Novo Nordisk 2Q14 – Diabetes Care up 3% to $3.1 billion; timeline for DEVOTE and Tresiba resubmission moves up once again – August 27, 2014

Executive Highlights

  • Diabetes Care revenue grew 3% as reported (8% in constant currencies) to $3.1 billion.
  • Levemir (insulin detemir) continued its momentum, growing 27% as reported to $660 million and capturing 24% of the US modern basal insulin market, up from ~20% a year earlier.
  • Victoza (liraglutide) sales totaled $560 million in 2Q14, up 6% year over year and 5% sequentially; growth was not as strong in the past due to a class-wide slowdown that management attributed to new drug classes and lingering effects of the pancreatitis / pancreatic cancer controversy.
  • Novo Nordisk now expects to conduct an interim analysis from the DEVOTE CVOT in late 2014 to early 2015, possibly enabling a resubmission to the FDA by mid-2015.

Novo Nordisk provided its 2Q14 financial update this morning in a call led by CEO Mr. Lars Sørensen. Our top highlights (top five financial highlights and top ten pipeline highlights) from the call are included below. Stay tuned for our expanded coverage of this action-packed presentation.

Top Five Financial Highlights:

1.  Total Diabetes Care sales rose 3% as reported (8% in constant currencies) to DKK 17.1 billion (~$3.1 billion) in 2Q14, driven by Victoza and Levemir; the US was responsible for a large share of growth, with strong performances in International Operations and China as well.

2. Novo Nordisk’s modern insulin portfolio grew 8% as reported (14% in constant currencies) to DKK 10.4 billion (~$1.9 billion) in 2Q14, driven largely by Levemir (up 27% to DKK 3.6 billion [~$670 million]).

3. Victoza (liraglutide) growth slowed to 6% due to a class-wide decline in growth in 2Q14, attributable to the pancreatitis controversy and the introduction of SGLT-2 inhibitors.

4. Sales of Tresiba (insulin degludec) reached DKK 141 million (~$26 million) in 2Q14, up from DKK 80 million (~$14 million) in 1Q14.

5. During Q&A, management noted that in addition to increased pricing pressure due to competition, average rebating is growing in the US as patients shift to more Medicare Part D-like plans.

Top Ten Pipeline Highlights:

1. Once again, Novo Nordisk advanced the planned timeline for the DEVOTE trial – an interim readout is expected in late 2014/early 2015, enabling possible FDA resubmission in mid-2015.

2. This timeline update means that the results of two new head-to-head trials comparing hypoglycemia with Lantus and Tresiba will not make it into the resubmission package.

3. During Q&A, management emphasized the importance of preserving DEVOTE trial integrity in the face of an interim data analysis; Chief Scientific Officer Dr. Mads Thomsen touched upon the recent FDA hearing on the topic.

4. The September 11th FDA Advisory Committee meeting for liraglutide 3 mg for obesity is right around the corner; the outcome, management speculated, could have spill-back effects on the perception and use of Victoza.

5. Xultophy (formerly IDegLira) received a positive CHMP opinion last month, paving the way for approval in Europe in one to three months.

6. In May, Novo Nordisk initiated ADJUNCT TWO, the second phase 3a study testing the GLP-1 agonist Victoza (liraglutide) in type 1 diabetes patients; the study will cap patients’ insulin dosages to isolate the additive effects of liraglutide.

7. Earlier this month, Novo Nordisk began the SUSTAIN 4 trial, the fifth of six phase 3 trials investigating the once-weekly GLP-1 agonist semaglutide; phase 3 for semaglutide (including the full CVOT) is expected to complete by the beginning of 2016. .

8. Earlier this month, Novo Nordisk completed Onset 4, the first phase 3 study for FIAsp (a faster-acting version of insulin aspart [NovoLog]) which confirmed the drug’s compatibility with insulin pumps.

9. OG217SC, the lead candidate in Novo Nordisk’s oral GLP-1 agonist program, remains in phase 2, while the company’s lead oral insulin candidate is nearing phase 2.

10. Novo Nordisk began a new phase 1 study on the ultra-long-acting basal insulin LAI287 (with potential for once-weekly dosing) in 2Q14.  

Top Five Financial Highlights

Table 1: Sales of Major Diabetes Products


2Q14 Revenue (billions)

Reported/Operational Growth from 2Q13

Reported Growth from 1Q14

Total Diabetes Care

DKK 17.1 (~$3.1)



Total Modern Insulins

DKK 10.4 (~$1.9)



   NovoRapid (NovoLog)

DKK 4.3 (~$0.78)




DKK 2.5 (~$0.46)




DKK 3.6 (~$0.66)



Human Insulin

DKK 2.5 (~$0.45)




DKK 0.14 (~$0.26)




DKK 3.1 (~$0.56)



1. Total Diabetes Care sales rose 3% as reported (8% in constant currencies) to DKK 17.1 billion (~$3.1 billion) in 2Q14. Sequentially, sales rose 7% as reported following an 8% sequential decline in 1Q14. As in past quarters, the main drivers of growth were Victoza (liraglutide) and Levemir (insulin detemir), particularly in the US market. The 3% reported growth for the portfolio in 2Q14 was comparable with the 1% growth seen in 1Q14 and the 4% growth seen in 4Q13, although it falls short of the consistent double-digit growth figures we were used to seeing from the company through mid-2013 (growth was 11% in 2Q13, 13% in 1Q13, 18% in 4Q12, and a striking 23% in 3Q12). Sales continue to be slowed by pricing and rebate pressure in the US and around the world (see item #5 below), and particularly in 2014 by the loss of an Express Scripts formulary contracts for Victoza and NovoLog (insulin aspart).

  • By geography, North America remains the main driver of growth, accounting for 49% of total Diabetes Care sales in 2Q14 and 61% of the company’s growth in 1H14 (figures were not provided for 2Q14 alone). International Operations and China also made significant contributions to growth, accounting for 24% and 23% of 1H14 growth, respectively.

2. Novo Nordisk’s modern insulin portfolio (NovoLog/NovoRapid, NovoMix, and Levemir) collectively grew 8% as reported (14% in constant currencies) to DKK 10.4 billion (~$1.9 billion) in 2Q14, driven largely by Levemir (up 27% to DKK 3.6 billion [~$670 million]). Sequentially, the modern insulins grew 10%, rebounding from an 8% decline in 1Q14. Novo Nordisk’s share of the global insulin market was 47% as of May 2014, declining slightly from 48% in May 2013; its share of the modern insulin and new-generation insulin market held steady at 46%. As in 1Q14, modern insulin growth was significantly stronger in North America (14% YOY as reported) than in ex-North America (1% YOY as reported).  

