Memorandum

Novo Nordisk 4Q12 – Diabetes Care up 21% in FY2012; Victoza sales up 58%; once- weekly insulin enters phase 1; Tresiba to launch in EU in 1H13 – January 31, 2013

Executive Highlights

  • Global Diabetes Care sales in 2012 rise 21% to DKK 61 billion ($11 billion); sales in 4Q12 total DKK 16 billion ($3 billion), up 18% as reported.
  • Victoza 2012 sales totaled DKK 9.5 million ($1.6 billion), up 58% as reported and 50% operationally; 4Q12 sales totaled DKK 2.7 billion ($471 million), up 29% as reported.
  • Several R&D updates: EU Tresiba launches to begin in 1H13; potential once-weekly insulin LAI287 entered phase 1; phase 3a for Victoza in type 1 diabetes to start in 2H13; IDegLira mid-2013 EU filing.

This morning, Novo Nordisk reported full-year 2012 and 4Q12 results in a call led by CEO Lars Sorenson. Global Diabetes Care revenue for full-year 2012 totaled DKK 61 billion ($11 billion), rising an impressive 21% from FY2011 and 15% operationally. Notably, Novo Nordisk broke up product sales by region for the first time (information below) – this helped us understand how strong North American sales and sales in China are. Sales of modern insulins in 2012 experienced robust growth (up 21% from FY2011) to reach DKK 35 billion ($6.0 billion), again driven by the strong growth of Levemir (27%) and NovoRapid (23%). In 4Q12, global Diabetes Care revenue rose 18% year-on-year to total DKK 16 billion ($3 billion), driven by 24% growth in North America. Sales of modern insulins in 4Q12 rose 20% year- on-year to DKK 9.5 billion ($1.6 billion), with NovoRapid, NovoMix, and Levemir achieving their highest quarterly revenue to date. Victoza achieved blockbuster status for the second year in a row, with full-year 2012 sales totaling DKK 9.5 million ($1.6 billion), up 58% on a reported basis, and also exceeding $1.0 billion in the US alone. Victoza 4Q12 sales reached DKK 2.7 billion ($470 million), up 29% from 4Q11. Sales continue to climb despite Victoza’s already high base. Notably, management said that it expects competitive pressure (i.e., Bydureon) to contribute to slower growth in 2013, a departure from its 3Q12 comments that Bydureon did not represent a substantial threat.

Management remarked that the FDA has not informed Novo Nordisk of when it will finish its regulatory review of Tresiba (insulin degludec) and Ryzodeg (insulin degludec/aspart); however, in Q&A management expressed confidence that the drugs would ultimately be approved following the positive vote during the November FDA Advisory Committee meeting. Novo Nordisk expects to launchTresiba in Denmark and the UK in 1H13 and in other European markets throughout the rest of 2013 and 2014, with Ryzodeg following about one year after. Launch of Tresiba in Japan will occur immediately following price negotiations, which are expected to conclude in 1Q13. Management also commented on its pricing strategy for Tresiba in Europe, noting that Novo Nordisk will likely be to opt for a relatively high premium to ensure a decent price overseas and in emerging markets.

Management provided several key R&D updates, noting that it just advanced a novel long-acting insulin (LAI287 ;NN1436) with the potential for once-weekly (!) dosing into phase 1. It’s certainly still early days, but a once-weekly insulin could be revolutionary for adherence if it is able to maintain reasonable hypoglycemia and weight profiles. Additionally, management reiterated that its ultra-fast- acting formulation of insulin aspart (FIAsp; NN1218) will enter phase 3 in late 2013 (we think it could be a bit sooner). Topline results from a pharmacology study for Victoza as an adjunct to insulin therapy in type 1 diabetes demonstrated that it decreased patients’ daily insulin requirement and did not compromise the glucagon response during hypoglycemia. Novo Nordisk also continues to investigate Victoza as an obesity medication, with the remaining three studies in the phase 3 SCALE program scheduled to finish in January, March, and June 2013. As previously announced, IDegLira completed its phase 3a DUAL program, and management forecasted a mid-2013 regulatory filing in Europe, as well as a US submission “as soon as possible” pending Tresiba’s US approval. No updates were provided on the other GLP-1 candidates, including the phase-3-ready once-weekly semaglutide.

FINANCIAL UPDATES – DIABETES CARE

  • For full-year 2012, worldwide Diabetes Care revenue totaled DKK 61 billion ($11 billion), growing an impressive 21% on a reported basis from DKK 50 billion ($9 billion) in FY2011. This was a relatively easy comparison given 10% growth in 2011, which followed a high growth rate of 22% in 2010. Management attributed 2012’s strong growth to increased sales of modern insulins (accounting for 55% of growth in local currencies; up from 41% in FY2011) and Victoza (making up 39% of growth; down from 55% in FY2011). Diabetes Care accounted for a 95% share of total Novo Nordisk growth, compared to an 83% share in FY2011. Notably, for the first time, Novo Nordisk broke out Diabetes Care sales by region (see table below).
  • Worldwide Diabetes Care revenue in 4Q12 totaled DKK 16 billion ($3.0 billion), rising 18% on a reported basis from DKK 14 billion ($2.5 billion) in 4Q11. This growth is on par with the 14% year-on-year growth observed in 4Q11, but lower than the 28% year-on- year growth in 4Q10. Sequentially, sales increased 6% from DKK 15.5 billion in 3Q12, on par with the 4% sequential growth observed in 3Q12 and 8% growth in 2Q12, and higher than the zero sequential growth in 1Q12. Diabetes Care sales in 4Q12 by region are reported in the table below.

Table 1: 2012 and 4Q12 Diabetes Care revenue by region

Diabetes Care Market

2012 Revenue in billions

Operational Growth from 2011

4Q12 Revenue in billions

Worldwide

DKK 60.9 ($10.5)

15%

DKK 16.5 ($2.9)

North America

DKK 26.7 ($4.6)

24%

DKK 7.6 ($1.3)

Europe

DKK 15.1 ($2.6)

3%

DKK 4.0 ($0.70)

International Operations

DKK 8.5 ($1.5)

20%

DKK 2.2 ($0.39)

Region China

DKK 6.2 ($1.1)

16%

DKK 1.5 ($0.27)

Japan & Korea

DKK 4.3 ($0.74)

-3%

DKK 1.1 ($0.19)

*The above assumes an exchange rate of 1 DKK to 0.1727 USD for full-year 2012 revenue and an exchange rate of 1 DKK to 0.1738 USD for 4Q12 revenue.

  • According to IMS moving annual total (MAT) data, the value of the diabetes care market totaled roughly DKK 225 billion (~$39 billion) as of November 2012. Over the past ten years, the global diabetes care market has grown at a compound annual growth rate (CAGR) of 9.9%, driven strongly by the injectables segment, which grew 15.1% (13.8% for insulin therapies), compared to the 5.9% growth for oral antidiabetic drugs. As of November 2012, Novo Nordisk continues to hold a leading 26% value share of the global diabetes market, up slightly from its shares of 25% in 3Q12, 24% in 4Q11, and 23% in 2Q10 – Novo Nordisk’s share has remained relatively stable (~23%-25%) over the past five years. For comparison, Sanofi trails Novo Nordisk with a ~17% share (up from ~7% in August 20o2), followed by Merck with a ~15% share (which has risen steeply from zero since Januvia’s launch) and Lilly with a ~12% share (down from ~17% in 2002). Interestingly, Takeda’s share has declined steeply in recent years, as illustrated by the drop from an 11% value share as of August 2012 to a 7-8% share as of November 2012. The decline appears attributable to the safety concerns surrounding pioglitazone, as well as the drug’s recent genericization; however, Takeda’s market share may increase following the January FDA approval of three drugs: 1) the DPP-4 inhibitor alogliptin (Nesina); 2) the alogliptin/metformin fixed-dose combination (Kazano); and 3) the alogliptin/pioglitazone fixed- dose combination (Oseni; the first DPP-4/TZD combination product approved in the US).
  • For full-year 2012, sales of modern insulins totaled DKK 35 billion ($6.0 billion), up 21% on a reported basis from FY2011 and up 15% operationally. This was a relatively easy comparison, as 2011 was the first year within the past six years in which modern insulins did not achieve an annual sales growth of 20% or more (growing only 8% from 2010). North America accounted for more than half of modern insulin growth, followed by International Operations and region China. NovoRapid contributed the majority of 2012 modern insulin sales with revenue of DKK 16 billion ($2.7 billion), followed by Levemir with DKK 10 billion ($1.7 billion) and Novomix with DKK 9 billion ($1.6 billion). Levemir’s reported growth of 27% was particularly strong, a fact management highlighted during the call. As noted above, Novo Nordisk reported product sales by region for the first time during the 4Q12 update (see Appendix A).

