Novo Nordisk 4Q15 – Diabetes/obesity portfolio up 9% operationally to ~$3.4 billion in 4Q15 and ~$12.7 billion in 2015; Levemir and Victoza key drivers; Topline LEADER results in March - February 3, 2016

Novo Nordisk provided its 4Q15 update this morning in a call led by CEO Mr. Lars Sørensen. The company’s diabetes and obesity portfolio posted sales of DKK 22.9 billion (~$3.4 billion) in 4Q15, up 9% year over year (YOY) in constant currencies and 19% as reported. For 2015 as a whole, the diabetes and obesity portfolio brought in DKK 85.6 billion (~$12.7 billion) in sales, up 9% YOY operationally and 22% as reported. In 4Q15, as in past quarters, growth was driven by Levemir (insulin detemir) and Victoza (liraglutide) – although in 4Q15, Levemir provided an even higher percentage of growth that usual. Specifically, of net quarterly growth for the company of ~ $150 million, Levemir growth was $120 million compared to a bit under $50 million for Victoza and about $25 million for NovoLog – completing that picture were declines in human insulin and NovoMix. Levemir accounted for 35% of growth while Victoza accounted for 20% and NovoLog accounted for 22%. Levemir grew a whopping 22% YOY in constant currencies (36% as reported) to DKK 5.0 billion (~$745 million) and Victoza grew 10% YOY in constant currencies (22% as reported) to DKK 4.9 billion (~$724 million). Levemir’s market share by total prescriptions is holding steady at 24% (implying to us there have been price reductions – given the overall growth, we otherwise would expect market share increases). Victoza had a very good quarter though as expected has lost some market share to new entrants to the class and now holds 56% of the US market share by volume. As a reminder, Lilly’s Trulicity had a great 4Q15 in its fifth quarter on the market and provided nearly half of Lilly’s quarterly growth, though just 7% of overall revenue! Novo Nordisk’s new-generation insulins (Tresiba, Xultophy, and Ryzodeg) posted DKK 461 million (~$68 million) in sales, up 76% YOY (as reported) and 23% sequentially from a fairly low base – the three provided growth of about $25 million from 4Q14 to 4Q15, about the same growth as NovoLog. Saxenda is continuing to grow its presumably low base and is now the market leader in terms of value though Novo Nordisk isn’t reporting numbers yet. Overall, our learning from the quarter was that Levemir drove much of the success, with Victoza and GLP-1 growth adding as well.  

On the R&D side, management shared that topline results from the LEADER CVOT for Victoza will be announced in March. There has been plenty of speculation about the potential for cardioprotection in this trial, though it is only designed to show non-inferiority. Novo Nordisk also announced the initiation of a phase 3b trial evaluating the durability of glycemic control with Xultophy vs. Lantus in 1,000 patients with type 2 diabetes not at target on oral medications. We also learned, in a bit of a surprise, that the company does expect an Advisory Committee meeting for Xultophy (submitted to the FDA in September) sometime this year. Read on below for our top ten highlights from the update, followed by Q&A.

Table 1: 2015 Financial results for Novo Nordisk’s major diabetes products    

Table 2: 2015 whole-company sales and growth by geography

Table 3: 4Q15 Financial results for Novo Nordisk’s major diabetes products    

Table 4: 4Q15 whole-company sales and growth by geography

Financial Highlights

• Novo Nordisk’s diabetes and obesity portfolio posted sales of DKK 22.9 billion (~$3.4 billion) in 4Q15, up 9% year over year (YOY) in constant currencies and 19% as reported; full year sales were up 9% YOY in constant currencies (22% as reported) to DKK 85.6 billion (~$12.7 billion). 4Q15 growth and full year growth occurred against a fairly easy comparison as sales growth for Novo Nordisk’s diabetes portfolio was 12.6% and 6.8% in 4Q14 and 2014, respectively. The overall 2015 performance was at the high end of Novo Nordisk’s 4Q14 forecast of 6%-9% sales growth in local currencies (18-21% as reported) for the year. Sequentially, sales rose 7% as reported in 4Q15.

Figure 1: Total Diabetes/Obesity Sales (1Q12-4Q15)

• As in previous quarters, North America and International Operations were the main drivers of growth. Notably, the company noted that beginning in 2016, it would be reorganizing its markets under a new regional set-up. Under the new split, the US will be a region on its own and Canada and Australia (formerly part of International Operations) will join Japan and Korea as “Region Pacific.” Under this new set-up, we assume it will become even clearer how US diabetes sales fare compared to ex-US sales, which might provide a better sense of how US performance drives overall performance of certain classes such as GLP-1 agonists and next-gen basal insulins.

Table 5: Whole-Company Sales and Growth by Geography (Full year 2015)

• By product, Levemir (insulin detemir) and Victoza (liraglutide) continue to be the primary drivers of growth, in line with what we’ve seen in recent quarters. The entire modern insulin portfolio (consisting of Levemir, NovoLog/NovoRapid [insulin aspart], and NovoMix) accounted for 48% of growth (as reported) in 2015. Within that, management highlighted Levemir as a primary contributor (26%), mostly from North American sales. Victoza accounted for 25% of 2015 growth and Tresiba accounted for 5%.