  • Levemir experienced very strong YOY growth in 2Q14, up 27% as reported (33% in constant currencies) to DKK 3.6 billion (~$660 million). This performance is Levemir’s strongest since 4Q12, when it also grew 27% (plus, growth in 2Q14 was from a higher base). Sequential growth was also robust at 16% as reported – this is the strongest sequential performance we have seen in the past few years.
    • Robust growth in North America accounted for much of Levemir’s strong showing – sales rose an extremely robust 54% year-over-year in USD (47% as reported) to DKK 2.4 billion (~$440 million). Levemir’s share of the US modern basal insulin market rose by two percentage points to 24% over the past year, continuing its strong momentum from previous quarters (the product held only an ~18% share of the US modern basal insulin market three years ago). Management did acknowledge in Q&A that the gain has been modest “in terms of actual patient capture,” but expressed confidence that Levemir’s strong clinical profile and the newly launched (in the US) FlexTouch pen should support a continued positive trajectory – the pen, which comes with increased capacity and a non-telescoping push button, had what looked like a substantially-funded debut at the ADA exhibit hall this year.
    • Levemir also performed well in China, growing 28% as reported to DKK 83 million (~$15 million). Performance in other markets was weaker – sales were flat at DKK 740 million (~$140 million) in Europe and declined 38% to in Japan and Korea to DKK 49 million (~$9 million).
    • In Q&A, several analysts expressed concern that a likely US launch for Tresiba in early 2016 could hurt Levemir’s momentum. We do suspect that at least part of Levemir’s strong US performance in 1H14 was due to a redirection of marketing resources that had been built up for Tresiba’s launch (which was expected to occur last year and into this year). Management assured analysts that the company should have sufficient sales force capacity to manage a Tresiba launch without causing Levemir to “collapse,” though we would expect that the “all in” effort to launch Tresiba will likely draw some resources away from Levemir, at least in the initial stages.
  • NovoLog (NovoRapid; insulin aspart) sales fell 1% as reported and rose 4% in constant currencies to DKK 4.3 billion (~$780 million). The YOY comparison was challenging, as NovoLog grew 12% as reported in 2Q13. Sequentially, sales rose 9% as reported, rebounding from a 12% decline in 1Q14. NovoLog sales likely continue to be hurt by the loss of the Express Scripts formulary contract to Lilly’s Humalog last year (Humalog sales, by contrast, grew 11% in 2Q14).
    • By region, growth was strongest in the International Operations division – sales rose 9% as reported to DKK 450 million (~$83 million). Growth was also positive in China (up 5% to DKK 140 million [~$25 million]) and Europe (up 4% to DKK 990 million [~$180 million]), but sales declined by 2% to DKK 2.5 billion (~$460 million) in North America and by 23% to DKK 190 million (~$34 million) in Japan and Korea.   
    • Management said the ability to launch a NovoLog FlexTouch pen in the US will be at least partially dependent on the launch of Tresiba. While the company believes the FlexTouch device is a convenient product that will hopefully have a positive impact on Levemir’s market share, management cautioned that it is too early to predict when the device could be launched for NovoLog and that “the tactical decision on when to add additional devices in the US will certainly also be dependent on the regulatory approach for Tresiba.”
  • NovoMix sales were flat as reported and rose 7% in constant currencies to DKK 2.5 billion (~$460 million); sequentially, growth was 5% as reported. NovoMix sales may have also been hurt by the Express Scripts formulary decision – sales in North America declined 7% to DKK 660 million (~$120 million) in 2Q14, although premix insulin has not generally shown growth in the US recently and we don’t anticipate that it will grow much, given the competition. Elsewhere, sales grew 17% to DKK 570 million (~$100 million) in China and 11% to DKK 520 million (~$95 million) in International Operations and fell 5% to DKK 580 million (~$110 million) in Europe and 27% to DKK 150 million (~$28 million) in Japan and Korea.
  • Sales of Novo Nordisk’s human insulin portfolio fell 11% YOY as reported (6% in constant currencies) to DKK 2.5 billion (~$460 million); sequentially, sales declined 4%. Growth was positive in North America for the first time since 2Q13, with sales rising 5% to DKK 540 million (~$100 million). Sales declined in all other markets for which they were reported, including in China, which has historically provided stronger human insulin growth than other markets. This was the first quarter since 2Q13 in which North American human insulin performance beat international performance.

Table 2: Insulin Market Share


Share of Total Insulin Market

Share of Modern and New-Generation Insulin Market

May 2014

May 2013

May 2014

May 2014

Global Total















International Operations















3. Sales of Tresiba (insulin degludec) reached DKK 141 million (~$26 million) in 2Q14, up from DKK 80 million (~$14 million) in 1Q14, the second quarter for which Novo Nordisk broke out Tresiba sales. This represents an impressive 76% sequential climb, though the magnitude of sales remains constrained by low reimbursement levels in some countries. In Japan, where Tresiba is reimbursed at a similar level to Sanofi’s Lantus (insulin glargine), the product has now achieved 21% market share. Uptake has also been strong in Switzerland, Mexico, and India (basal insulin market shares of 19%, 12%, and 11%, respectively). In other countries with more restricted market access for Tresiba compared to Lantus, penetration has been more modest (Tresiba’s market share is 4% in Sweden, 2% in Denmark, and 1% in the United Kingdom).

  • Novo Nordisk is pricing Tresiba at a fairly high premium, which it feels is reflective of the product’s value vs. currently available insulins – CEO Mr. Lars Sørensen commented extensively on this decision during the company’s 3Q13 update, saying that it should not be up to the US to pay for innovation. The company firmly believes that Tresiba’s long duration of stable action, potential for flexible dosing, and reduction in hypoglycemia merit a premium. We aren’t sure that US payers will accept this but we are glad Novo Nordisk is being open about its strategy.
  • During Q&A, a number of investors asked how the Tresiba launch will impact Levemir’s momentum. Management said Tresiba appears to be drawing market share proportionally from the existing basal insulin market, meaning most of the increased share is due to patients switching from market leader Lantus. Management noted during Q&A that the company may not need a large augmentation in its sales force to handle the Tresiba launch – this statement tells half of the story, as the company made significant staff investments prior to the FDA CRL in early 2013 that are still in place. However, management did forecast a boost in promotional spend around the time of a Tresiba launch, assuming it is approved.

4. Management shared IMS Health data demonstrating a marked slowdown in the GLP-1 agonist class in 2Q14 – the MAT volume growth rate was around 8% as of June 2014 compared to 20-25% in early 2013. Management attributed the decline to increased competition from other drug classes and lingering concerns from last year’s controversy over incretins and pancreatitis risk. Victoza remains the market leader among GLP-1 agonists and has held its ~66% share amidst the slowdown; in 2Q14, global Victoza sales rose 6% as reported (11% in constant currencies) to DKK 3.1 billion (~$560 million), a far cry from the consistent double-digit growth the product demonstrated through the end of last year.