  • Sales of modern insulins continued to rise in 4Q12, with NovoRapid, NovoMix, and Levemir achieving their highest quarterly revenue to date (figures in table below). According to our model, sales of NovoRapid increased 23% from 4Q11 and 6% sequentially to reach DKK 4.3 billion ($750 million). For comparison, sales of Lilly’s Humalog decreased 7% year-on-year but increased 7% sequentially to $616 million (the year-on-year decline was driven by a 19% drop in US revenue, largely attributable to increased rebates, as well as Humalog’s removal from the CVS Caremark formulary). Sanofi has not reported sales for Apidra or Lantus, as its 4Q12 earnings update is scheduled for February 7. According to IMS MAT data (given by Novo Nordisk), NovoRapid has a 50% brand share of the fast-acting insulin market as of November 2012. NovoMix sales grew 14% year-on-year and 4% sequentially; according to IMS MAT data from November 2012, the brand holds a 44% share of the premixed insulin segment. Sales of Levemir rose an impressive 29% year-over-year and 10% sequentially to reach DKK 2.7 billion ($470 million), though Levemir only holds a 20% share of the long-acting insulin segment as of November 2012, due to the market dominance of Lantus. Since it’s launch, Levemir has failed to attract significant patient/HCP uptake like Lantus, and according to our model, its value share appears to be leveling off (8% in 2006, 12% in 2007, 16% in 2008, 17% in 2009, 19% in2010, and 19% in 2011).

Table 2: Sales of modern insulin by product in 4Q12 and full-year 2012.

Product

2012 Revenue in billions

Reported Growth from FY2011

4Q12 Revenue in billions

Reported Growth from 4Q11/3Q12

Modern Insulins

DKK 35 ($6.0)

21%

DKK 9.5 ($1.6)

20% / 7%

NovoRapid

DKK 16 ($2.7)

23%

DKK 4.3 ($0.75)

23% / 6%

NovoMix

DKK 9.3 ($1.6)

13%

DKK 2.5 ($0.43)

14% / 4%

Levemir

DKK 9.8 ($1.7)

27%

DKK 2.7 ($0.47)

29% / 10%

*The above assumes an exchange rate of 1 DKK to 0.1727 USD for full-year 2012 revenue and an exchange rate of 1 DKK to 0.1738 USD for 4Q12 revenue. 

  • In 2012, Novo Nordisk reported <10% growth for human insulins (5%), protein- related products (9%), and oral antidiabetic products (NovoNorm, Prandin, and PrandiMet; 7%). Sales figures for all three products over 4Q12 and FY2012 are reported in the table below. As a reminder, during Novo Nordisk’s 3Q12 update, management attributed thepositive growth in human insulin sales to a slowdown in the conversion from human insulins to modern insulins, primarily driven by a reduction in forced discontinuations of human insulins. Management also acknowledged a push toward the use of human insulins in Europe and expected this trend to influence the competitive landscape, as well as the future pricing of modern insulins and novel insulins. No further comments were given during the 4Q12 update.

Table 3: 2o12 and 4Q12 sales of human insulins, protein-related products, and oral antidiabetic products.

Product

2012 Revenue in billions

Reported Growth from FY2011

4Q12 Revenue in billions

Reported Growth from 4Q11/3Q12

Human insulins

DKK11 ($2.0)

5%

DKK3.0 ($0.52)

8%/8%

Protein-related products

DKK2.5 ($0.43)

9%

DKK0.62 ($0.11)

9%/-4%

Oral antidiabetic products

DKK2.8 ($0.48)

7%

DKK 0.67 ($0.12)

3%/-7%

*The above assumes an exchange rate of 1 DKK to 0.1727 USD for full-year 2012 revenue and an exchange rate of 1 DKK to 0.1738 USD for 4Q12 revenue. 

  • 2013 Financial Guidance: In 2013, Novo Nordisk forecasts sales growth of 8-11% in local currencies (3.5%-6.5% on a reported basis). The company expects a positive contribution from sales of modern insulins, Victoza, and Tresiba (primarily in the US, EU, and Japan), countered by a negative impact from challenging price environments, generic competition to oral antidiabetic products, and greater competition in diabetes care and biopharmaceuticals. During Novo Nordisk’s 3Q12 report, management remarked that growth in 2013 would be strongly driven by markets in China and International Operations (growing at around 15%), as well as North America. Novo Nordisk expects 2013 operating profit growth of around 10% in local currencies (3% on a reported basis), which reflects the “significant costs” related to the expected launch of Tresiba, to the expanded US sales force, and to sales and marketing investments in China and select markets in Internationals Operations.*The above assumes an exchange rate of 1 DKK to 0.1727 USD for full-year 2012 revenue and an exchange rate of 1 DKK to 0.1738 USD for 4Q12 revenue.

THE GLOBAL INSULIN MARKET

  • Global Insulin Market – Value/Growth: According to IMS MAT data from November 2012, the value of the global insulin market totals DKK 111 billion (~$20 billion), and has grown at a compound annual growth rate of 16% by value from DKK 54 billion (~$10 billion) in 2007 – this growth reflects an 9% mix/price contribution and a 7% volume contribution (due to the increasing prevalence of diabetes and obesity, rising diagnosis and treatment rates, the intensification of insulins regimens, a growing number of obese and overweight patients, and an aging population). Currently, the growth of the global insulin growth remains around 5% by volume.
  • Global Insulin Market – Composition: According to IMS MAT data from November 2012, the global insulin market is roughly evenly divided between fasting-acting insulins (33% volume share; up from 32% in 2007), premixed insulins (29% share; down from 34% in 2007), and long- acting insulins (38%; up from 34% in 2007). Modern insulins make up 85% of the global insulin market (valued at DKK 94 billion [~$16 billion]), with human insulins making up the remaining share (valued at DKK 17 billion [~$3 billion]).
  • Company Market Share: Novo Nordisk remains the leader in the global insulin market with a 49% volume share of the total market and a 46% volume share of the modern insulin market.

Table 4: Insulin market by region

Region

Value of total insulin market (in billions)

Composition

B=basal; FA=fasting-acting; P=premixed

Share of Global Market (by volume)

Global

DKK111 ($20)

38% B / 33% FA / 29% P

--

North America

DKK69 ($12)

48% B / 37% FA / 14% P

35%

Europe

DKK25 ($4.4)

37% B / 39% FA / 24% P

37%

International Operations

DKK7.4 ($1.3)

28% B / 19% FA / 53% P

17%

Japan and Korea

DKK5.7 ($1.0)

30% B / 36% FA / 34% P

4%

China

DKK3.6 ($0.6)