Insulins

• Novo Nordisk’s modern insulin portfolio (NovoLog/NovoRapid, NovoMix, and Levemir) collectively grew 11% YOY in constant currencies (22% as reported) to DKK 13.6 billion (~$2.0 billion) in 4Q15, driven largely by Levemir (up 22% operationally and 36% as reported to DKK 5.0 billion [~$745 million]). Sequentially, sales were up 9% for the modern insulin portfolio. For full year 2015, the modern insulin portfolio posted sales of DKK 50.1 billion (~$7.5 billion), up 7% operationally and 21% as reported from 2014.

Figure 2: Modern Insulin Sales (1Q12-4Q15)

• Novo Nordisk suggested that Levemir and NovoLog’s strong performance in North America, in particular, can be attributed to market share gains and underlying insulin market growth. In contrast, Sanofi management had suggested in its 4Q15 update that basal insulin and overall insulin market growth was slowing. Based on reported revenue from Novo Nordisk, Sanofi, and Lilly, we found that the basal insulin analog market fell 9% YOY as reported in 4Q15 to $2.6 billion and fell 8% YOY as reported in full year 2015 to $10.2 billion. The rapid-acting insulin market grew 6% YOY to $1.8 billion in 4Q15 but remained flat for 2015 overall with $6.4 billion in total sales. The combined basal and rapid-acting insulin market fell 5.3% in 4Q15 and 5.8% in full year 2015. For comparison, the market grew 7.9% in 2014. By product, Sanofi’s Lantus (insulin glargine) held 65% of the basal insulin market by value in 4Q15 and its next-generation Toujeo (U300 insulin glargine) held 4% of the market. Levemir held 28% of the market, Tresiba held 3%, and Lilly/BI’s new biosimilar insulin glargine Abasaglar held a mere 0.3% of the market. Within the rapid-acting insulin market, NovoLog (insulin aspart) accounted for 48% of sales in 4Q15, Lilly’s Humalog (insulin lispro) accounted for 45% of sales, Sanofi’s Apidra (insulin glulisine) accounted for 7% of sales, and MannKind’s Afrezza (inhaled insulin; formerly partnered with Sanofi) accounted for 0.1% of sales in 4Q15.

Table 6: Novo Nordisk Insulin Market Share

Figure 3: Total Basal Insulin Market Sales (1Q06-4Q15)

Figure 4: Total Rapid-Acting Insulin Market (1Q06-4Q15)

Levemir

• Levemir (insulin detemir) sales were up 22% operationally and 36% as reported to DKK 5.0 billion [~$745 million]). Sequentially, sales were up 8%. For full year 2015, sales totaled DKK 18.3 billion (~$2.7 billion), up 13% operationally and 29% as reported from 2014.

Figure 5: Levemir Sales (1Q12-4Q15)

• Management noted that North America is a main driver of Levemir’s growth. Sales of the product in that market grew 31% operationally (53% as reported) to DKK 3.7 billion (~$550 million) in 4Q15. Full year North American Levemir sales grew 19% operationally (15% as reported) to DKK $13.3 billion (~$2.0 billion). US growth was driven by underlying volume growth in the insulin market and Levemir market share gains, according to management. IMS data from the presentation slides indicate that Levemir held 24% of the basal insulin market in the US by volume as of November 2015. This is the same market share that Levemir held as of August 2015, as shown in the company’s 3Q15 update. However, it appears that Sanofi’s Lantus (insulin glargine) has lost a bit of market share to its next-generation basal insulin Toujeo (U300 insulin glargine). Lantus now holds 65% market share, up from 66% as of Novo Nordisk’s 3Q15 update, and Toujeo now holds 1% of the market compared to 0.5% in 3Q15. On the other hand, management also noted that Levemir’s growth rate in Europe and in Japan and Korea has been impacted by the launch of Tresiba (insulin degludec) in those regions. We’ll be interested in see how the recent launch of Tresiba in the US affects the dynamics of the basal insulin market here in the coming quarters. Presumably so much Levemir growth is happening in the US due to some contract wins – it would be interesting to know to what extent those were driven by price.

Tresiba

• Sales of the Tresiba (insulin degludec) franchise reached DKK 461 million (~$68 million) in 4Q15, up 23% sequentially from DKK 376 million (~$56 million) in 3Q15; full year sales totaled DKK 1.4 billion (~$214 million) in 2015, more than doubling from DKK 658 million (~$117 million) in 2014. These figures include sales for Ryzodeg (insulin degludec/insulin aspart) and Xultophy in addition to standalone Tresiba. YOY growth for 4Q15 was 76% as reported.

Figure 6: New-Generation Insulin Sales (1Q14-4Q15)

• Novo Nordisk shared that Tresiba has now launched in 39 countries, highlighting the product’s recent launch in the US in particular. Management characterized the initial feedback from patients and providers for Tresiba as “encouraging” and noted that coverage for the product is increasing with ongoing formulary negotiations. In the US launch announcement, the company shared that Tresiba has secured Lowest Brand Co-pay status with CVS Health’s national formulary. We also learned from CMO of North America Dr. Todd Hobbs that Tresiba was able to secure formulary access for Humana’s Medicare Part D and Medicare Advantage plans. Management stated in the 4Q15 update that the company had not expected to achieve Medicare access this soon and is “slightly ahead of our expectations when it comes to formulary access.”
• Outside of the US, Novo Nordisk showcased a graph of Tresiba’s basal insulin market value share in 12 countries, ranging from 3% to 33%. At the top end was Japan, where Tresiba enjoys a full third of the market nearly three years after launch. On the low end, Tresiba holds a mere 3% of the basal insulin market in the UK, also nearly three years after launch.  As in the past, the company stated that Tresiba’s performance is stronger in countries where it is reimbursed at a similar level to Lantus (insulin glargine) and its performance is weaker in countries where access is more limited. However, management highlighted Tresiba as a strong driver of growth in Europe and International Operations overall. Novo Nordisk shared that Tresiba also launched in Spain this month.