  • Victoza sales in 2Q14 were relatively strong in the US (up 10% to ~$370 million) as well as China (up 15% to ~$8 million) and International Operations (up 13% to ~$34 million). Sales were particularly weak in Japan and Korea (down 31% to ~$11 million) – management shared during Q&A that the recent Victoza weakness in Japan was due to the poor judgment of some local providers who put type 1 diabetes patients on a GLP-1 agonist and withdrew their insulin, which led to a number of cases of DKA and one death.
  • Management admitted in Q&A that there is no “silver bullet” to reverse the GLP-1 agonist slowdown but expressed confidence in the class’ long-term potential. Management asserted that “there’s no doubt that GLP-1 therapy is a superior therapy compared to OADs from a clinical point of view” and that such a strong clinical profile will improve the class’ position as more patients and providers become aware of its benefits. While we certainly agree that the GLP-1 agonist class offers a very appealing efficacy and safety profile, we expect that competition from SGLT-2 inhibitors will present a meaningful challenge for the foreseeable future, as that class offers solid A1c-lowering efficacy with a side effect profile that might be more attractive to some patients and the convenience of oral administration. However we have heard more and more at scientific meetings about the power of using SGLT-2 inhibitors and GLP-1 agonists in combination, as the latter class can help blunt the compensatory rise in glucagon secretion seen with the use of SGLT-2 inhibitors.

Table 3: GLP-1 Agonist Market Share


GLP-1 Agonist Share of Total Diabetes Care Market (by value)

Victoza Share of GLP-1 Agonist Market (by value)

May 2014

May 2013

May 2014

May 2014

Global Total















International Operations















5. During Q&A, management commented on the broader US pricing environment. The company has seen rebate pressure in the US increase, due in part to increased competition in segments where products are older and more established (such as rapid-acting insulin, where Novo Nordisk lost one of its major Express Scripts formulary contracts). Pricing pressure has also increased as more employers switch to Medicare Part D-like plans with higher rebates (the same effect as changing payer mix). 

  • Management suggested that exclusive formulary contracts should be less of a factor in the basal insulin market than in the rapid-acting segment, which is more mature and seeing less current innovation. Management also noted that it would be very difficult to obtain for a product with a market share below 20%, as “that would simply create too much turmoil.”  

Top Ten Pipeline Highlights

1. Once again, Novo Nordisk advanced the planned timeline for the DEVOTE trial and Tresiba’s US resubmission – the company now plans to conduct an interim analysis at the “turn of the year 2014/2015,” which could enable a resubmission in 1H15. As of the company’s 1Q14 update, the guidance was for an interim analysis in mid-2015 and a resubmission in late 2015. The step up in the timeline is primarily due to a higher event rate (possibly closer to 3%) in the trial compared to the initial more conservative estimate of ~2%. Based on the revised timeline, and assuming the product is approved, management expects that a launch would occur in early 2016. Recruitment has also proceeded faster than expected, and the completion of the full trial is now expected within three to four years following trial initiation (late 2016 to late 2017).

  • Management noted that an NDA for Ryzodeg (a combination of Tresiba and NovoLog) would be resubmitted to the FDA along with the NDA for Tresiba – we are not sure how soon after the resubmission Novo Nordisk might submit its GLP-1 agonist/basal insulin combination Xultophy (IDegLira; insulin degludec/liraglutide), and in our view, given the upside, it cannot be soon enough.
  • During Q&A, management disclosed that the event rate of over 3% see in the LEADER CVOT for Victoza (liraglutide) was also far above the company’s initial conservative estimate of ~2%. When the entry for LEADER was originally posted, the estimated primary completion date was in January 2016. Late last year, the company updated the estimate to October 2015. Given the difference in the event rate that management has been discussing, we wonder if the endpoint for that trial will be advanced once again. Management did note during Q&A that LEADER is not expected to read out in the next three to four quarters, meaning that the current estimate of late 2015 on may be reliable.
    • Management did not specifically disclose the event rate seen in DECLARE, but the fact that DECLARE’s population is older and insulinized suggests that the event rate in DECLARE could be at least in the same range as LEADER’s >3%. One caveat in making direct comparisons (which management noted during Q&A in 1Q14) is that the LEADER trial is global while DECLARE is more US-focused.
    • Novo Nordisk is not the only company that planned trials with conservative assumptions about event rates – we noticed earlier this year that Lilly/BI advanced the estimated primary completion date of EMPA-REG OUTCOME, the CVOT for its SGLT-2 inhibitor Jardiance (empagliflozin), from 2018 all the way up to 2015, due to a higher than expected event rate. We wonder if the issue of conservative initial event rate estimates might be a factor for other ongoing CVOTs as well.
    • Although in one sense it is exciting that some CVOTs like DEVOTE are progressing faster than expected, in another sense the short duration of these trials means that they are less likely to demonstrate cardioprotection. Chief Scientific Officer Dr. Mads Thomsen addressed this point during Q&A, noting that a short trial with a high event rate theoretically is still sufficiently powered to rule out harm, but will not run for long enough for CV benefits to become apparent. He noted that in long-term studies like ACCORD, UKPDS, and the VADT, benefits were only seen after five or even ten years. The way that most CVOTs are being executed (enriched population, large numbers, shorter duration) means that they will not address some of the most exciting questions (like cardioprotection), although it is completely understandable from a resource standpoint why companies are pursuing relatively shorter trials.
  • In terms of other next generation basal insulins in development, Sanofi’s Toujeo (new U300 insulin glargine formulation) leads the pack, as it is currently under regulatory review. Lilly’s peglispro is currently in phase 3, with a submission forecast for 1Q15. Additionally there are a number of “biosimilar” basal insulins approaching the market in the US and EU. Lilly/BI’s insulin glargine was filed in the EU in mid-2013 and in the US in December 2013, but a Sanofi lawsuit could push approval back until 2016. Merck and Samsung Bioepis have a biosimilar insulin glargine entering phase 3 (read our report), and Biocon/Mylan have a biosimilar insulin glargine on the market in over 10 countries (mostly in emerging markets) and entering phase 3 globally.

2. The revised timeline means that the results from SWITCH 1 and SWITCH 2, two crossover phase 3 trials initiated at the beginning of this year to compare hypoglycemia between Tresiba and Lantus, will likely not be included in the FDA resubmission. Management instead suggested during Q&A that the company plans to resubmit the Tresiba NDA without that supporting data, and to file the SWITCH results post-approval for addition to the product’s label. Differentiation is harder to achieve in the realm of basal insulin analogs, and positive findings from the SWITCH studies would have been very valuable to have in the product’s label at the time of approval and launch. Although the DEVOTE trial will provide some data on hypoglycemia that could potentially indicate a benefit over Lantus, the FDA did not appear to be entirely convinced of Tresiba’s hypoglycemia advantage during its initial review, and that having hypoglycemia comparisons on the label will need to wait until Novo Nordisk can submit the SWITCH data to the FDA.