21% B / 13% FA / 66% P

7%

From IMS MAT data as of November 2012

NOVO NORDISK’S INSULIN PERFORMANCE BY REGION

  • North America: Sales of modern insulins, human insulins, and protein-related products increased a reported 29% during full-year 2012, driven by the increased market penetration of modern insulins, as well as a “modest growth” in human insulin sales (this stands in contrast to Novo Nordisk’s 3Q12 update, when sales of human insulins negatively impacted growth). As of November 2012, Novo Nordisk possesses a 41% volume share of the US total insulin market (on par with 3Q12 and up from 40% in 4Q11), as well as a 38% volume share of the US modern insulin market (consistent with 3Q12 and up slightly from 36% in 4Q11). Sanofi remains the market leader of the US global insulin market (with a 44% volume share, consistent with 3Q12), while Lilly has an 18% share. In the US, the percentage of Novo Nordisk’s modern insulin volume sold via the FlexPen has remains at 50%, compared to 50% in 3Q12, ~48% in 2Q12, ~47% in 1Q12 and~44% in 2Q11, ~41% in 2Q10, ~39% in 2Q09, and >37% in 4Q08, >35% in 4Q07, >33% in 4Q06. We imagine the market still has room for greater pen penetration, as the use of pens is significantly higher in the modern insulin markets of Europe and China.
  • Europe: Sales of modern insulins, human insulins, and protein-related products in full-year 2012 remained unchanged relative to FY2011, again driven by revenue from NovoRapid and Levemir, partly countered by declining sales of human insulins (consistent with the 3Q12 update). As with the past several quarters, European sales have been negatively impacted by the continued low volume growth of insulins (<3%) and by pricing reforms across several European markets (not specified). As of November 2012, Novo Nordisk holds a 50% volume share of the total European insulin market (down slightly from 51% in both 3Q12 and 4Q11), as well as a 50% volume share of the modern European insulin market (consistent with 3Q12 and 4Q11). For comparison, Sanofi and Lilly have 31% and 18% volume shares of the EU modern insulin market, respectively (Novo Nordisk lost a 1 percentage point market share to Sanofi in 4Q12). The use of the devices in Europe (primarily NovoPen and FlexPen) remains high, accounting for 96% of Novo Nordisk’s insulin volume (on par with the 3Q12 update).
  • International Operations: Sales of modern insulins, human insulins, and protein- related products increased 19% on a reported basis in full-year 2012. Growth was again driven by modern insulins, with a “solid” contribution from human insulins. Novo Nordisk possesses 58% of the total insulin market in International Operations (flat with 3Q12 and 4Q11) and a 55% share of the modern insulins market in this region (flat with 3Q12 and down slightly from 56% 4Q11). In the modern insulin market, Lilly trails Novo Nordisk with an 18% volume share, followed by Sanofi (14%) and Biocon (3%; market shares have not changed since 3Q12). At this time, the use of devices accounts for roughly 58% of Novo Nordisk’s insulin volume in major private markets (not defined), consistent with 3Q12 and 2Q12.
  • Region China: In full-year 2012, sales of modern insulins, human insulins, and protein-related products increased a strong 27% (for comparison, growth was 29% in the first nine months of 2012 and 27% in the first half of 2102). Novo Nordisk met its guidance for 15% operational growth in 2012. Consistent with previous quarters, modern insulins were the main drivers of growth, while sales of human insulins “only grew modestly.” As a reminder, during the 3Q12 update, management mentioned increased competition in China, particularly in the human insulin market, and forecasted a negative impact from the pricing of modern insulins. IMS MAT data from November 2012 indicates that Novo Nordisk possesses a 60% volume share of the total insulin market (down from 61% in 3Q12 and 62% in 4Q11), as well as a 65% share of the modern insulin market (down from 66% in 3Q12 and 67% in 4Q11). Novo Nordisk remains the uncontested leader of the modern insulin market, and is trailed by Lilly (14% volume share), Shanghai Fosun (10%), Tonghua Dongbao (6%), and Sanofi (5%; market shares have not changed since 3Q12). The use of the devices (primarily NovoPen) remains high, accounting for 97% of Novo Nordisk’s insulin volume in China.
  • Japan and Korea: Sales of modern insulins, human insulins, and protein-related products increased by 1% on a reported basis in full-year 2012; for comparison, growth was 4% in the first three months of the year, suggesting declining sales (or a strengthening of the yen) in 4Q12. Consistent with 3Q12, management attributed the low growth to a challenging competitive environment and to volume decline in the Japanese insulin market. Novo Nordisk continues to be the market leader in Japan, with a 55% volume share of the total insulin market (down from 56% in 3Q12 and 59% in 4Q11), as well as a 51% share of the modern insulin market (flat with 3Q12 but down from 53% in 4Q11). For comparison, while Sanofi and Lilly have nearly equal shares of the modern insulins market (26% and 24%, respectively), five years ago Lilly led Sanofi by roughly 17 percentage points (~27% share vs. 10% share). Currently in Japan, 98% of Novo Nordisk’s insulin volume is sold via devices, primarily the FlexPen.

FINANCIAL UPDATES – VICTOZA

  • Victoza reached blockbuster status for the second year in a row with full-year 2012 sales of DKK 9.5 billion ($1.6 billion), up 58% as reported from DKK 6.0 billion ($1.2 billion) in 4Q11. While 58% year-on-year growth is still quite strong, it represents a slow down from the 120% growth observed in 2011. Naturally, the much higher base plays a role (as Victoza reached blockbuster status in 2011), but management also commented that “competitive pressure” (presumably Bydureon) would play a primary role in continuing to slow Victoza growth. This is a departure from the 3Q12 update, when management stated that it did not expect Bydureon to significantly affect market dynamics. Victoza provided 39% of Novo Nordisk’s growth in 2012, with modern insulins supplying much of the rest (55% of growth).
  • In 4Q12, Victoza sales reached DKK 2.7 billion ($470 million), up 29% as reported from DKK 2.1 billion ($380 million) in 4Q11, driven mostly by growth in North America and International Operations. While still quite robust, and from a very high base, this is the lowest quarterly year-on-year growth that Victoza has seen since its launch; we imagine that the likely future ramp-up of once-weekly Bydureon sales will only continue to slow Victoza’s growth (now that the transfer of international rights from Lilly to BMS/AZ is nearing completion). That said, Victoza’s brand recognition, greater efficacy, and ease of administration may still win many patients over despite Bydureon’s less frequent administration. Additionally, we’ll be interested in how Victoza sales will be influenced by the introduction of other once- weekly GLP-1 agonists. The once-weekly candidates closest to market are Sanofi’s Lyxumia and GSK’s albiglutide (both submitted to the FDA in January, meaning that Victoza could face more competition in 2014), Lilly’s dulaglutide (filing expected in 2013), and BMS/AZ’s once-weekly exenatide suspension (advanced into phase 3 in 4Q12). The GLP-1 market still has plenty of room to grow – as management has repeatedly highlighted, the entry of GLP-1 products to the market in the past has expanded the entire GLP-1 market in addition to creating more competition.
  • Table 5: Sales of Victoza by region

 

4Q12 Revenue in billions

Reported Growth from 3Q12

2012 Revenue in billions

Reported Growth from 2011

Worldwide Victoza Sales

DKK2.7 ($0.47)

8%

DKK9.5 ($1.6)

58%

North America

DKK1.7 ($0.29)

3%

DKK5.9 ($1.0)

60%

Europe

DKK0.71 ($0.12)

22%

DKK2.4 ($0.42)

50%

International Operations

DKK0.18 ($0.03)

21%

DKK0.61 ($0.11)

71%

Japan and Korea

DKK0.12 ($0.02)

-0.8%

DKK455 ($79)

52%

Region China

DKK0.021 ($0.04)

-16%#

DKK70 ($12)

---

*The above conversions assume an exchange rate of 1 DKK to 0.1738 USD for 4Q12 and a rate of 1 DKK to 0.1727 USD for 2012.

# This likely underestimates sequential growth in China (i.e., the real growth is likely less negative) given that Novo Nordisk only began reporting Victoza sales by geography for 4Q12, and our previous estimate of China 3Q12 sales was based on Novo Nordisk’s estimate that China made up 1% of total Victoza sales. Now, given actual geographical sales, it appears that 2012 China sales made up 0.7%, not fully 1%, of total 2012 Victoza sales, but we cannot determine how this may have fluctuated throughout the year.