Figure 7: Tresiba Market Share

• In less positive news, management confirmed that Tresiba was withdrawn from the German market on January 15, 2016. The company announced Tresiba’s withdrawal from Germany in July 2015 but shared at its Capital Markets Day in November that it was engaged in ongoing negotiations with German authorities. During Q&A in the 4Q15 update, management shared that the company had “small hopes” that Tresiba could remain on the German market with the renewed discussions. However, the offer was similar to the initial offer, which would have priced Tresiba similarly to human insulin. This is extremely dispiriting for patients, particularly those who have nighttime hypoglycemia as a major issue. We believe patients will be more adherent to new and better long acting insulins and fervently hope that the manufacturers will be figuring out better ways to use big data to assess whether this is true.
• Kelly had her own first experience with Tresiba this morning. She sought to test the claim of good formulary access this morning, going to her 24-hour Walgreen’s in San Francisco’s Castro district (said to be the most profitable Walgreen’s in the US) before leaving for ATTD – indeed, her doctor called in the prescription yesterday with no other instructions. Kelly was surprised to be greeted by “Oh yes! We’ve got one thing for you! Here’s Tresiba – looks like a $25 co-pay” – the new insulin required no phone calls to Aetna, our company’s insurance provider, from either Kelly’s esteemed doctor, Dr. Nancy Bohannon, or by Walgreen’s. Toujeo was initially harder to get approved by Aetna (she had to “fail” Levemir – such poor choice of language, Mr. Bertolini!) and has a $100 co-pay, and initially had a hard time using the “card” although Kelly was extremely impressed by the COACH program that Sanofi is using – and overall, though it’s hard to compare Tresiba to Toujeo after one day, she prefers the Tresiba pen (“it’s easier to dial and much easier to push.”  She hasn’t in her n=1 trial had a chance to comment on hypoglycemia yet but says both are SO much (“miles!”) better than NPH, the “long acting” insulin she used to take and also seem much better than Lantus and Levemir – the delivery device alone for Toujeo is a major win over Lantus in her view as is COACH. She acknowledged it’s mostly formulary access these days that determines what patients buy, not patient or even doctor preference though patients can use workarounds if they are lucky to have insurance as strong as ours. Either way, Kelly is very happy that both next-gen insulins have been launched and hopes the data collection will go well so that as many patients as possible gain access and so that uninsured and underinsured patients can gain access to older insulins at lower prices. We need a thriving commercial market for all new products that are valuable to patients so that manufacturers continue to increase R&D investment (which Novo Nordisk did not in 2015 – it would’ve been easier to do so had Tresiba not had the very long regulatory tie up in the US as well). 
• Management highlighted the recent positive results from the SWITCH 2 trial demonstrating hypoglycemia reductions with Tresiba (insulin degludec) vs. Lantus. The trial (n=721 patients with type 2 diabetes) demonstrated a significant 30% reduction in severe or blood glucose-confirmed symptomatic hypoglycemia, a significant 42% reduction in severe or symptomatic nocturnal hypoglycemia, and a significant 51% reduction in severe hypoglycemia with Tresiba vs. Lantus. The 46% reduction in severe hypoglycemia during the maintenance period (the second half of the 32-week treatment period with each drug) was not statistically significant, but management attributed this to the relatively low numbers of hypoglycemic events and stressed that the numerical reduction was similar to that observed during the entire treatment period. Management emphasized that SWITCH 2 and SWITCH 1 (a similar trial in type 1 diabetes that just reported this week) were designed specifically based on FDA feedback and that they should be able to support some sort of label update for Tresiba.

Xultophy

• Novo Nordisk is initiating a phase 3b trial to evaluate the durability of glycemic control with Xultophy (insulin degludec/liraglutide) vs. Sanofi’s Lantus (insulin glargine). The open-label DUAL VIII trial aims to enroll 1,000 patients with type 2 diabetes not at target on oral medications. Participants will be randomized to receive either Xultophy or Lantus once daily for 104 weeks in addition to their usual treatment; doses will be adjusted individually. The primary endpoint is time to inadequate glycemic control (presumably A1c >7%) and need for treatment intensification. Secondary endpoints include time to A1c >6.5%; change in A1c, weight, fasting plasma glucose, and nine-point SBMG profile after 26 weeks; insulin dose after 26 weeks; percentage of responders (A1c <7% and A1c <6.5%) after 26 weeks and 104 weeks; and number of severe or confirmed symptomatic hypoglycemic events after 26 weeks and 104 weeks. The trial is currently recruiting participants and is expected to complete in October 2018. We would guess that this trial was designed primarily with payers in mind, as it could support reimbursement of Xultophy for patients earlier in the disease progression. Given what we assume will be an expensive list price for Xultophy, it is conceivable that payers might otherwise limit coverage to patients who have already failed on basal insulin in an effort to reduce costs.
• Novo Nordisk expects an FDA Advisory Committee meeting for Xultophy and a decision in 3Q15. The product was submitted in September 2015, so this is consistent with a standard review timeline. We are a bit surprised that an AdComm would be required for a combination of two approved drugs, but Xultophy is technically the first product in its class to reach the FDA. We wonder if this means there will be an AdComm for Sanofi’s LixiLan (lixisenatide/insulin glargine) as well – we would assume so. LixiLan was submitted in December and is being reviewed under an expedited process, meaning the timeline for a decision should be similar to Xultophy.