3. During Q&A, company Chief Scientific Officer Dr. Mads Thomsen commented extensively on the broader FDA paradigm for evaluating cardiovascular safety with diabetes drugs – his points touched upon DEVOTE as well as the FDA’s recent hearing on the subject. He emphasized that Novo Nordisk has been working closely with the FDA to make sure that there is a complete firewall between the individuals conducting the interim data analysis and the current trial investigators – this was one of the biggest areas of consensus during the hearing. CEO Mr. Lars Sørensen added that the shrinking amount of time between the drug’s possible approval date and the estimated full DEVOTE trial endpoint (late 2016 to late 2017) reduces the likelihood that any disclosed results would interfere with the continuation of the trial.

  • More broadly, in response to a question about the FDA’s hearing, Dr. Thomsen speculated that the FDA is holding its hearing to seek a compromise between the need for public visibility into the drug approval process and the need to protect industry’s ability to use interim analyses to support approval. Dr. Thomsen said nothing to suggest that Novo Nordisk will be speaking at the hearing (it did not), but he raised some interesting possibilities on how the FDA could find a new compromise, such as redacting data from package inserts or conducting closed or semi-closed Advisory Committee meetings for Class 2 resubmissions.
  • Dr. Thomsen noted that the FDA has not provided clear guidance on a few elements of its 2008 CV Guidance. For instance, he mentioned during Q&A that the FDA has never defined its threshold for a “reassuring point estimate” that must accompany the ruling out of a hazard ratio of 1.8. As a result, Novo Nordisk simply has to cross its fingers and hope that the data will end up looking favorable. Broadly speaking, drug companies would likely appreciate additional clarity on this and other elements of the 2008 CV Guidance, including on the interim data issue (which we hope, following the hearing, the FDA is working on).

4. With the FDA Advisory Committee for liraglutide 3 mg for obesity right around the corner (on September 11th), management provided some brief suggestions during Q&A on what the agency’s primary concerns might be. Dr. Thomsen noted that the entire obesity pharmacotherapy field has been haunted by the “three Cs”: CNS (central nervous system side effects), cardiovascular, and cancer. However Dr. Thomsen suggested that some of the most significant discussion would center around the imbalances in pancreatitis and gallbladder disorders seen in the SCALE phase 3 program. He expressed hope that these factors will be labeling issues rather than approvability issues.

  • In a response to another question, management suggested that the regulatory outcome for liraglutide 3 mg for obesity could influence broader provider sentiment on the overall GLP-1 agonist class. This could perhaps cause either a positive or negative spillback effect on Victoza, which raises the stakes for the upcoming Advisory Committee meeting somewhat. As we understand it, liraglutide 3 mg itself was more at attempt to use an existing agent to meet the unmet need of a slightly different population than the beginning of a large move into obesity for Novo Nordisk.
  • Full results from the SCALE Diabetes trial were presented at this year’s ADA the 56-week trial assessed the efficacy of Novo Nordisk’s liraglutide 3.0 mg for weight management in 846 overweight or obese adults with type 2 diabetes. At baseline, study participants had an average weight of 105.9 kg (223 lbs.), BMI of 37 kg/m2, and A1c of 8.0%. Treatment with liraglutide 3.0 mg led to significantly greater weight loss at 56 weeks compared to placebo. Specifically, participants in the liraglutide 3.0 mg arm achieved an average weight loss of 5.9% from baseline, compared to 4.6% with liraglutide 1.8 mg and 2.0% percent with placebo. Treatment with liraglutide 3.0 mg also resulted in clinically meaningful reductions in A1c level as well as in systolic blood pressure. The most common side effects were gastrointestinal issues like nausea and diarrhea. Liraglutide was generally well tolerated, with no new safety signals identified.
  • The other new-generation obesity therapy progressing through regulatory review is Orexigen/Takeda’s Contrave (naltrexone/bupropion). During Orexigen’s 2Q14 update, management expressed confidence that the drug would be approved following reassuring data from the Light Study CVOT. However, liraglutide 3 mg is mechanistically quite different from Contrave and the other new-generation obesity medications available (including Vivus’ Qsymia [phentermine/topiramate] and Eisai/Arena’s Belviq [lorcaserin]), which are primarily neuromodulators.

5. The exciting GLP-1 agonist/basal insulin combination Xultophy (insulin degludec/liraglutide; formerly IDegLira) continues to make progress, receiving a positive CHMP opinion in the EU. That decision places a likely approval in the next couple of months. If approved in Europe on that timeline, Xultophy will become the first GLP-1 agonist/basal insulin combination to reach patients – Sanofi’s LixiLan (Lantus/Lyxumia) is on track for a regulatory submission in late 2015. In the US, as we understand it, Novo Nordisk is waiting to receive added clarity on the review of the updated Tresiba NDA, meaning that the specific filing timeline for Xultophy has not been firmly decided. In our view, the most disappointing thing about the FDA’s Tresiba CRL last year was that it pushed Xultophy back so long – the clinical data we have seen on the combination from long-term phase 3 study has been quite compelling.

  • At this year’s ADA, Dr. Stephen Gough (University of Oxford, Oxford, UK) presented one-year follow-up data from the remarkable DUAL I trial. The results showed that the substantial A1c reductions, weight benefit (vs. insulin), and hypoglycemia benefit (vs. insulin) seen after 26 weeks were preserved through to 52 weeks (A1c reductions after 52 weeks were 1.8% on IDegLira, 1.4% on insulin degludec and 1.2% on liraglutide).

6. In May, Novo Nordisk initiated ADJUNCT TWO, the second phase 3a study testing the GLP-1 agonist Victoza (liraglutide) in type 1 diabetes patients. The randomized, double-blind superiority trial is expected to enroll around 800 type 1 diabetes patients, randomizing them to liraglutide or placebo in addition to their regular insulin therapy. The trial is designed to solely study the additive impact of liraglutide, as investigators will cap patients’ daily insulin dose at the average dose at baseline. The trial is currently recruiting, and has an estimated primary completion date of April 2015 ADJUNCT ONE, which is currently ongoing, began late last year and is expected to end in May 2015 – it is a treat-to-target study with a slightly longer duration (52 weeks for the primary endpoint).

7. Earlier this month, Novo Nordisk began the SUSTAIN 4 trial, the fifth of six phase 3 trials investigating the once-weekly GLP-1 agonist semaglutide. SUSTAIN 4 will compare semaglutide to Sanofi’s basal insulin Lantus (insulin glargine) for 30 weeks, enrolling over 1,000 insulin-naïve type 2 diabetes patients. SUSTAIN 5, the last phase 3 SUSTAIN trial to begin, is expected to start enrolling patients later this year. Out of the five trials now registered on, the non-CVOTs have endpoints in mid-to-late 2015, while the CVOT (SUSTAIN 6) has an end date in January 2016, which is fairly soon after the projected end of the rest of the phase 3 program. However, some trials may be progressing faster than expected, because at this year’s ADA company analyst call management suggested that the first data should be available at the end of 1Q15. The company is also conducting a preclinical set of studies to better understand the component that facilitates transepithelial uptake of the drug in the gut.