  • Novo Nordisk provided an overview of Victoza’s global progress and, for the first time, reported revenue by region (see above table; in the past we had calculated regional revenue using Novo Nordisk’s estimates of percent of total Victoza sales by region). Revenue in 2012 and 4Q12 were driven mainly by sales in North America (62% of global sales for both 2012 and 4Q12) and Europe (26% for both 2012 and 4Q12), followed by International Operations (6% in 2012 and 7% in 4Q12), Japan and Korea (5% in both 2012 and 4Q12), and China (0.7% in 2012 and 0.8% in 4Q12). Management noted that Victoza has now been launched in 60 countries; since September, Victoza has launched in South Korea and two smaller, unspecified countries. According to IMS MAT data from November 2012, Victoza remains the global GLP-1 marketleader with a 68% value market share, compared to 58% in 2011 and 30% in 2010. GLP-1 agonists’ share of the total diabetes market has increased to 6%, compared to 4.5% in 2011 and 3.2% in 2010. We expect this share to continue to increase over time as newer, more convenient applications emerge.
    • Global GLP-1 market: According to IMS MAT data from November 2012, the global GLP-1 market increased at a compound annual growth rate of 34% in DKK over the past five years. The value of the global GLP-1 market totaled nearly DKK 14 billion ($2.4 billion) in November 2012, with North America constituting the majority of the market (69%), followed by Europe (24%), International Operations (4%), and Japan and Korea (4%).
    • North America accounted for 62% of Victoza sales in both 2012 and 4Q12 (see table above). For 2012, North American sales increased 60% as reported and 48% operationally. The GLP-1 class now represents 7.3% of the total US diabetes care market by value (up from 5.8% in 2011). Novo Nordisk noted that, despite competition from Bydureon, Victoza continues to be the driver of GLP-1 market growth in the US and continues to capture more of the US GLP-1 market– it now has a 62% share (compared to 60% in 3Q12, 58% in 2Q12, and 52% in 4Q11). We’ll be interested in seeing whether Victoza will keep increasing its US market share with the full BMS/AZ diabetes sales force behind Bydureon and Byetta. Similar to 3Q12, Victoza represents 1.3% of total prescriptions in the US diabetes care market, and roughly two-thirds of prescriptions are for the three-pen pack.
    • Europe accounted for 26% of Victoza sales in both 2012 and 4Q12 (see table above). Sales in Europe increased by 50% as reported and 48% operationally for 2012. Consistent with previous quarters, France, UK, Italy, and Spain were the primary drivers of European growth. In Europe, the GLP-1 segment has increased to 6.7% of the total diabetes care market (from 6.3% in 3Q12 and 5.0% in 2011), and Victoza is by far the GLP-1 market leader with a 76% share by value (up from 75% in 4Q12 and 68% in 2011). In the past, Victoza’s leading position in part reflected Lilly’s lower-than-expected effort in promoting Bydureon and Byetta overseas, and we will be watching whether the European GLP-1 market changes as Lilly transitions the international exenatide rights to BMS/AZ (according to Lilly, the transition is expected to complete by March 31, 2013).
    • International Operations accounted for 6% of 2012 and 7% of 4Q12 Victoza sales (see table above). Sales rose an impressive 90% as reported and 98% operationally for 2012. Novo Nordisk notes that the growth was driven by strong performances in Brazil and a number of Middle Eastern countries (unspecified), along with an easy comparison to 2011. GLP-1 now represents 3.0% of the total diabetes market in International Operations (flat with 3Q12 and up from 1.2% in 2011). Victoza holds an impressive 80% market share by value across International Operations (slightly down from the 81% share in 3Q12 and up from 60% in 2011).
    • Japan and Korea accounted for 5% of both 2012 and 4Q12 Victoza sales (see table above). Sales in this region increased 39% as reported and 29% operationally in 2012. The GLP-1 market now represents 2.3% of the total diabetes care market by value in Japan (up from 2.2% in 3Q12 and 1.6% in 2011). Victoza holds 77% of the market share in the Japanese GLP-1 market (down from 79% in 3Q12 and 87% in 2011). During the 3Q12 update, management noted that an unfortunate situation occurred during Victoza’s launch in which, due to wrong medical advice, patients experienced harmful medicaloutcomes when switching from insulin to Victoza. No update was provided on this front during the 4Q12 update.
    • China accounted for 0.7% of 2012 and 0.8% of 4Q12 Victoza sales. This was Victoza’s first full year on the Chinese market (following its launch in 4Q11), and Novo Nordisk notes, consistent with previous remarks, that early market feedback is positive and that hospital listings are developing satisfactorily. The GLP-1 class in China is still not reimbursed, but its small share of the total diabetes market is still expanding – it accounted for 0.5% of the diabetes market in 2012, compared to 0.4% in 3Q12 and 0.2% in 2011. Victoza still holds a minority market share in the Chinese GLP-1 market (44%) but this has increased from 32% in 3Q12 and 23% in 2Q12.

Table 6: Market share of the GLP-1 segment and Victoza, by region

  GLP-1 agonists' share of the total diabetes care market (by value) Victoza's share of the GLP-1 market (by value)
 

November 2012

November 2011

November 2012

November 2011

Global

6.0%

4.5%

68%

58%

US

7.3%

5.8%

62%

52%

Europe

6.7%

5.0%

76%

68%

International Operations*

3.0%

1.2%

80%

64%

Japan

2.3%

1.6%

77%

87%

China#

0.5%

0.2%

44%

4%

Data based on IMS November 2012 data

*Data for 12 selected markets representing approximately 60% of diabetes sales in the

#Data for mainland China, excluding Hong Kong and Taiwan

REGULATORY UPDATES – DEGLUDEC

  • Management stated that the FDA has not informed Novo Nordisk of when it will conclude its review for Tresiba (insulin degludec) and Ryzodeg (insulin degludec/aspart). As a reminder, a November 8, 2012 FDA Advisory Committee voted 8-4 to recommend approval of Tresiba and Ryzodeg with a post-approval cardiovascular outcomes trial. Management stated that it could not comment on a potential US label at this time. Our detailed report on the Advisory Committee meeting can be found at http://www.closeconcerns.com/knowledgebase/r/c53dd554. During Q&A, management affirmed that the FDA warning letter for the aseptic filling facility in Denmark (see “Financials” section above) is not directly linked to the Tresiba approval process, but did note that if the FDA is unsatisfied with Novo Nordisk’s recently-submitted response, one possible sanction it could take would be to postpone approval of new applications.
  • Management repeatedly highlighted Tresiba and Ryzodeg’s recent EU and Japanese approvals. As a reminder, management believes that Tresiba’s EU label will likely serve as a“reference label,” setting precedents for Tresiba’s designated use in other markets such as the US. The EU label is quite favorable, with three key benefits: 1) it states that Tresiba is associated with less overall and nocturnal hypoglycemia compared to other long-acting basal analogs; 2) it states that Tresiba has four-times lower day-to-day variability compared to Lantus; and 3) it highlights degludec’s 42-hour duration of action, which we view as a big boon for Novo Nordisk and patients, as it grants patients greater flexibility in the timing of their daily insulin dosing (Novo Nordisk still advises patients to take degludec once daily at the same time, as phase 3 data show that degludec taken less frequently is less efficacious compared to glargine and leads to higher rates of hypoglycemia). For further details on the EU approval, please see our January 22, 2013 Closer Look at http://www.closeconcerns.com/knowledgebase/r/938acfdf.
    • The Japanese label is similar to the EU label except: 1) it notes that while a 42- hour duration has been demonstrated in global trials, only a 26-hour duration of action has been tested in Japanese patients; and 2) it only lists superiority for nocturnal hypoglycemia, not overall hypoglycemia.
  • It appears that Novo Nordisk will ask for a price premium for Tresiba - we would guess ~10-15%, but this is purely speculative. During Q&A, management commented on pricing strategy, stating that it would opt for a relatively high premium in Europe – they noted that while such a strategy could slow down European penetration, it would set the foundation for a higher price overseas and in emerging markets. During Novo Nordisk’s 3Q12 update, management remarked that pricing premiums could be justified based on phase 3 data on efficacy, hypoglycemia, and quality of life (no further details provided). We’ll be interested in seeing cost-effectiveness data for efficacy, as degludec’s phase 3 trials were “treat-to-target” studies that could only demonstrate non-inferiority to Lantus. Quality of life, we imagine, is dependent on several factors including the lowered risk for hypoglycemia and Tresiba’s flexible dosing. Regarding hypoglycemia, we have not seen any health-economics data to date, but this analysis would likely encompass the costs saved from preventing hypoglycemia-induced hospitalizations, as well as the costs saved from preventing complications, as a lower fear of hypoglycemia may result in better glycemic control. For reference, during a Novo Nordisk corporate symposium at CODHy Barcelona 2012, Dr. Marc Evans (Cardiff University, Cardiff, UK) presented a survey of 1,400 patients that found that the direct economic consequences of non-severe hypoglycemia led to an additional cost of €1,800 ($2,400) per person per year (for details on Dr. Evan’s presentation see page 38 of our CODHy report at http://www.closeconcerns.com/knowledgebase/r/bf2864bb). Management has previously noted that Tresiba could reduce the cost of blood glucose measurements based on the simplicity of the treatment – we presume this reflects the expectation that degludec’s more stable PK/PD profile will lead to less intra-patient glucose variability, allowing for less frequent blood glucose measurements.
  • Management noted that it expects to launch Tresiba in Denmark and the UK in 1H13, with other European markets following during the rest of 2013 and 2014; consistent with the Japanese launch schedule, Ryzodeg will follow about one year after Tresiba. In the past, management has stated that it believes that familiarity with the degludec molecule will increase HCPs’ willingness to prescribe Ryzodeg. The EU reimbursement process appears relatively arduous, as Novo Nordisk will meet with hundreds of local commissioning groups (212 in the UK alone), speak to their budget holders, and try to gain inclusion on local formularies. Despite Tresiba’s label, management has previously expressed the possibility that the payors may initially cover Tresiba as only a “third line option” (behind Lantus and Novo Nordisk’s Levemir).
  • Management announced that it expects Tresiba price negotiations in Japan to conclude in 1Q13, with launch following immediately after. As background, Tresiba and Ryzodeg gained Japanese approval in September and December, respectively. Management noted that the Ryzodeg launch timing would be determined following Tresiba’s price listing. In the past, management has said that they plan to launch Ryzodeg a year after Tresiba.
  • Both Tresiba and Ryzodeg are expected to launch in 2H13 in Mexico, where they gained approval in 3Q12.
  • Management noted that since filing Tresiba and Ryzodeg in the EU and US in September 2011, the drugs have also been submitted for regulatory review in nine additional countries, including Switzerland, Canada, South Africa, India, Australia, Brazil, and Russia. Tresiba alone was also submitted in Argentina in December 2012.
  • Novo Nordisk could face future competition from Lilly and Sanofi’s candidates, as well as from glargine biosimilars.
    • Lilly’s novel PEGylated insulin lispro (LY2605541; phase 3) has been shown in phase 2 trials to result in weight loss (as opposed to the weight gain typically associated with insulin use) and less hypoglycemia compared to Lantus. However, this candidate also led to higher levels of ALT and AST liver enzymes (and non-significant increases in triglycerides), raising potential safety concerns. On January 8, Lilly and BI announced that BI decided to terminate its collaboration with Lilly on the novel insulin because of “independent strategic portfolio considerations.” While Lilly emphasized during its 4Q12 update that BI’s exit does not reflect any concerns for LY2605541’s clinical profile, we wonder if the potential liver signal influence BI’s decision. Lilly has stated that it will continue executing pre-planned trials for LY2605541 and that internal readouts of data from initial phase 3 trials are expected in 2013. For details, please see our Lilly 4Q12 report at http://www.closeconcerns.com/knowledgebase/r/9da23fa6.
    • Lilly/BI’s insulin glargine candidate is currently in phase 3, with a potential regulatory submission in 2013. Lilly has previously mentioned that it does not plan on disclosing data for this candidate “at this time.” As Lantus’ patent expires in 2015, Tresiba will likely have roughly a two-year head start for reaching patients and providers– indeed patients used to the flexible dosing may be hesitant to switch to a glargine biosimilar despite its potentially lower price. At this moment, it’s difficult to speculate on the glargine candidate’s market impact, as so many aspects of biosimilar regulation and commercial uptake remain uncertain.
    • Sanofi is developing a new insulin glargine formulation in phase 3 (phase 3 trials are expected to complete August 2013 through February 2014, according to ClinicalTrials.gov). In Sanofi’s 4Q11 update, management stated that the new formulation will have a flatter PK/PD profile and require a lower injection volume than other basal insulins, due to the fact that it dilutes at a different rate than Lantus and contains more insulin in a smaller volume. We’ll be keeping an eye out for new details during Sanofi’s 4Q12 update on February 7.