NovoLog/NovoRapid

• NovoLog (NovoRapid; insulin aspart) sales rose 8% YOY in constant currencies (18% as reported) to DKK 5.7 billion (~$840 million) in 4Q15; full year sales rose 6% YOY in constant currencies (19% as reported) to DKK 20.7 billion (~$3.1 billion). Sequentially, sales rose 11% as reported in 4Q15. NovoLog did quite well in 2015, with double-digit sales growth each quarter in comparison to slight declines or single-digit growth in 2014. Full year 2015 growth of 19% was against an easy comparison of 4% growth in full year 2014. This is likely due in part to NovoLog’s more favorable formulary positioning in 2015 following the loss of a major Express Scripts contract to Lilly’s Humalog (insulin lispro) in 2014.

Figure 8: NovoLog/NovoRapid Sales (1Q12-4Q15)

• By region, growth was strongest in China, where NovoLog sales rose 15% in constant currencies (27% as reported) to DKK 214 million ($32 million) in 4Q15. In International Operations, sales rose 12% in constant currencies (8% as reported) to DKK 561 million (~$83 million). Sales rose 9% in constant currencies (25% as reported) in North America to DKK 3.5 billion (~$524 million) and rose 4% in constant currencies (5% as reported) in Europe to DKK 1.1 billion (~$166 million). In Japan/Korea, sales were flat in constant currencies (up 8% as reported) at DKK 240 million (~$35 million).

Other Insulins

• NovoMix sales rose 3% YOY in constant currencies (9% as reported) to DKK 2.8 billion (~$418 million) in 4Q15; full year sales rose 2% YOY in constant currencies (13% as reported) to DKK 11.1 billion (~$1.7 billion). Sequentially, sales rose 11% as reported in 4Q15 against an easy comparison, as sales fell 3% in 3Q15. By region, NovoMix sales were strongest in International Operations and China, where sales grew 16% and 10% in constant currencies, respectively (14% as reported in International Operations to DKK 661 million [~$98 million] and 21% as reported in China to DKK 728 million [~$108 million]). The higher demand for premixed insulin in these markets typically contributes to NovoMix’s stronger performance in these regions. Operation growth was in the low single-digits or negative in North America (1%, +14% as reported), Europe (-11%, +6% as reported), and Japan/Korea (-9%, +11% as reported).

Figure 9: NovoMix Sales (1Q12-4Q15)

• Human insulin sales fell 5% YOY in constant currencies (flat as reported) to DKK 2.8 billion (~$410 million) in 4Q15; full year sales fell 1% in constant currencies (rose 9% as reported) to DKK 11.2 billion ($1.7 billion). Sales were flat sequentially in 4Q15. The only region in which the portfolio experienced positive operational growth was International Operations; sales fell in North America, Europe, China, and Japan/Korea.

Figure 10: Human Insulin Sales (1Q12-4Q15)

Victoza

• Victoza (liraglutide) sales rose 10% in constant currencies (22% as reported) YOY to DKK 4.9 billion (~$724 million) in 4Q15; full year sales were up 18% operationally (34% as reported) to DKK 18.0 billion (~$2.7 billion). Sequentially, sales rose 5%. Management highlighted Victoza as a main driver of growth for the diabetes portfolio and the company overall. Victoza’s strong performance was largely attributed to underlying volume growth of the GLP-1 agonist class rather than gain in market share. IMS Health data included in Novo Nordisk’s presentation slides showed US volume growth for the class at 25% by the end of 2015, a strong recovery from the ~8% growth at the beginning of 2015 and a ~5% increase in growth from the last update. This brings class growth back in line with the 20%-25% growth seen in early 2013 and we expect that only to continue when basal insulin/GLP-1 combinations are introduced in the US. Outside of the US, notably, management highlighted Victoza’s role in driving growth in Europe (offsetting weakness in insulin) and in Japan and Korea.
• Victoza appears to be losing some market share in the US by volume, though it remains the clear leader in this growing field. IMS Health data showed that, as of December 2015, Victoza held 56% of the US GLP-1 agonist market by total prescriptions (TRx). For comparison, AZ’s Byetta (exenatide twice daily) and Bydureon (exenatide once-weekly) held a combined 24% of the market, Lilly/BI’s Trulicity (dulaglutide) held 11%, and GSK’s Tanzeum (albiglutide) held 8%. Market share for both Victoza and exenatide appears to be declining, while Trulicity’s is on a clear upward swing and Tanzeum’s is plateauing. As we noted in the intro, 4Q15 Lilly results showed Trulicity growth as representing nearly half of Lilly’s quarterly growth – very impressive given that Trulicity represents only 7% of its revenue. As a sidenote, our sense is that patients and providers both really ilke the delivery system, which we loved when we first saw – the  prestigious design firm IDEO built this with Lilly as we understand it. For more on GLP-1 preferences and market, contact Richard.wood@d-qa.com as he surveys many hundreds of patients quarterly who are on GLP-1.  For comparison, a similar graph showed Victoza holding 66% of the market share by TRx in 4Q14. Trulicity appears to have finally pulled ahead of Tanzeum in terms of market share by TRx. The two products were neck-and-neck in 2Q15 and 3Q15, though Trulicity’s sales already tripled Tanzeum’s in 4Q14 – its first quarter on the market – due to GSK’s decision to price Tanzeum at a steep wholesale discount. 4Q15 sales of Trulicity reached $113 million ($249 million for full year 2015). While that’s dwarfed by Victoza’s sales, Trulicity is still has only been on the market for a little over a year and its launch trajectory seems very strong. The GLP-1 agonist market will only get more crowded as more entrants with greater efficacy or innovations in dose timing or delivery device (like Trulicity’s) arrive on the market in the near-term. Novo Nordisk management stated that the company will continue to invest in Victoza’s commercialization “to create the strongest possible platform for when we launch semaglutide.” In addition to Novo Nordisk’s own phase 3 injectable and oral formulations of semaglutide, we expect to see Intarcia’s implantable exenatide mini-pump ITCA 650 submitted for FDA approval in the first half of this year. Sanofi/Hanmi have efpeglenatide, an ultra-long-acting GLP-1 agonist, in phase 2b, which could support up to once-monthly dosing.