  • The fact that semaglutide will not be the first once-weekly GLP-1 on the market does not temper management’s enthusiasm for the candidate. Previously, management has forecast that semaglutide will have greater efficacy than Victoza, which (if proven true) could also give it a leg up on other once-weekly GLP-1 agonists. That list includes AZ’s Bydureon (exenatide), GSK’s newly-launched Tanzeum (albiglutide), and Lilly’s Trulicity (dulaglutide), which is currently under regulatory review. During Novo Nordisk’s ADA analyst call, management suggested that the first half of 2015 will be an exciting time for semaglutide.
    • A key point of differentiation among once-weekly GLP-1 agonists is on device design and administration protocol; the two currently available once-weekly GLP-1 agonists require reconstitution. The Bydureon pen (approved earlier this year) represents a significant improvement over the previous Bydureon tray formulation, but the reconstitution process (now encapsulated entirely within the pen) takes approximately 15 minutes (mostly warm-up time after removing from the refrigerator). The Tanzeum pen also encapsulates the entire reconstitution process within one device, and also requires 15 minutes of waiting time (30 minutes for the higher dose) as part of the administration process. Novo Nordisk management have in the past suggested that the biggest potential competitor to Victoza’s leadership of the GLP-1 agonist field is Lilly’s Trulicity (dulaglutide), which comes as a ready-to-use suspension that will be administered through an auto-injector. Semaglutide is also a ready-to-use suspension, and both AZ and GSK are working on suspension formulations of Bydureon and Tanzeum, respectively.
    • Taking the device element to a whole new level is Intarcia’s ITCA-650, a once-yearly implantable exenatide osmotic mini-pump. ITCA-650 is in phase 3 trials and reported striking phase 3 results at ADA showing >3% A1c reductions in a high baseline A1c population (>10%).

8. Earlier this month, Novo Nordisk completed a pump study (the first of four phase 3a trials) on FIAsp, a faster-acting version of the NovoLog (insulin aspart) molecule – the trial, Onset 4, was a six-week randomized double-blind parallel-group trial comparing FIAsp to NovoLog in 36 insulin pumping adults with type 1 diabetes. Both insulins demonstrated pump compatibility and there was no apparent difference in the number of microscopically confirmed infusion set occlusions.

  • PK/PD data on FIAsp presented at ADA was quite impressive, with 3.6 minutes to first appearance compared to 9.8 minutes for insulin aspart (NovoLog). The time to 50% maximum concentration for FIAsp was 27 minutes vs. 35 minutes for insulin aspart, which translated to a 22 mg/dl lower postprandial glucose excursions after one hour (26 mg/dl lower after two hours).

9. The lead candidate in Novo Nordisk’s oral GLP-1 agonist program, OG217SC (NN9928, an oral formulation of semaglutide) remains in phase 2. The compound was the first oral GLP-1 agonist to reach phase 2, and Novo Nordisk has in the past suggested that it can take some small shortcuts with the compound’s development because it is already conducting an extensive phase 3 program for injectable semaglutide. The compound uses the Eligen carrier technology from Emisphere to protect the peptide as it passes through the harshly acidic stomach, allowing it be absorbed in the gut. Another oral formulation of semaglutide (OG217GT/NN9928) and two oral formulations of liraglutide (OG987GT/NN9926 and OG987SC/NN9927) remain in phase 1. Other phase 2 oral GLP-1 agonists include Transtech’s TTP054 and Oramed’s ORMD-0901.  

  • According to Novo Nordisk’s 1Q14 update, the company plans to advance its primary oral insulin candidate OI338GT (NN1953) into phase 2a in 2015. Novo Nordisk has completed three phase 1 trials on the candidate. Earlier this year, Novo Nordisk discontinued development of its two other oral insulin candidates, OI287GT and OI362GT. OI338GT is a novel long-acting basal insulin analog that uses Merrion’s GIPET carrier system to carry it from the GI tract into circulation. The product’s phase 1 testing included three clinical pharmacology studies in a total of 118 healthy volunteers and people with type 2 diabetes, supporting its safety and dose-dependent glucodynamic effects. According to the company’s 1Q14 update, Novo Nordisk is planning for OI338GT’s phase 2a trial to be a proof-of-principle trial investigating glucose lowering and safety, including rates of hypoglycemia, after individual titration of OI338GT in people with type 2 diabetes. Based on results of this phase 2a trial, larger phase 2b proof-of-concept studies may be initiated.
  • Other companies developing oral insulins include Biocon (phase 2), Oramed (phase 2), and Aphios (preclinical), among others. Novo Nordisk has the advantage of having the most experience in insulin and a large amount of resources to support this project – in 2013, management estimated that $3 billion might be required to fully develop oral insulin and oral GLP-1 agents. Although not quite an insulin tablet, MannKind’s inhalable insulin Afrezza was recently approved and licensed by Sanofi. Generex is also developing a buccal insulin spray, Oral-Lyn.

10. In 2Q14, Novo Nordisk began a new phase 1 study on its ultra-long-acting basal insulin LAI287 (NN1436), which has the potential for once-weekly dosing. This follows the completion of an initial phase 1 study in mid-2013. The study ( Identifier: NCT02148861) will enroll around 48 type 2 diabetes patients, comparing LAI287 to Tresiba. It is currently recruiting, and has an estimated primary completion date of March 2015.

  • We have not heard much from management about this candidate, and in general the once-weekly insulin competitive landscape is not yet too crowded. In June of last year, PhaseBio announced that it had initiated phase 1 testing of its once-weekly basal insulin Insumera (PE0139), while AntriaBio is moving towards a clinical program for its PEGylated basal insulin AB101.