PIPELINE UPDATES – OTHER INSULINS

  • Management announced the initiation of a phase 1 trial for a novel long-acting insulin, LAI287 (NN1436), which has the potential for once-weekly (!) dosing. It’s certainly still early days, but a once-weekly insulin could be revolutionary for adherence if it is able to maintain reasonable hypoglycemia and weight profiles. The trial will investigate the drug’s safety, tolerability, pharmacokinetics, and pharmacodynamics in 80 healthy volunteers and people with type 1 diabetes (ClinicalTrials.gov Identifier NCT01730014). When asked during Q&A about how LAI287 was engineered, management stated only that it was an analog (i.e., the primary amino acid sequence of the human insulin backbone has been modified) with an attached side chain that promotes stability – we understand that this is also the case for degludec, so we remain curious how, specifically, LAI287 is able to achieve such a long half life. Management did suggest, though, that LAI287’s engineering built off of knowledge gained in engineering the relatively long half-life of Tresiba – a perfect example, we think, for demonstrating the importance of incremental change. For reference, management stated that the half-life of Lantus, once it reaches circulation, is only roughly 20 minutes; Tresiba’s is four hours.
  • Management reiterated that it will advance an ultra-fast-acting formulation of insulin aspart (FIAsp; NN1218) into phase 3 development, as announced in December. The phase 3a program (aptly named “onset”) is scheduled to initiated in late 2013 and include roughly 3,000 patients with type 1 or type 2 diabetes. Notably, FIAsp will be investigated in both a pump infusion and multiple daily injection setting in the “onset” program. Since the rapidity of insulin action has been one major barrier to the development of closed-loop systems, FIAsp could help accelerate these efforts, as it would be the first injectable ultra-fast- acting insulin to market should it be approved (it would be behind MannKind/Affreza’s ultra-fast- acting inhalable human insulins). We enthusiastically await faster-onset insulin analogs, as we imagine they could also dramatically improve glucose control and adherence to insulin regimens over currently available analogs.
    • FIAsp’s late 2013 phase 3 initiation places a regulatory submission around late 2014 or early 2015, at the earliest. FDA filing for Mannkind’s Afrezza is expected in 3Q13. Halozyme has a coformulation of PH20 and rapid-acting analogs that is “phase 3 ready,” and phase 2 data for Biodel’s ultra-rapid-acting recombinant human insulin BIOD-123 is expected in 3Q12.
  • Management did not mention Novo Nordisk’s two basal oral insulins in phase 1: OI1362GT (NN1954) and OI338GT (NN1953); no ongoing studies are listed on ClinicalTrials.gov. As a reminder, both candidates use Merrion Pharmaceutical’s GIPET delivery technology – in November 2008, Novo Nordisk signed a development and licensing agreement with Merrion for the development and commercialization of oral insulin, in particular the prandial oral insulin NN1952 (which has been discontinued). Other oral insulin candidates include Tamarisk Technologies O’insula (one international trial of unknown status), Biocon’s prandial insulin IN-105 (phase 2/3), Oramed’s ORMD 0801 (phase 2), and Diabetology’s Capsulin OAD and Capsulin IR (both in phase 2). Our December 21, 2012 Closer Look provides a comprehensive review of the oral insulin landscape: http://www.closeconcerns.com/knowledgebase/r/5439ad6a.