Table 7: GLP-1 Agonist Market Share

Figure 11: GLP-1 agonist market share by TRx

• 4Q15 growth for Victoza was noticeably lower than in previous quarters in 2015 – part of this reflected that it was a very challenging comparison. Operational growth for the first three quarters of the year ranged from 18% to 25% (36%-47% as reported), compared to 10% (22% as reported) in 4Q15. In constant currencies, this growth is even lower than Victoza’s sluggish 1Q14 earnings (though, admittedly, from a much higher base). When pressed on this in Q&A, management noted that a rebate adjustment in the US during the quarter made the sales for the quarter look weaker than they actually are and emphasized not putting too much stock in quarterly sales numbers. Management was also quick to point out that the company has not seen any major impact from competing products in Europe. We’re not too worried about Victoza and believe GLP-1 will be a major growth engine for Novo Nordisk for years to come.
• The GLP-1 agonist class as a whole grew 24% YOY in 4Q15 to $1.1 billion and grew 22% YOY in 2015 to $3.95 billion. By value, Victoza led the market with 66% of the market share in 4Q15. AZ’s exenatide franchise (Bydureon and Byetta) held 21%, Trulicity held 10%, Tanzeum held 2%, and Sanofi’s Lyxumia (lixisenatide) held a mere 1% of the market (though this is partly a reflection of the product not yet being available in the US). The breakdown for full-year 2015 was similar: market share was 68% for Victoza, 23% for the exenatide franchise, 6% for Trulicity, and 2% for Tanzeum.

Figure 12: GLP-1 Agonist Sales (1Q06-4Q15)

• Topline results from the LEADER CVOT for Victoza (liraglutide) will be announced in March. The trial completed in December 2015 and we expect that the full results will be presented at ADA in June. There has been much speculation about the potential for LEADER to demonstrate a cardiovascular benefit, particularly following the positive EMPA-REG OUTCOME results for Lilly/BI’s Jardiance (empagliflozin). Novo Nordisk has expressed cautious optimism on this subject in the past, citing GLP-1 agonists’ potential effects on the underlying mechanisms of atherosclerotic disease and LEADER’s greater individual patient exposure compared to trials of other agents in the class. However, the company has also repeatedly emphasized that the trial is only designed to show non-inferiority, so we are keeping our expectations in check as always. Analysts out there are betting that there is perhaps a 10% chance it shows cardioprotection.
• During Q&A, Chief Scientific Officer Dr. Mads Thomsen offered interesting commentary on the potential for different types of cardiovascular benefits with Victoza vs. Jardiance. He argued that based on Victoza’s effects on vascular physiology and preliminary evidence from small studies, it is more likely to act as a cardioprotective agent earlier in the progression of type 2 diabetes, or even at the obesity/prediabetes stage. On the other hand, Jardiance’s volume depletion effects may be most beneficial in patients with more advanced disease, particularly those who already have heart failure. He suggested that any signal of benefit in LEADER could therefore support the use of Victoza earlier in the type 2 diabetes treatment algorithm than is typical today. However, Dr. Thomsen also noted that LEADER enrolled a sicker population similar to that in EMPA-REG OUTCOME, meaning any earlier preventive effects may not become apparent.