Table 4: Novo Nordisk Diabetes and Obesity Drug Pipeline

Compound Name




Other Remarks

Tresiba (insulin degludec)

Ultra-long-acting basal insulin analog

Type 1 and type 2 diabetes

Approved OUS; CRL in the FDA

Re-submission to the FDA expected in early 2015 after interim analysis of DEVOTE CVOT

Ryzodeg (insulin degludec/ insulin aspart)

Pre-mixed basal/bolus insulin

Type 1 and type 2 diabetes

Approved OUS; CRL in the US

Re-submission to the FDA expected in early 2015 after interim analysis of DEVOTE CVOT

Xultophy (IDegLira; insulin degludec/ liraglutide)

GLP-1 agonist/basal insulin fixed-ratio combination

Type 2 diabetes

Filed in the EU May 2013; positive CHMP opinion one month ago

Awaiting Tresiba decision in US

Liraglutide 3 mg

GLP-1 agonist


Filed in the US and EU December 2013

 FDA Advisory Committee on September 11th

FIAsp (NN1218)

Ultra-rapid-acting insulin analog

Type 1 and type 2 diabetes

Phase 3

All four phase 3 Onset trials initiated, Onset 4 complete


Once-weekly GLP-1 agonist

Type 2 diabetes

Phase 3

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

LATIN T1D (liraglutide for type 1 diabetes)

GLP-1 agonist

Type 1 diabetes

Phase 3

Phase 3 initiated in 4Q13, ADJUNCT 2 recently initiated

OG217SC (oral semaglutide; NN9928)

Oral GLP-1 agonist

Type 2 diabetes

Phase 2

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

OI338GT (NN1953)

Oral basal insulin analog


Phase 1

Phase 2 expected to begin in 2015

LAI287 (NN1436)

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


Phase 1

Completed first phase 1 trial in July 2013, new phase 1 trial initiated in 2Q14

OG987GT (oral liraglutide; NN9926)

Oral GLP-1 agonist


Phase 1

Uses Merrion’s GIPET oral delivery technology

OG987SC (oral liraglutide; NN9927)

Oral GLP-1 agonist


Phase 1

Uses Emisphere’s Eligen carrier technology

OG217GT (oral semaglutide; NN9928)

Oral GLP-1 agonist


Phase 1

Uses Merrion’s GIPET oral delivery technology


Questions and Answers

Q: Could you give us more details on the event rates you've seen in DEVOTE so far? Is there anything unusual in that rate compared to what you're seeing in the LEADER trial, for example, or in other CV outcomes trials that have reported for DPP-4 inhibitors? Do you have any thoughts on why the event rate is now higher than you had planned?

A: We had originally been on the slightly conservative side when we discussed the planned MACE event rate at last year’s EASD meeting. We also planned for an event rate slightly above the 2% per annum for LEADER as well. As you are aware, we did demonstrate last year that we were above 3% for the LEADER trial.

When you compare the LEADER population with the DEVOTE population, by definition, the DEVOTE population is an insulinized population because without that we could do the trial without patients taking insulin. That means they are older. They’ve had the disease for a longer time and they are sicker – when you look at all of the cardiovascular risk factors they are on the high side compared to LEADER if anything.

However, this was not a given at the initiation of the trial, but it is something that we can conclude now based on a significant amalgamation of baseline data from the population. I cannot give you an exact event rate, but only say its significantly higher than the originally planned for 2%. If we look at other CVOTs such as the incretins, the DPP-4 inhibitors, they tend to very often be in the level of 2% because these are earlier patient populations. We’re probably on the high side in this one, which I think is quite fine; these people are somewhat older. But we have a strong power in the trial, at least towards the end of the trial, by definition.

Q: You have some crossover trials to establish beyond doubt the profile of Tresiba vs. Lantus. If the DEVOTE trial means you can file earlier, would you wait for those trials? How would those trials impact the review with the FDA in terms of thinking about the benefit side of their evaluation, and getting onto the label?

A: This has some implications for which data we can bring to the FDA when we start discussing the outcome of this interim analysis. If it is correct, we have this ongoing hypoglycemia trial, which were intended to put more solid platform under out claims of hypoglycemia. When can we expect that to be available, Mads?

A: I should mention that one of the serious adverse events that we are following in the DEVOTE trial is obviously severe hypoglycemia. That is being followed, but it’s powered to watch the cardiovascular aspect, as we’re all aware. The BEGIN-SWITCH trials – the SWITCH 1 is with patients with type 1 diabetes and the SWITCH 2 is with patients with type 2 diabetes – are trials that were originally planned to coincide the ability to co-submit with the DEVOTE trial interim data. This is no longer going to be the case because we’re ahead of plans, so to speak.

If the hazard ratio is comforting for the FDA, the DEVOTE interim data will form the basis for the first label and hence the approval of the product and launch in the US. Then as soon as we have hypoglycemia data in these blinded trials, they will be submitted to update the US label.

Q: Going back to DEVOTE, what are your expectations for the August 11 FDA cardiovascular study panel, and could it have any impact on your filing timelines?

A: When we look at the Advisory Committee meeting that is happening on August 11, we have the situation where the FDA is seeking guidance on how they kind of can seek a compromise between being open so that the public community has insight into what's going on in the drug development process, and also protecting the innovative industry in such a way that the industry is at all able to develop products and do interim analyses. This has to do with the redaction process – how can you redact out the data from a package insert if you get the approval based on the interim. What is the dialogue? If you want an Advisory Committee as part of the Class 2 resubmission or the cardiovascular submission, how do you conduct that and that can you do it open, semi-closed, or closed? There are many points of discussion there.I feel we are in a pretty good shape regarding the way we're handling DEVOTE and doing so in agreement with the FDA.

Q: Would you remind us of the dialogue you've had with the FDA around the interim data just to check off there's no risk similar to what we've seen with ELIXA and Contrave’s Light Study?

A: As you can imagine, at the point in time when Novo Nordisk received the complete response data on February 8 last year, we were initially from day one thereafter in very, very intimate dialogue with the FDA about how to conduct DEVOTE. We have also recently been in a very intensive dialogue about exactly how to do such interim analyses, making sure that you have a complete firewalling between the interim analysis folks and those doing the trial going forward. You cannot have the situation where you're by any means and measures compromising the definitive trial result by doing the interim. That was the whole lixisenatide dilemma, as I understand it being communicated. Of course, there has been a very close dialogue on that point.

Q: Do you have any color at this stage of what the upper definition of a “reassuring point estimate” is with the FDA for cardiovascular adverse events?

A: That has never been defined by the FDA in any documents or at any meetings. What has been defined, as you are aware, that as an interim cutoff, the upper bound of the 95% confidence interval must not exceed 1.8. But it has to be, as you correctly point out, accompanied by a reassuring point estimate, and we don't know exactly what that is. I'm not sure anybody does, or at least it's never been communicated. We're simply crossing our fingers and hoping that the interim data will turn out in a favorable way. As you know, the opposite situation is obviously that one would have to wait until the definitive trial has concluded, and we've also concluded today that that would be three to four years from trial initiation. But so far we're obviously betting on the interim data, which we will have around the turn of this year.

A: It's by advancing the timeline on the interim analysis we also indicate that the completion of the trial is advanced. So, if you take the timeline that we've indicated now, if we have a confidential dialogue with the FDA, before they make a final decision on those data, we will be in a position that the gap between then and when we have concluded the trials will be small; therefore the risk of destroying the remaining of the trial is minimized.

Q: One thing that always confuses me with cardiovascular outcomes on long-term therapies is, if you're getting results very quickly, how does the FDA reconcile that with a long-term rest-of-your-life therapy such as Tresiba? Do you have discussions on how you look at the events over time, or is it simply just the hazard ratios and the guidelines that they've set?