PIPELINE UPDATES – LIRAGLUTIDE FOR TYPE 1 DIABETES AND OBESITY

  • On par with previous forecasts, Novo Nordisk provided updates on Victoza (liraglutide) as adjunct to insulin therapy in type 1 diabetes (NN9211). Management discussed new topline results of a trial investigating the pharmacology of liraglutide in conjunction with insulin for patients with type 1 diabetes. The study demonstrated that liraglutide does not compromise the glucagon response during hypoglycemia or other counter-regulatory responses – this had been a potential concern given that GLP-1 analogs reduce the elevated glucagon levels observed in both type 1 and type 2 diabetes. In addition, liraglutide decreased type 1 patients’ daily insulin requirement and body weight in a dose-dependent manner over the four- week trial. Furthermore, management stated that CGM data suggest using liraglutide in type 1 may improve time-in-zone. Management stated that the safety profile was similar to that seen in patients with type 2 diabetes.
    • The phase 3a program in ~2,000 people with type 1 is expected to begin in 2H13. During Q&A, management forecasted a potential end-of-2016 approval, should everything go according to plan.
  • Novo Nordisk continues to investigate Victoza as an obesity medication in the phase 3a SCALE program – management did not provide any updates during the prepared remarks, but reviewed many details during Q&A. As a reminder, the SCALE program includes four trials, one of which completed in 3Q10 (SCALE-Maintenance), and three that are currently ongoing (details below) and are expected to complete by or in June 2013. During Q&A management discussed several aspects of the drug’s development and commercialization:
    • Positioning the drug: Management stated that although its trials have abided by the FDA’s request to include participants with a BMI >30 kg/m2 (or >27 kg/m2 with a significant comorbidity), the target population will likely be patients with a BMI > 35 kg/m2 and comorbidities. Additionally, management has expressed the intention of positioning the drug for prediabetes, should follow-ups of trial participants who had prediabetes at baseline show that liraglutide can postpone the progression to diabetes. We are encouraged to see companies begin to explore potential approaches to treating prediabetes, and we hope that such efforts motivate the FDA to establish a prediabetes regulatory pathway; however, we imagine that taking an expensive and injectable therapy will be too great of a commitment for prediabetes patients and that an easier-to-take drug like metformin may prove a better initial option for this patient population.
    • Pricing: Since liraglutide for obesity will be used at a higher dose than that used for type 2 diabetes (3.0 mg for obesity vs. 1.8 mg for type 2), Novo Nordisk expects the price to vary linearly with the dose increase (i.e., it expects the obesity application to be 40% more expensive than the current price for diabetes).
    • Market opportunity in the US: Management suggested that the market for liraglutide in obesity is “similar to the size of the current diabetes population.” While Victoza is the GLP-1 market leader for type 2 diabetes (with a 68% share of the global market and a 62% share of the US market), we’ll be interested in how an injectable therapy competes with oral medications such as Arena/Eisai’s Belviq and Vivus’ Qsymia, especially with the potential added concerns of nausea and/or hypoglycemia. For reference, during his presentation at TOS 2012, Dr. Louis Aronne (Weil Cornell Medical College, New York, NY) summarized published data on obesity drugs showing that the weight loss observed with liraglutide (4.4 kg [9.7 lbs]) appears similar to that for Belviq (4.8 kg [10.6 lbs]) butmuch less than that for Qsymia (12.2 kg [26.9 lbs]) – while cross-drug evaluations are challenging, the data suggest that a single-agent drug such as liraglutide may be less efficacious than dual-agent drugs. Full details of this presentation can be found on page 33 of our TOS 2012 report at http://www.closeconcerns.com/knowledgebase/r/ecc11190.
    • As stated in the past, management does not expect to have to conduct a pre- market cardiovascular outcomes trial for liraglutide’s obesity indication.
  • The current SCALE program is investigating liraglutide in three patient populations: 1) obesity and prediabetes (ClinicalTrials.gov Identifier: NCT01272219; primary completion date is listed as March 2013); 2) type 2 diabetes (ClinicalTrials.gov Identifier: NCT01272232; primary completion date is listed as January 2013 on ClinicalTrials.gov, though Novo Nordisk’s slides suggest that it has not yet completed [“expected completion in H1 2013”]); and 3) sleep apnea: a 32-week trial investigating the effect of liraglutide 3 mg in obese subjects with sleep apnea, but not diabetes (ClinicalTrials.gov Identifier: NCT01557166; primary completion date is listed as June 2013).
  • To our knowledge, liraglutide is the only GLP-1 agonist under investigation for an obesity indication.

PIPELINE UPDATES – IDEGLIRA

  • Management discussed data from the phase 3 DUAL II study for IDegLira (fixed- ratio insulin degludec/liraglutide), as announced in December. Consistent with previous estimates, management forecasted a mid-2013 regulatory filing in Europe and a US submission “as soon as possible” pending Tresiba’s approval. As a reminder, IDegLira provided an average A1c reduction of 1.9% from a baseline average of 8.7% over 26 weeks, a statistically greater reduction that was seen with Tresiba (specific percentage undisclosed). Furthermore, patients on IDegLira lost an average of 2.5 kg (5.5 lbs). Importantly, the maximum dose of Tresiba was fixed in both arms so that the additional benefit of adding Victoza could be determined. In the DUAL I trial, which reported in August, patients randomized to IDegLira experienced a mean A1c reduction of 1.9% from a baseline of 8.3%, as well as an average 0.5 kg(1.1 lb) weight loss. Management emphasized that the safety and tolerability profiles observed inDUAL II were consistent with previously reported data – in DUAL I, IDegLira resulted in less hypoglycemia than Tresiba, and <10% of patients on IDegLira experienced nausea (for details on DUAL I results, see our Novo Nordisk 2Q12 report at http://www.closeconcerns.com/knowledgebase/r/e1e0624b).
  • As a reminder, IDegLira is a fixed-ratio combination of the ultra-long acting basal analog Tresiba (degludec) and the once-daily GLP-1 agonist Victoza (liraglutide). IDegLira’s mid-2013 submission timeline in the EU puts it far ahead of Sanofi's fix-flex lixisenatide/Lantus product, for which the final phase 3 trial is expected to complete in February 2014, and which Sanofi expects to file in 2014 with a 2015 launch if approved. IDegLira will allow patients to change the dose of either degludec or liraglutide at a fixed ratio with the other, whereas Sanofi’s pen device will allow for titration of Lantus at a fixed dose of lixisenatide.

PIPELINE UPDATES – OTHER GLP-1 CANDIDATES

  • Management provided no updates on its three oral GLP-1 candidates in phase 1: OG987SC (NN9927; Eligen technology,), OG987GT (NN9926; GIPET technology), and OG217SC (NN9924; Eligen technology) are long-acting oral GLP-1 analogs for type 2 diabetes. During the 3Q12 update, management remarked that Novo Nordisk had advanced several oral-GLP-1 candidates into clinical development to optimize use of the tablet coatings that will facilitate the transfer of the GLP-1 agent across the gut barrier into systemic circulation. In addition to leveraging the GIPET technology (acquired through the Merion partnership) and the Eligen technology (acquired through the Emisphere partnerships), Novo Nordisk is also working on its own carrier technology, though little information has been released at this time. li>
    • Other companies with oral GLP-1 agonists in development include Zydus Cadila (ZYOG-1; phase 1), Arisaph (preclinical nasal and oral candidates) and Oramed (ORMD 0901; phase 2).
  • Management did not mention the novel once-weekly GLP-1 agonist semaglutide, though accompanying slides confirmed that phase 3 would be initiated this year; as a reminder, Novo Nordisk announced in June 2012 that it would advance semaglutide into phase 3 over once-weekly liraglutide. The decision was in part based on phase 2 data presented at EASD 2012: over 12 weeks, in 411 patients, greater A1c reductions were observed with semaglutide 1.6 mg titrated vs. once-daily liraglutide 1.8 mg titrated (1.7% vs. 1.3%, respectively), as well as greater reductions in body weight (3.6 kg [7.9 lbs] vs. 1.4 kg [3.1 lbs]). However, Novo Nordisk will investigate a lower dose of semaglutide for the phase 3 trials, due to an unacceptable side effect profile with the 1.6 mg dose. For reference, semaglutide 0.8 mg titrated (the second-highest tested dose) provided an A1c reduction comparable to liraglutide 1.8 mg titrated (1.4%), but greater weight loss (2.4 kg [5.3 lbs]). The data released at this time makes it difficult to predict how semaglutide compares in safety, efficacy, and tolerability with other GLP-1 agonists set to enter the market (described below in order of closeness to market), since the exact dose of semaglutide to be advanced to phase 3 studies has not been disclosed. To our knowledge, Novo Nordisk plans to position semaglutide in parallel with Victoza, as an alternative GLP-1 option for patients who prefer once-weekly dosing.
    • 1. Sanofi confirmed at JPM 2012 that it has submitted the once-daily lixisenatide (Lyxumia) to the FDA – as a short-acting GLP-1, Lyxumia will be in less direct competition with semaglutide than it will be with other short-acting agonists such as Byetta.
    • 2. GSK submitted its once-weekly GLP-1 agonist albiglutide to the FDA in mid-January, placing a final FDA decision around 4Q13/1Q14; an EU submission is expected in early 2013. As background, the phase 3 Harmony 7 study showed that albiglutide led to smaller improvements in A1c and weight compared to liraglutide, but was more tolerable – given that semaglutide was chosen for its presumptive superiority over once-daily liraglutide, it appears likely that it will present a more favorable profile than albiglutide, barring any serious safety/tolerability issues.
    • 3. Lilly’s once-weekly dulaglutide has completed initial phase 3 trials and could be submitted for regulatory review in 2013. Lilly expects to disclose phase 3 data at scientific meetings this year, but in phase 2 trials, dulaglutide has demonstrated A1c-lowering of about 1.0-1.2% from relatively low baselines of <7.1%-7.9% in phase 2 trials, which is roughly on par with Bydureon.
    • 4. During it’s 4Q12 update, BMS announced that the once-weekly exenatide suspension formulation was advanced into phase 3 during 4Q12. Amylin had previously announced that it would pursue a line extension regulatory approach for this formulation, building upon the already available data for Bydureon and Byetta (much like the Bydureon program built upon Byetta data) – it is uncertain whether this approach could shorten the regulatory approval process.
    • PhaseBio initiated a phase 2b study for its once-weekly GLP-1 agonist Glymera in August 2012. Although specifics were not given, PhaseBio reported positive phase 1/2a results in November 2011 suggesting that Glymera might achieve greater A1c reductions than Victoza if levels reached over the four-week study can be maintained.