Saxenda

• Saxenda (liraglutide 3.0 mg) now leads the branded obesity pharmacotherapy field with ~31% market share by value, according to Novo Nordisk. This is likely due in part to its comparatively higher cost, as Novo Nordisk chose to price Saxenda linearly to Victoza at $1,068 per month. The high cost has also likely contributed to the sluggish uptake as reimbursement for obesity medications is relatively low in general. That said, management shared that Saxenda has contracted with a number of undisclosed pharmacy benefits managers and is growing its share within each formulary (that is not too hard to say since in all cases this is from a low or nonexistent base). The company also noted that the obesity market experiences a seasonality in which prescription volumes are generally higher in the first half of each year. In terms of total prescriptions (TRx), Saxenda is on an upward trend but still lags behind other prescription obesity medications with only 11% of total prescriptions. For comparison, Orexigen/Takeda’s Contrave (naltrexone/bupropion extended-release) holds the majority of the TRx volume, followed by Vivus’ Qsymia (phentermine/topiramate), and Eisai/Arena’s Belviq (lorcaserin). Novo Nordisk also shared that Saxenda has been launched in Canada, Denmark, and Italy in addition to the US.
• Novo Nordisk has not yet broken out Saxenda numbers. Saxenda sales were grouped with NovoNorm and needles under “other diabetes and obesity care.” As a whole, the group’s sales totaled DKK 1.2 billion (~$183 million) in 4Q15 and DKK 4.7 billion (~$703 million). We hope to see specific sales broken out in future updates – we suspect this will happen if and when the product becomes a somewhat larger contributor to growth (right now it’s probably well less than 1% of growth).

Figure 13: Saxenda market share

Pipeline Highlights

Faster-Acting Insulin Aspart

• Novo Nordisk highlighted the recent EU and US regulatory submissions for faster-acting insulin aspart (Faster aspart) for both type 1 and type 2 diabetes. Assuming a standard ~10-month review cycle, an FDA decision should arrive in October 2016. The filing is supported by results from the phase 3 Onset 1 (type 1 diabetes; n=1,143) and Onset 2 (type 2 diabetes; n=689) trials demonstrating equivalent A1c reductions and superior postprandial control with Faster aspart vs. NovoLog (insulin aspart). In Onset 1, Faster aspart produced significantly greater A1c reductions vs. NovoLog when dosed at mealtime and comparable reductions to pre-meal NovoLog when dosed 20 minutes after a meal. Overall hypoglycemia rates were comparable with the two products, but Faster aspart did cause more hypoglycemia in the postprandial period. In its 3Q15 update, Novo Nordisk reported that findings from an additional 26-week treatment period in Onset 1 were consistent with the initial results.

Semaglutide

• Novo Nordisk believes the available phase 3 results for once-weekly semaglutide support a potential best-in-class profile. Management highlighted the recent results from SUSTAIN 2 vs. Merck’s Januvia (sitagliptin) and SUSTAIN 4 vs. Lantus and noted that semaglutide has demonstrated a significant benefit on A1c and weight against every comparator studied thus far. The company noted that the remaining SUSTAIN trials, SUSTAIN 5 (vs. placebo as an add-on to basal insulin and/or metformin) and SUSTAIN 6 (CVOT), will report within the next three months; SUSTAIN 5 reported just this week with similarly positive results. According to ClinicalTrials.gov, SUSTAIN 6 has now completed.

Table 8: SUSTAIN phase 3 trial program for semaglutide

• Novo Nordisk also announced that it plans to initiate a phase 2 trial of semaglutide in NASH in 2H16. NASH has attracted a great deal of interest from pharmaceutical companies in recent years, for good reason – its prevalence is rising along with obesity rates and there are currently no approved treatments. This trial is being added to what is already quite a comprehensive development program for semaglutide: in addition to the six SUSTAIN trials, Novo Nordisk is conducting two phase 2 trials of semaglutide dosed once daily and planning a ten-trial phase 3 program for an oral formulation.

Oral GLP-1 Agonists and Insulins

• Novo Nordisk shared that it has discontinued development of its two phase 1 oral formulations of liraglutide, OG987GT (NN9926) and OG987SC (NN9927). This follows the advancement of its oral semaglutide formulation into phase 3 in August. This decision is not surprising, as we had not heard any updates on the plans for either candidate in quite some time and semaglutide appears to have superior efficacy compared to liraglutide as an injectable formulation.
• Novo Nordisk did not offer any commentary on its PIONEER phase 3 program for oral semaglutide. We got our first detailed look at the PIONEER program at the company’s November Capital Markets Day. The program will include 10 trials involving >9,300 patients that will all be conducted over the next two years. As of now, PIONEER 3 (oral semaglutide vs. Merck’s Januvia [sitagliptin]) is the only trial posted on ClinicalTrials.gov; it is currently recruiting participants for a February 2016 start date and an estimated completion date of August 2018. This trial and PIONEER 2 (vs. Lilly/BI’s Jardiance [empagliflozin]) suggest that Novo Nordisk may be looking to position its oral GLP-1 agonist earlier in the treatment algorithm than current injectable GLP-1 agonists, as an alternative to DPP-4 inhibitors and SGLT-2 inhibitors.

Table 9: PIONEER phase 3 trial program for oral semaglutide

• The company has completed a phase 2a trial of its primary oral insulin candidate OI338GT (NN1953). A phase 1 trial of a newer oral insulin candidate (OI320GT/NN1957; added to the pipeline in 3Q15) is also complete. Novo Nordisk did not offer any additional commentary on these candidates in its 4Q15 update and we’re curious to see if the company plants to advance both candidates all the way through the pipeline or if the new formulation could eventually supersede the more advanced candidate.

