-valley fluctuation within one percentage point if we administer the molecule on a daily basis. That will mean two things: that we have even more exquisite coverage at even more peripheral target tissue receptor levels and that the peak-related side effects that GLP-1s can be associated with should be reduced. So we believe as a company that there will be, in the long-term future, a market for oral GLP-1, for once-daily GLP-1, and for once-weekly GLP-1. And believe it or not, we’re even looking into a once-monthly GLP-1. And to cater to the best possible needs of the patients we do that to tweak out whatever benefit we can use in such a treatment paradigm.

It is clear that we do believe semaglutide as a molecule is more powerful than Victoza, especially on the weight side. And the trick here is that cutting the weekly dose into seven daily doses, we’re striving to go even beyond the efficacy you know for the once-weekly 1 mg semaglutide dose.

Q: Could we have an update on your thoughts heading into LEADER? We’ve seen the CV benefit with empagliflozin. It’s postulated to be driven by a diuretic effect, which Victoza doesn’t have. Has your optimism of previous quarters changed in light of EMPA-REG?

A: It’s really good news for patients, and also for our friends at Lilly and Boehringer Ingelheim, that they have these data on empagliflozin. But you can argue it does not reduce the likelihood of the LEADER trial coming out positive for a couple of reasons. One is that now history tells us it can be done to actually achieve a CV benefit. The other is that this has nothing to do with the disease pathology, because the way the curves dissect out initially from the get-go in the trial tells me that it’s more like a diuretic effect or something. And the interesting thing is what if you load up all patients with maximum doses of thiazide diuretics, which do have a cardiovascular benefit? Would you then still see the benefit of empa? Many questions remain to be answered.

My view is that it’s more interesting to have an agent that can interfere in the vessel wall pathology that’s due to the glycemic and the dyslipidemic and the hypertensive maladies that are ongoing in the type 2 diabetic body. That would be a long-term investment for the patient because you will get the legacy or the memory effect of the drug that was once used. That’s unlike a diuretic, which works when it’s given and then it stops working the day after it’s given the last time. On the one hand, this will go hand-in-hand. If we show good data then they will actually be additive to those of empa, which is also good news for the patient. But my hope is we have something where there remains a legacy effect, potentially years after people have encountered Victoza.

Q: You saw quite weak insulin growth in 2011, and one of the reasons was the high focus on the Victoza launch. Now that you’re putting all your sales force on Tresiba in 2016, could we see a similar impact on Victoza next year? We also see that Lilly has started their DTC campaign recently.

A: We prefer not to give guidance on individual products for next year. What we can say about Victoza is you’re right that there will be a period of time where they’ll be full force on Tresiba and then we’ll revert to the current structure we have in the US. We will all along in 2016 have a specialized sales force on Saxenda. This is something we do to avoid cross-promotion and avoid compliance risk. We believe there will be solid growth in the GLP-1 market and we hope that the strength of the gold standard in this market, Victoza, will remain after a short detour to promote Tresiba. Also, when the force is detailing Tresiba they have a second detail that will be Victoza for many of the reps. It’s not like it’s completely out of the market.

Q: Do you believe some of the positive growth we’ve seen in Victoza over the last six to eight months has something to do with the approval of Saxenda in obesity? Could there be higher prescribing on Victoza given the low formulary access for Saxenda? So that would also change when the formulary access increases for Saxenda?

A: I don’t think there is a link between the approval of Saxenda and the growth in the marketplace. I think we can detect the growth coming all the way back to the time when FDA and EMA eradicated the lingering concern on pancreatic safety after the Butler paper. We believe that is the key driver. And also of course that there are more companies out in the street talking about the virtues of the GLP-1 as therapy. These are the main factors as far as we see.

Pipeline

Q: Could you give some more detail on the side effects for the new long-acting insulin?

A: We have few subjects because it’s only a phase 1 program. The molecule is an insulin analog that has been designed, both the backbone and the side chain, to give a rather unique mechanism of protraction so that it circulates in the body with a terminal half-life of 185 hours. The issue is that when you have so few subjects as you do in this kind of phase 1 program, you always see side effects and adverse events. And when we report to you, we don’t have the full picture. So rather than go into detail with such small numbers, I’d rather say that it’s very natural that you look into the side effects you have seen, investigate those maximally, and then make a decision as to progression into phase 2. So I won’t be more specific than that at this point, but obviously as the program hopefully continues into further development we can start being more specific once we’re in the phase 2 mode. Phase 1 is limited patient exposure.

Another thing we need to discuss is the hesitation with Tresiba from clinical investigators about how we can titrate a long-acting insulin without running the risk of hypoglycemia and when we get to steady state. We are indeed having these discussions, albeit at a higher level, with this insulin. With that said though, I was personally surprised when I saw the results that we went up against the best insulin in the game, which is insulin degludec, and we actually saw that the peak-to-trough and the overall variability was reminiscent of that achieved with once-daily degludec. However, the titration issue remains because with a half-life of more than one week, the time to steady state and when you can start titrating does become something we need to consider before a phase 2 program.

-- by Emily Regier and Kelly Close