QUESTIONS AND ANSWERS

Q: Regarding the FDA inspection you mentioned and the subsequent warning, could you elaborate in terms of whether it has had any impact on the Tresiba approval process? Furthermore, are there any read through to the talks with EMA?

A: In regards to the warning letter, it has had no impact. It was a US inspection, so it has no impact on the situation vis-à-vis Europe, and it was a routine cGMP [the FDA’s Current Good Manufacturing Practices] inspection, which took place at several locations of our manufacturing sites, to which some observations was made, to which Novo Nordisk has responded.

Apparently, the response from one of the aseptic filling plants here in Denmark has not been sufficient to satisfy the agency and hence, in December 2012, we received a warning letter covering two specific observations and some points under these two specific observations. It is a little bit unusual that we announced and we then received a warning letter in advance that the FDA has not posted the warning letter on their website, which they usually do.

And therefore, we would like to restrict our comments to the technicalities of this warning letter, as we are in the process of ensuring that our response, which was submitted towards the end of the December, is complete and satisfactory to the FDA. Until we've had such dialogue, we intend not to comment on it. It has had no impact on our marketed product and we don't anticipate that it will. It has not been linked to Tresiba either, but we have to be, of course, cautious in the sense that the FDA has certain sanctions toward companies spanning all the way from closing warning letters, to re-inspections, to operating plants under the consent decree to postponing approval of new applications. However, these are the measures the FDA has, but we cannot speculate on any of those yet until we've had a dialogue on our recently submitted response.

Q: In terms of the Tresiba approval process with FDA, could you elaborate on the dialogue you're having and maybe compare it with the dialogue you had with the FDA on Victoza?

A: That's a little bit of a broad question. What I can say, though, is that if you go back to those days in 2009 when we had the advisory committee for Victoza, there is no doubt that it was pretty much about risk – i.e., there was really no true clinical discussion about Victoza’s benefits in terms of the clinical views. Quite frankly, a lot of the discussions after the Victoza advisory committee also had to do with the overall clinical review, but which at that point clearly had only been partly underway. Hence, there was not only a focus on all the topics of that particular advisory committee (including C-cell safety and so on), there was also a lot of very detailed clinical discussions right up until the very happy end on January 25, 2010, when we got the approval. I would say it's a fair statement that for Tresiba, there has been more discussion of benefit and risk – i.e. it was a true benefit/risk evaluation that took place at the advisory committee on November 8.

And, obviously, some of those issues that were discussed – hypoglycemia benefits, potential for cardiovascular risk, etc. – had obviously, as you can imagine, popped up as discussion items and things where we have had queries from the agency even since the advisory committee, as a natural part of that process. Obviously, Novo Nordisk has adapted to the discussions – we are at the point where it's natural that the company does submit its various suggestions, including the package insert language and so on. But obviously at this point, we are not discussing the specifics of the label, and I think that is as far as I would go at this point. It is a constructive dialogue. It is progressing.

Q: On the fast-acting insulin analog, could you clarify what's taking delaying the start of phase 3 until the end of this year? Also, is there any plan for a premixed product with Tresiba?

A: The FIAsp project has seen a whole range of clinical activities where we ended up in the final round between two frontrunner candidates. Once we select one of these – this has been taking place very recently – we need to make sure since we're leapfrogging over phase 2 and directly into to phase 3, that we also have sufficient amount of stability data for the company to conduct a full-blown phase 3 program.

This is a new product. It's a new brand. This is not a follow-on compound. We are developing this as a new brand. And obviously, you will need the stability data for this new product, and normally it's not an issue because you have phase 2 and there’s ample time to do phase 3 stability studies. Here, we are actually leapfrogging into phase 3, which gives us less time to do stability. When we say toward the end of 2013, I can tell you one thing: we'll do it as fast as we can. I'll be more specific on the phase 3 start later on in the year.

Whether FIAsp can be used together with Tresiba is really a very good question and Novo Nordisk is never leaving any stone unturned when it comes to engineering or formulating various combinations of our protein, so we can even discuss that at some later point.

Q: Regarding the Victoza obesity program, as we approach the phase 3 data, could you clarify the positioning you intend for the product? Can you broadly discuss the BMI populations and the extent of diabetes overlap?

A: We are approaching the final date where we will get the phase 3 data on obesity. We have actually for some time discussed and communicated our intended positioning, which of course depends on the data. We intend this product to be used for individuals with a BMI of 35 and above, but also in people with comorbidities such as elevated lipids, high myocardial infarction, high blood pressure, etc. And so this is a population which is severely obese, who probably have had weight problems for several years, have participated in several programs to try and lose weight, and for whom weight has become a health risk to them, and something needs to be done unless they progress to diabetes or progress to an even more serious metabolic or cardiovascular situation.

So, in these individuals, we believe we can justify the use of potent pharmacological entities such as liraglutide. And we think that we can justify the relatively costly intervention, because you have to remember that it is used in a higher dose than what we typically use today in diabetes; hence, you should also count on it being linearly priced, such that the application for obesity will be some 40% more expensive than the current daily consumption for people with type 2 diabetes.

It is the intention that in the long term, we will follow up with those individuals in the cohorts that had pre-diabetes, which can be recognized from their clinical parameters. We will check whether or not their diabetes changes or whether we can postpone the progression from pre-diabetes to diabetes over longer periods of time. So, those are the types of patients we'd like to position the drug in. And this is a significant opportunity if you look at it from a market perspective, because if we take the United States, we're assuming that that population which has this drug is similar to the size of the current diabetes population in the US. So, needless to say, there is big need and there is a big potential, if we can show that it's sustainably effective and that the side effect profile is benign. Just to corroborate on the drug population, even though we've abided with the FDA's demand to just enroll anybody with a BMI above 30 or anybody with a BMI above 27 with a comorbidity, the fact the matter is that the patient we just described is the average patient in the trials. So, the BMI average will above 35 and the amount of people with pre-diabetes will at least be 50% at baseline.

Q: What is the uniqueness of the sleep apnea study [on liraglutide for obesity] in this population and what discussions have you had around the results of that study with payers?

A: Sleep apnea is a very specific obesity comorbidity. The physical kind of sleep apnea we are talking about is related to severe obesity, and the way we're doing that is going with some very focused specialist centers where we are measuring all the hallmarks of sleep apnea. We are seeking to improve those in a study in a limited population, but with sophisticated metrics. This study will actually report slightly later than the others, but it will still report around the summertime of 2013. And it will be part of the health economic package that we'll be submitting later on.

Q: Regarding Tresiba, we've seen the label in Europe. You've now got the label and it looks to have all the benefits such as nocturnal hypoglycemia on there. I’m wondering about your pricing strategy in Europe, how that's panning out, and what you think that's going to be.

A: Clearly, the biggest potential that we have for Tresiba is going to be in the United States. It's also where we are having a more liberal pricing environment. Obviously, we need to justify the price to the different managed care organizations and the insurance companies, but we will get to that as we move along. When we talk about Europe, we have stated all along that we need to see a premium on Tresiba in Europe, because the European price level is, to some extent, linked to our export private markets. Many of the export private markets in emerging markets are referring back to the home country, both for approval and for reimbursement status and price. Thus the balance for us is that we probably need to take a higher premium in Europe – which to some extent can slow down the penetration and the rollout in Europe – out of consideration to get a decent price in overseas markets. This is probably the best mix.

Q: On liraglutide in type 1 diabetes, could you review the timeline and give some elaboration on the commercial opportunity you see?

A: First, there is no doubt that when you look at the commercial potential, everybody who has type 1 diabetes worldwide today is typically under multiple daily injections with insulin. The potential is here for a new paradigm where you use the GLP-1 analog – in this case as a rapid site – as an adjunct to insulin.