A: Speaking as a has-been pharmacologist, I would define two scenarios: One where you want to prove a benefit, and the other where you want to prove a harm. If you're trying to prove a benefit we know from the ACCORD, the VADT, the BCCP, the UKPDS, and all the other trials that you do need to see an exposure time of five-plus years, sometimes 10-plus years, before there's a realistic chance to have seen the atherosclerosis process progress so much in the control group that it dissects out statistically significantly from the active group. So that indeed requires many years of exposure.

When we are looking at the harm situation, it is the case that once a drug is in steady state in the circulation in the organism, you will expect it to behave the same after one month, two months, three months, one year, two years, or three years. When you're trying to prove no harm, it becomes a game of having adequate exposure, having it in enough people to get the requisite number of events.

Yes, what we are presenting to the FDA, I hope, pending the outcome of the interim analysis, is obviously something that has been discussed with the FDA. I would say that it is true that when DEVOTE is to conclude only three to four years from trial initiation, we are in a situation that if this had been a trial like LEADER, where those among us who are optimistic think there could be some chance of an upside, that would be highly unlikely to happen in DEVOTE due to the shortness of the trial.

Q: Now that you can potentially launch Tresiba faster in the US, what should we expect about the sales force expansion and the very, very strong performance you have seen on Levemir in the US? It may not be good to remove the foot from the pedal for Levemir – how will you balance this? When should we expect you to increase sales forces?

A: When it comes to the size of the sales force, again, it is difficult to predict the future because it also depends on the dynamics up against the key competitors and what they are doing with their sales forces. It becomes a bit of a speculation. However, if you speculate that the current market dynamics are unchanged so that the sales forces we are up against remain more or less unchanged, then it's our estimate that with the current structure we have, we will have ample sales force capacity to undertake the Tresiba launch without any significant increase in the U.S. sales force.

A: We saw some cost reductions in 2014 as a result of not launching Tresiba; I would imagine that the promotional spend would be increased compared to the current situation.

A: That is, of course, quite clear. Assuming that we eventually get an approval for Tresiba in the US, and in connection with the launch there will be a significant spend on promotional activities.

A: This is likely to be the most significant launch that we've ever made as you can appreciate. This is one of the biggest potentials we have at least in the short-term.

Q: As it relates to the growth for Tresiba in Japan and the impressive 21% share that you have there, how much of that is coming from Levemir versus how much are you gaining from glargine?

A: Of course, the fear when you launch a product like this is that you'll be cannibalizing yourself – then you will not gain much additional business from it. I'm happy to inform you that that is not the case. What we see is that out of new patients, the portion that comes from Levemir more or less corresponds to the Levemir market share we have. As you know, we are not the leading basal insulin unfortunately in Japan. That is Lantus, and that also means that the majority of new patients that go on to Tresiba, they come from Lantus. Then again, it's also important to get people that are new to medication. We also get that, and we look at some from OADs, but the majority of patients that start on Tresiba every month are coming from Lantus.

A: So consequently if you take weekly market shares, then we turned the declining trend in the basal segment to a stabilized situation and, in fact, can we expect to see a growth going forward?

A: We have changed the basal segment into a growing market share. We are growing our basal insulin value market share in Japan now.

Q: With respect to the potential launch of Tresiba in the US next year, given the positive momentum that you received from Levemir, how do you expect to manage that launch to maintain Levemir momentum and launch Tresiba?

A: I just have to correct one assumption that you used in your question, which was that we would be able to launch Tresiba next year in the US; our assumption is that we’ll launch in early 2016, giving us enough headroom for the regulatory process and for compliance data.

A: If you imagine what we see when we launched Tresiba, it is that we gain strong market share growth. The sourcing comes sort of from the proportion of products in the current market. And since we are, in most markets, the small one in the basal segment and glargine is the big one, then even if you sort of just convert patients from existing therapy on to Tresiba, you gain market share because you move a lot more patients from Lantus to Tresiba than you move from Levemir to Tresiba. Those dynamics are very positive, and that is why you see that there's a net positive effect in the total basal market share for Novo Nordisk in the markets where we launched Tresiba. What you should expect if and when we get FDA approval is that we will launch full blast, so to speak, and we will go all in on Tresiba. However, because we go all in on Tresiba doesn't mean that Levemir collapses. We will do a classical launch where you typically first part of the launch you go all in on the new product, and that will be our plan.

Q: It appears that the Levemir FlexTouch pen has been able to grab some good margin. When do you expect the FlexTouch will be available for your NovoLog franchise in the US? Do you think this could make a change for the competitive balance between NovoLog and Humalog?

A: We are very happy to just have launched the FlexTouch, a pre-filled device that also injects when you touch the push button and doesn't move. It’s a convenient product for the patient. It's too early to say exactly what the effect in the marketplace will be. Obviously, we're hoping to see a positive dynamic on market share, but it is too early to tell. The data is currently too limited. It's also too early for us to say anything firm about NovoLog and at what point in time we potentially would change the device from FlexPen to FlexTouch for NovoLog. Basically, it remains to be seen how this plays out, and we'll take our decision based on what we see in the marketplace.

A: Of course, the total capacity we have for FlexTouch will have a high priority of being able to support the Tresiba launch in the US. So the tactical decision on when to add additional devices in the US will certainly also be dependent on the regulatory approach for Tresiba in the US.

Q: For Levemir, now that you've gained a more critical mass in the US, does that make it easier for you to win contracts? Is that also explaining why we suddenly see this significant market share move in 2014, of course besides the increased marketing spend on Levemir? What is actually driving underlying Levemir, and do you see that continuing at least until the Tresiba launch?

A: In the US, the Levemir market share is around 24% right now. What we've seen during the last year is a growth of roughly two percentage points, from around 22% to around 24%, roughly depending on how you measure in detail. You could say that in absolute terms, a 2% market share gain is not really very dramatic. But the basal insulin segment in the US is the biggest value segment in the world in terms of insulin. So, of course, it is significant in terms of growth of our volume in the marketplace.

However, in terms of actual patient capture, it is a modest gain. We are convinced that with the very strong clinical profile of Levemir, the good weight profile of Levemir, and now having clearly a very strong offering in terms of a pre-filled auto-dosing device, that we can continue the trend on our Levemir market share.

In terms of contracting, it's not a dramatic change whether you have 20% or 22%, but of course, if you're below 20%, it's very hard to imagine then that PBMs will go exclusive with you because that would simply create too much turmoil. We are, in general, not seeing exclusives in the basal segment. We see that more in the more mature fast-acting segment where there's been less innovation – or rather innovation is older.

Q: Victoza has definitely slowed down a bit. Could you try to explain what you can do both in the US, but also outside the US, to try to grow the GLP-1 agonist market? Are larger rebates required also outside the US, or significant discounts to get European reimbursement?