Other Pipeline Candidates

• Novo Nordisk has initiated a phase 1 trial of the liver-preferential rapid-acting insulin NN1406. We first learned about this candidate at the company’s November Capital Markets Day, when management indicated that phase 1 was expected to begin before the end of 2015. The double-blind trial aims to enroll 48 male patients with type 1 diabetes who will be randomized to receive a single dose of NN1406 or NovoLog. The study is primarily intended to evaluate safety and will also assess pharmacokinetics (PK) and pharmacodynamics (PD) over 24 hours. It is currently recruiting and expected to complete in May 2016. NN1406 is designed to more closely mimic physiologic insulin by increasing distribution to the liver relative to the periphery, hopefully producing weight and hypoglycemia benefits. We will be watching closely for any safety issues similar to those that derailed Lilly’s once-promising liver-selective basal insulin peglispro.
• Novo Nordisk confirmed that it has initiated a phase 2 trial of anti-IL-21 antibody and liraglutide for beta cell preservation in newly-diagnosed type 1 diabetes. We first caught wind of plans for this trial in May 2015. According to ClinicalTrials.gov, the 304-participant trial is expected to complete in November 2018. We’re glad to see Novo Nordisk’s continued commitment to investigating GLP-1 agonist-related therapies for type 1 diabetes, especially in light of the lackluster results from its phase 3 ADJUNCT ONE and ADJUNCT TWO trials of liraglutide in type 1 diabetes. The company also confirmed that it has terminated its phase 3 LATIN T1D development program for liraglutide in type 1 diabetes in this update.
• While not mentioned on the call, we noticed that Novo Nordisk has added a GLP-1/GIP dual agonist (NN9709) to its phase 2 pipeline. Novo Nordisk acquired Dr. Richard DiMarchi’s (Indiana University, Bloomington, IN) companies Calibrium and MB2 in September 2015. The acquisition presumably continues Dr. DiMarchi’s work in dual and triple agonists. It’s possible that NN9709 is the same as Roche’s phase 2 GLP-1/GIP dual agonist RG7697, which was also developed with Dr. DiMarchi. Roche had removed the candidate from its pipeline in 2Q15, though the company was committed to completing the trial and Novo Nordisk had stated that it would evaluate the compound once the trial is complete.
• Novo Nordisk is also investigating its phase 1 gut hormone PYY 1562 (NN9748) for type 2 diabetes. The company added the same candidate to its phase 1 pipeline for obesity in 3Q15 (with the identifier NN9747). The company’s early-stage obesity pipeline also includes the glucagon analog G530L (NN9030) and the long-acting amylin analog NN9838. A once-weekly basal insulin analog (LAI287/NN1436) remains in phase 1 as well.

Select Questions and Answers

Q: On LEADER, what are the commercial implications for you if you were to show a CV benefit versus not showing a benefit given the EMPA-REG data out there?

A: What we have seen so far is that in earlier-stage diabetes and in obesity and prediabetes we have seen a hazard ratio that has consistently been below one for Victoza, albeit the numbers are very small. However, with that said, the patients being studied here are significantly more atherosclerotic. They are more reminiscent of the population that we saw in the EMPA-REG study. That is a population where many of them have emerging heart failure. Many of them have had heart attacks previously. That basically means that the effect of empagliflozin – which is basically reduction of volume overload – that occurs in such patients is something that is not necessarily likely to happen in an earlier stage population.

Whereas the way we believe GLP-1 agonists may work (if they work at all) in a long-term study like LEADER, is more based on the pathophysiology of the vessel wall disease. By that I mean that the earlier we treat, the more effect we would expect to see in terms of benefit on cardiovascular performance. So my argument would be that if we can see something in LEADER, that would be a rationale for more upstream use of liraglutide. We have to bear in mind that it is today used very often as a third or even sometimes fourth line of treatment.

Q: You had to give a slightly high rebate for Victoza in 4Q15 in the US. Is that because you do see some impact from payers trying to play you against Lilly with Trulicity? Also, what is happening in the European franchise? Are there any trends we should be aware of? Or is it just quarterly swings we see on Victoza?

A: Yeah, the numbers for 4Q15 were weak, in particular in the United States. This is of course is why we continuously try to notify the market to be a little bit cautious about reading too much into quarterly numbers, because we did do a rebate adjustment for Victoza in the US, relating to sales from previous quarters in the US. That negatively impacted the 4Q15 numbers to make them weaker than they are in reality. There were no major changes in Europe. There was strong performance of our franchise in Europe and no major impact from competing products.

Tresiba

Q: On Treisba, can you give us any earlier data on the formulary access that you've achieved in the commercial and the Part D setting and any updates regarding Treisba in Germany?

A: We have achieved the planned formulary access in the commercial markets. We also have, slightly earlier than anticipated, received some access in Part D as well. If you look at it from that perspective, we are slightly ahead of our expectations when it comes to formulary access.

For Tresiba in Germany, our negotiations did not pan out unfortunately. We had small hopes based on an opening by the German authorities to revisit the pricing discussion. But at the end of the day, they came back with basically the same offer that we were offered initially. Basically, we would have had to accept the pricing based on human insulin, which was not really something that we could accept. Therefore, we have started ceasing the distribution and helping the authorities and the medics to transfer patients to the best of our ability to minimize harm for them.

Q: On SWITCH, if we looked at the ADA definitions of hypoglycemia, would the hypoglycemia benefit be equally compelling?