The rationale for that, which we've confirmed in the clinical study that is just completed, is that by reducing the excess glucagon that is also present in type 1 patients, you will first of all reduce the need for insulin. You will also have the appetite suppressant effect that makes you lose a bit of weight and become more insulin-sensitive, which further improves the situation. What we have seen both in anecdotes and investigator-initiated trials, but also in our own, is that it seems as if the time spent in the near-normal glycemic zone is enhanced. So for the patient, there's a lot of value in actually getting into better control or reducing the amount of hypoglycemic effects at the same prevailing HbA1c level, while also harvesting benefits on body weight and insulin sensitivity and insulin dose.

So that's what we're going to set out to study in two major trials – one will be bigger than the other because it's a superiority trial. They were negotiated with the FDA and EMA very recently, and the first one will start in the second half of this year (that's the one year trial), while the other one (the six-month trial) will start soon thereafter. This way, they are both destined to complete with the estimated approval, if we are lucky at the end of 2016, but as soon as we can. So we're really winning the series. There are millions of people with type 1 diabetes out there. We'll have to look at the profile emerging from phase 3 before I can be more specific about the sales, but of course, it's sizeable.

Q: Will you be required to do pre-marketing CV trials for Victoza in obesity? Is there any progress in the discussions with the FDA. Have you had more discussions around this, and what should we expect now when we see the data that reports out very soon?

A: Well having discussed with the regulators, it's our assessment that the obvious initial things to do are to combine and do a meta-analysis of what exists of pre-existing trials where CV has been measured, including the LEAD program, the SCALE program, the phase 3b trials for Victoza in diabetes. And it is with the admittance from regulators and ourselves that while the doses are not the same (i.e., 1.8 mg versus 3 mg) the body weight is also not the same. Hence, the exposure typically is only 30%, maybe 40% different between the two populations. And there is an understanding that the overarching pathophysiological elements underlying large-vessel disease are somewhat similar whether you are a 45- year-old obese woman or a 60-year-old diabetic male, for that matter.

So that's the initial approach. If, somehow, that creates unclarity and uncertainty, a next step that has by no means been concluded could, of course, be to look into the interim analysis of the LEADER trial, the CV outcome trial for liraglutide. LEADER will have been very well progressed at that point, because the event rate we are witnessing in that trial are at least on par with what we had predicted when we initiated the trial. There's already a commitment to update the European Medicines Agency vis-à-vis such an interim anyhow during the process before 2016. So that is a multi-step process, and we'll take the amount of steps that are needed to satisfy regulators, hopefully with good outcomes. These tests would not involve a pre-approval CV outcome trial. That, in my personal view, would only occur in the event that the point estimate of all the things I just mentioned were to be in excess of one. And it would take quite some drive from where we are today to bring it above one, if you recall from the CV meta-analysis that was done as part of the Victoza approval process.

Q: How is LAI287 is engineered? Is it acylated? Presumably, it's not PEGylated. Does it have the potential to possibly serve the needs of all patients requiring basal insulin therapy?

A: LAI287 does actually represent a case where we've had to look into our own publications and the textbooks about why even the longest-acting insulins tend to have a very brief circulation half-life. The longest one I know about is Tresiba, with a circulation half-life of four hours, and for Lantus its about 20 minutes.

That is because of a curious mechanism of the body, which is insulin receptor mediated uptake, i.e. endocytosis followed by degradation. So, what we've done is map that process, and then we've looked at how can we manipulate that system. And the way to do that chemically is by both introducing a modification at the level of the human insulin skeleton (i.e. creating a true analog) while also modulating the character of the substituent that you're attaching (i.e. the fatty acid side chain).

And this has not been trivial, because we needed to make a step change in the half-life that we've seen from, for instance, Tresiba to create an analog such as LAI287. This has been all benefit from that. In doing everything we've done, you also engineer a degree of metabolic stability into the molecule that may even lend itself to the use for all regimens, in which case you will have the benefit of the long circulation half life, which can then buffer against the day-to-day variability for such an agent when given orally. [Editor’s note: to our knowledge, Novo Nordisk speculates that LAI287’s stability could potentially allow the drug to be delivered through alternative routes of administration (e.g., oral delivery) or to be combined with other drugs such as a once-weekly GLP-1 agonist – again, this is purely on a speculative basis].

Q: On Tresiba, how confident are you that an approval will come in 2013? I ask that question from the point of view of the significant review times the FDA has taken for some diabetes medicines in recent history. Victoza is one example – you’ve already talked about that. Another example is Takeda’s alogliptin, where the FDA has approved it with a commitment for post approval cardiovascular safety studies, albeit after a very lengthy delay.

A: You have to then mind a couple of things. Some examples with very long review times are also precipitated by complete response letter that were interjected. In the case of Victoza, the review time was 20 months (which we thought was a long time, to be frank), but that was without a complete response letter. And my view is that we are in this dialogue with the FDA and unless we get a complete response letter, the dialogue has progressed to an extent that going beyond 2013 without the complete response letter would be somewhat of a surprise to me. Of course, I cannot speculate on the reason of a complete response letter, but I can speculate that we feel that we have a very compelling case for a very safe analog. This is also based on the various aspects that were discussed at the advisory committee meeting. And I think we presented that case quite clearly as a part of the outcome.

Furthermore, the FDA usually follows the recommendation from the advisory committee, so it would be kind of surprising if we were to receive a complete response letter. So we are anticipating that we're going to launch Tresiba all over the world during the half of this year.

Questions for the company:

1) If Tresiba’s half-life in circulation is four hours, and its duration of action is up to 42 hours, this presumably means that Tresiba’s half-life in the subcutaneous space is around 38 hours.

If so, and the half-life in circulation is so much longer than that of Lantus, does that mean that more degludec molecules must be circulating in the blood in order to have an equal effect to Lantus? Does this change a person’s basal insulin requirement?

 

APPENDIX A: MODERN INSULIN SALES BY REGION

 

Worldwide

North America

Europe

International Operations

Region China

Japan & Korea

Modern Insulins

2012 sales (billions)

DKK 35 ($6.0)

DKK 17 ($2.9)

DKK 9.1 ($1.6)

DKK 4.2 ($0.73)

DKK 2.1 ($0.37)

DKK 2.6 ($0.45)

Operational growth from FY2011

15%

22%

4%

18%

32%

0%

4Q12 sales (billions)

DKK 9.5 ($1.6)

DKK 4.7 ($0.82)

DKK 2.4 ($0.42)

DKK 1.2 ($0.20)

DKK 0.54 ($0.094)

DKK 0.67 ($0.12)

NovoRapid

2012 sales (billions)

DKK 16 ($2.7)

DKK 9.0 ($1.6)

DKK 3.7 ($0.64)

DKK 1.4 ($0.24)

DKK 0.37 ($0.064)

DKK 1.2 ($0.20)

Operational growth from FY2011

16%

21%

6%

25%

35%

3%

4Q12 sales (billions)

DKK 4.3 ($0.75)

DKK 2.5 ($0.44)

DKK0.99 ($0.17)

DKK 0.40 ($0.067)

DKK 0.10 ($0.017)

DKK 0.31 ($0.054)

NovoMix

2012 sales (billions)

DKK 9.3 ($1.6)

DKK 2.5 ($0.43)

DKK 2.5 ($0.64)

DKK 1.7 ($0.30)

DKK 1.6 ($0.27)

DKK 1.0 ($0.18)

Operational growth from FY2011

7%

10%

-4%

13%

28%

-2%

4Q12 sales (billions)

DKK 2.5 ($0.43)

DKK0.70 ($0.12)

DKK0.65 ($0.11)

DKK 0.46 ($0.080)

DKK 0.40 ($0.070)

DKK 0.27 ($0.046)

Levemir

2012 sales (billions)

DKK 9.8 ($1.7)

DKK 5.3 ($0.91)

DKK 2.8 ($0.49)

DKK 1.1 ($0.20)

DKK 0.17 ($0.030)

DKK 0.39 ($0.067)

Operational growth from FY2011

21%

32%

9%

16%

72%

-2%

4Q12 sales (billions)

DKK 2.7 ($0.47)

DKK 1.5 ($0.26)

DKK 0.75 ($0.13)

DKK 0.31 ($0.054)

DKK 0.045 ($0.008)

DKK 0.10 ($0.017)

 

-- by Jessica Dong, Nina Ran, and Kelly Close