A: This is, of course, a different situation to Levemir that we just discussed. Here we have a very high market share – we basically have approximately two-thirds of the market, and we continue to have a very strong position being the preferred GLP-1 product. We expect to maintain that position going forward. So this is more a question of the total market dynamics. There's no doubt that GLP-1 therapy is a superior therapy compared to OADs from a clinical point of view, and it's a very small part of the patient population that is currently on GLP-1 therapy. I don't think there is a silver bullet here. It's about getting the knowledge out to more and more doctors, and getting the knowledge out to patients. It's about making sure that this treatment option is being considered all around the world. I'm convinced that we will see long-term growth of the GLP-1 segment because of the very strong clinical benefits that the product offers.

A: So can we say that we’re not really contemplating forgoing the exclusivity of that product by rebating at this point in time?

A: I just have a clinical comment because we can't quantify to what extent the market still has this original pancreatic cancer scare. Obviously, do be aware that the more we see the emergence of large scale cardiovascular outcome trials such as LEADER and the one for Bydureon, and so on, will hopefully provide convincing safety data in a large and long-term scale, that will take out this last residual of scare of using the products among some doctors. And then of course, Novo Nordisk is constantly life-cycle managing Victoza both by creating whole new indications, such as for type 1 diabetes indication with the LATIN program, but also looking into further segmentation combination therapies, and other strategies, for instance what we have been doing in Japan.

A: The LEADER trials are not going to report in the next three to four quarters, and the LATIN trial will not be done and on the label. That, of course, is not going to impact the short-term growth.

A: There is one short-term item that's going to happen. We are going to have an outcome on liraglutide 3 mg for obesity. This is, of course, the same molecule, liraglutide, that's being used in Victoza but for a different indication. It will be very exciting to see the outcome and also to see the sort of regulatory process with FDA. There's probably no doubt that the market will, to some extent, look at what happens with liraglutide in that indication – that could also influence the sentiment on the total use of GLP-1.

A: You’re saying you’d have a spill-back effect on the diabetes application?

A: Most likely.

A: We're all crossing everything we have here and waiting for September 11.

Q: As you look at the upcoming Advisory Committee for the 3 mg use for liraglutide in obesity, can you please remind us what do you expect will be the key concerns from the FDA perspective? What are the likely key concerns for the FDA there given what we already know of the side effect profile for the approved anti-obesity medications?

A: Generally speaking, the whole field of obesity pharmacotherapy has been haunted by what I would call the three Cs: CNS, cardiovascular and cancer. In the case of GLP-1s also C-cells. Obviously, these four Cs will be discussed, as we also saw at the Victoza Advisory Committee meeting many years ago. I do expect, though, that a significant discussion will be on the numerical imbalance that is seen on the two topics we've discussed with you previously –namely, the pancreatitis and gallbladder disorders. We hope, and would optimistically expect, that these are things that will be label concerns and handled as they are already, for instance, for pancreatitis today in diabetes, but it all boils down to one thing: How does the FDA overall see the benefit-risk profile in the segment that is to be treated? That's as much as we can say now, but we'll know quite a bit more a month and a few days from now.

Q: Do you have any comments on the contraction of the GLP-1 agonist market in Brazil and Japan?

A: We had two very specific situations here. In Brazil, we had an extraordinarily strong launch of Victoza, and this may have been affected by off-label use because there were some magazine articles in Brazil that occurred without our involvement where Victoza was praised as an anti-obesity product. We have worked hard with the Brazilian healthcare authorities to ensure that there is no off-label use of the product in Brazil, and it is likely that this has affected some of the development in Brazil. And we think that we are at a place where we can say that the current use in Brazil is predominantly on-label, and we’re doing everything we can to ensure that.

In Japan, it's a completely different story. Some of you might remember that when we launched Victoza in Japan, it was not indicated for combination use with insulin. And there was a very tragic situation where some key opinion leaders and some specialists put patients with type 1 diabetes on Victoza and removed their insulin; that is not what you're supposed to do, of course. That led to some very serious cases of ketoacidosis and, actually, also a death. We had to spend more than half a year just safety promoting the fact that you cannot use Victoza with insulin, which basically stopped the launch before it really got started. We have a nice possibility coming up in that we have done the clinical trials to prove that insulin and Victoza in combination is safe and efficacious therapy. We are getting that approved in Japan. Maybe we will see a better development long-term of Victoza in Japan.

Q: On the US pricing environment, managed care is trying to squeeze various categories going forward. We're just now seeing albiglutide launch at a relatively low price. Do you have any predictions for the future US pricing environments and where it goes – up, down, flat?

A: There are sort of a two different developments that are interesting. If we talk about rebate levels, it is clearly our notion that the rebate levels are increasing, but they would depend on the competitiveness of the individual segment; it will be a different situation in the GLP-1 segment, for instance, compared to the basal insulin segment, compared to the short-acting insulin segment. This variability is due to the maturity of the products and the competitive nature of the participants.

It's no secret that Lilly was very aggressive on bidding for the short-acting insulin business with Express Scripts. On the other hand, we were not very aggressive in bidding on Victoza with Express Scripts and got excluded from that. How this plays out will depend on the individual segment. However, in general, the demand for rebates to stay competitive, to stay in the market, is going up.

There's another move which is slightly different; that is that the extent to which the commercial insurance is changing because employers adopt Medicare Part D-like claims, transferring patients to a different channel with a different rebate level. This is, in a way, changing the mix of where our customers are getting their insurance. We have seen that to some extent already now, and the extent to which that continues is a very interesting one because that is likely to create some pressures for the pharmaceutical industry in general, as well.

Q: Regarding Xultophy in Europe, have you had any preliminary pricing discussions with parties in Europe? If so, could you give us some information regarding those?

A: We can't comment on details here, but we can say that in general what we're going for is a straightforward combination price, taking the milligram and the unit-to-unit price for insulin, the milligram price for the Victoza part, and then that will be the price we will be going with. As to how many countries that will lead to market access and launch – that remains to be seen. I'm sure there will be markets where we can launch very effectively. I'm sure there will be markets where it will be uphill, and it will take several years in Europe before we can get market access.

Q: There are some pretty obvious dynamics at play for next year. You've got a competitor launching – or hoping to launch – a new basal, making it clear publicly that they're going to throw a lot of resource behind that. You have your own potential launch plans or gearing up to your launch plans for Tresiba and multiple pipeline readouts during the first half of 2015? Would it be wrong to assume significant step-up in sales and marketing dollar spend next year compensated by a more stable level or even a decline in R&D?

A: We have a tradition of actually providing a glimpse of insight into the next year when we get to our third quarter results. I can share with you the thought on how the cost development will be from a helicopter perspective. I think it is going to be likely when you look at the overall operating margin that you will not see a significant expansion in 2015, for the exact reason as you were referring to. We're not going to see the same expansion in the R&D ratio next year. On the other hand, we will probably see – hopefully – a ramp up in the S&D cost that as an expression of improved launch opportunities in the US, which is where the predominant part of cost is going to be utilized.


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