On the condition of hypoglycemia, the FDA actually had four major points for us to consider in the construction of the new SWITCH trials, one being the timing of the dosing, i.e. morning versus evening. This we addressed by randomizing 50/50 morning and evening for both products in a blinded way. Another point was the hypos have be not just be measurable but also symptomatic, etc. and here we have actually only measured symptomatic 56 milligrams per deciliter and the severe hypoglycemic episodes were actually adjudicated by a committee. There was also a notion that the population was perceived to be unreasonable or not ideal because we had weeded out those at highest risk of hypoglycemia. This did not take place in this trial as we looked into different ways of being at risk of hypoglycemia, either of having a hypo in 12 weeks, a severe hypo in 12 months, have had insulin treatment for more than five days or moderate renal insufficiency. All of these patient sub-populations were part of the trial. So I think it's difficult to find any flaws, to be quite frank, and all hypos patients are indeed symptomatic.

Q. What should we make of the lack of benefits seen in the maintenance phase of the trial?

A: We mentioned the maintenance period for both the primary and confirmatory secondary endpoints and found statistically significant and clinically meaningful reductions in both severe and blood glucose confirmed symptomatic hypoglycemia as well as a similar measurement just for nocturnal hypoglycemia.

The issue with the severe hypoglycemia in the maintenance period is not that there was not a reduction, because it was actually a 46% reduction in the event rate. However, the numbers were low because you're only measuring in the 16-week maintenance period. Fortunately, there was a pre-specified other analysis that measured severe hypoglycemia throughout the period. When that is done, the reduction percent-wise is similar – 51% versus 46%. But here the numbers naturally accrued are larger and then it’s clearly a statistically significant benefit in the occurrence of severe hypoglycemia.

Big Picture

Q: One way to read the long-term target is that you will also invest quite significantly in R&D. Could I just ask you to contextualize this a little bit, given that you have the SUSTAIN program rolling off and you've got DEVOTE coming to an end. In terms of numbers of patients in phase 3 trials, I would have thought it would help you. Yet you seem to suggest that your R&D budget is going up. So why and where?

A: With the completion of three cardiovascular trials, LEADER, DEVOTE and SUSTAIN 6, we are coming to the end of a 25,000 patient commitment, you can argue. However, that has to be countered by a 9,300 patient commitment in the PIONEER program. Also, do bear in mind that at least this year, the shared workload associated with completing all the CV outcome trials and initiating a huge phase 3 program for oral semaglutide is keeping everybody busy.

One important thing to bear in mind is that, while we do not have the same heavy load coming from phase 2 into phase 3 in the next coming couple of years, we on the other hand have a huge load of early marketed assets, whether it's the three degludec products or whether it's the future semaglutide or Saxenda and whatnot. All of them deserve to have fully professional and dedicated phase 3b programs and even emerging real world evidence surrounding those assets. That means that you will see a relative increase in the proportion of development cost associated with the protection and expansion of the business opportunity for these five, six, seven new products in the diabetes and metabolism studies.

I think one should also assume that going forward our clinical programs will be more complex. We need to include in our global clinical program a regional input to take into consideration the market access challenges we have in different markets and also there is an increasing need for real world data to demonstrate the value of the drugs to get them on the market as quickly as possible. So, all other things being equal, I think it's going to expand the cost to develop new drugs, especially in areas like ours where the effect is only accrued over time, and therefore monitoring the benefit will require more patient years than we've had in the past.

Q: On the selling, general, and administrative expenses expansion – if that line was to grow in line with sales over the next four to five years, that's an additional DKK 1.5 billion in absolute spend. Is that about right? Where do you deploy that?

A: Our selling and distribution costs will increase in line with the growth of sales. As for where we will spend it, we may need to expand our sales force in the United States. We have currently employed almost our entire sales force on Tresiba. We're still supporting Victoza, which is a successful brand. We would like to continue to push Victoza in the US to create the strongest possible platform for when we launch semaglutide. We will hopefully be in a position to launch Xultophy later this year. We are also hoping for faster aspart to be approved. We have Ryzodeg on the shelf, to be launched at an appropriate point in time. So I think there will be requirements for further adjustments of our sales force. This could be in certain markets outside the US but also in US markets. We'll also have the traditional expansion of our international organization in the emerging markets.

Q: We know that 2016 and 2017 are going to be huge investment years and that US pricing dynamics are going to be more challenging. That said, your product mix is going to improve. It seems to me that over the long term, if you're growing, there's going to be scope for further margin expansion. Is there some sensitivity about communicating that given the current US political backdrop on drug pricing or am I missing something?

A: If you look to our accounts that will be released on the annual report, you will actually see that the rebate percentage of our US sales have gone up to 56% and that, to me, is a sign that while list prices may be adjusted, the majority of them are actually being offset by increased rebate levels. So overall, not anticipating any very significant impact on the margin.

Q: You've given some color on the double-digit diabetes growth embedded in the long-term outlook, but I wonder if you could touch on some of the other key variables that you see that influence that outlook? When do you expect an inflection from China?

A: The current suppression of growth in China is partly due to the ongoing pricing reform, so we expect this to unfold into 2017. We probably need to go through that before we could potentially see return. We see a good volume growth. We also need to reposition ourselves vis-à-vis our position in the basal insulin segment. To that end, we filed Tresiba last year in August. So you could also say that before we get that approved and into the market, the impact will not be visible. I wouldn't want to predict an exact timing of it, but we're probably going to see suppressed growth in 2016 and potentially also into 2017.

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