1Q16 and 2Q16 Diabetes and Obesity Industry Roundup – Record sales on growth in SGLT-2s, GLP-1s, CGM, and pumps; pressure in BGM and insulin – October 5, 2016

Based on results from the nearly 30 public companies with revenue that we regularly track, our 1Q16 and 2Q16 Diabetes and Obesity Industry Roundup provides the high-level business and financial trends from the first half of 2016.

This report is divided into four sections: (i) overall industry highlights; (ii) diabetes drugs; (iii) diabetes technology; and (iv) obesity. Each includes sub-section analysis by therapeutic area, followed by charts with current and historical sales figures, growth, and market share estimates. We note that the data represent our best estimates in many cases since a number of companies do not disclose financial information in great detail.

Overall Industry Performance

1. Diabetes revenue across drugs (excluding generics) and devices totaled an estimated ~$11.0 billion in 1Q16 and ~$12.2 billion in 2Q16, rising 3% year-over-year (YOY) in the first quarter and rising 6% YOY in the second quarter. This uptick in the industry occurs against an easy comparison, with 2015 having been the most challenging year for the diabetes field since we began tracking its performance in 2004. Newer drugs and technologies propelled the field’s growth: GLP-1s (36% share of industry growth in 2Q16), SGLT-2s (31%), DPP-4s (22%), CGM (7%), and insulin pumps (3%). The field experienced pressure in the insulin market and continued headwinds for the three public BGM companies.

Diabetes Drugs

2. The overall insulin market (basal, rapid-acting, and human insulins) fell 2% YOY in 1Q16 to $5.0 billion and remained flat in 2Q16 at $5.4 billion. The basal insulin analog market remained flat YOY in both 1Q16 and 2Q16, totaling $2.4 billion and $2.6 billion respectively, and its performance was largely attributable to continually declining revenues for Sanofi’s flagship product Lantus (insulin glargine). That said, Lantus continues to lead the market both in terms of sales revenue and new-to-brand prescription share (NBRx) and the launch of Novo Nordisk’s Tresiba (insulin degludec) in the US has thus far appeared to most impact Levemir’s (insulin detemir) market share. The rapid-acting insulin analog market fell 7% YOY in 1Q16 to $1.4 billion and remained flat YOY in 2Q16 at $1.5 billion – here we think more than just pricing pressure is going on. Namely, this is also a move away from MDI. Sales of human insulin from all three insulin manufacturers were flat in 1Q16 at $794 million and rose 1% YOY in 2Q16 to $775 million.

3. The DPP-4 inhibitor class grew 3% YOY in 1Q16 and 8% YOY in 2Q16, reaching $2.1 billion and $2.6 billion in sales, respectively. Merck’s Januvia franchise (sitagliptin) continued to lead the class by value, capturing 62% of the market in 1Q16 and 63% in 2Q16. We heard differing stances on DPP-4 inhibitor products from various pharmaceutical companies during the 2Q16 earnings season, matching the conflicting commentary we’ve heard from diabetes experts on the future of this drug class. We wish it could be tested for pre-diabetes since it could help so many.

4. The GLP-1 agonist class grew substantially in both 1Q16 and 2Q16 at 26% YOY and 29% YOY, respectively. These YOY growth rates are the highest for the class since 2013 and represent a continued rebound from when GLP-1 agonist sales slowed in mid-2014 to mid-2015. Overall GLP-1 agonist class revenue totaled $1.1 billion in 1Q16 and $1.2 billion in 2Q16. Novo Nordisk’s Victoza (liraglutide) maintained its market-leading position both in terms of revenue and total prescriptions (TRx), and Lilly’s Trulicity (dulaglutide) continued making inroads in the first half of 2016 – we expect the market as a whole will continue to grow with the positive LEADER cardiovascular outcomes results for Victoza and with new entrants to the class, including Novo Nordisk’s once-weekly injectable semaglutide, its once-daily oral semaglutide, and Intarcia’s implantable ITCA 650 (exenatide mini-pump). While we are very excited about fixed-dose combos of basal insulin and GLP-1, we continued to be concerned about payers’ willingness to cover the cost of combination therapy for patients (this is speculative as we’ve done no research yet here).

5. The SGLT-2 inhibitor class experienced impressive YOY growth of 47% and 49% in 1Q16 and 2Q16, reaching $606 million and $715 million in total revenue by our estimates – quite impressive sequential growth. By our estimates, J&J’s Invokana (canagliflozin) franchise continued to dominate the class with 54% of the total market share by value in both 1Q16 and 2Q16.

6. The negative sales trajectory for the now-generic thiazolidinedione class (TZDs) continued in 1Q16 and 2Q16. Lilly ceased breaking out Actos (pioglitazone) sales with its 1Q16 update and Takeda stopped breaking out its share of Actos revenue in 2Q16.

Diabetes Technology

7. Pooled global revenue for the “Big Three” BGM companies (Roche, J&J, Abbott) totaled ~$1.1 billion in 1Q16 and ~$1.3 billion 2Q16, falling 13% and 2% YOY, respectively. The YOY declines continue to be driven by US weakness, where combined sales plummeted 28% YOY in 1Q16 and 16% YOY in 2Q16. International sales fared better, declining 6% YOY in 1Q16 and growing (!) 4% YOY in 2Q16. (Note: This includes some non-BGM revenue for each company.)

8. By our estimates, the worldwide CGM market (Dexcom + estimated Medtronic) grew to ~$215 million in 1Q16 (up 42% YOY) and ~$228 million in 2Q16 (up 32% YOY). The latter was an all-time record in our model, and this trajectory puts CGM on track to nearly hit $1 billion in sales in 2016. Wow! The field is seeing strong momentum and competition as next-gen products come to market, and these figures exclude Abbott’s FreeStyle Libre (125,000+ EU users) since the company continues to exclude reporting of this product that has so much momentum.

9. We estimate the insulin pump market (excluding J&J and Roche) grew ~9% YOY in 1Q16 (~$483 million) and ~6% in 2Q16 (~$458 million). These were the two strongest quarters of YOY pump growth since late 2014 as Medtronic’s 640G drove international growth and Tandem and Insulet saw strong US sales. The field may soon see an acceleration with Medtronic’s ongoing launch of the MiniMed 630G, Tandem’s 4Q16 launch of t:slimX2, and an early 2017 launch for the long-awaited MiniMed 670G that was approved considerably earlier than expected.

Obesity

10. The obesity market is best understood from two angles – with and without Novo Nordisk’s Saxenda (liraglutide 3.0 mg). Overall, total sales rose 78% YOY in 1Q16 and 117% YOY in 2Q16, but this growth was driven exclusively by Saxenda, without which sales declined 13% YOY in 1Q16 and 20% YOY in 2Q16 – a very negative performance. Saxenda captured a quite unbelievable 51% of market share in 1Q16 and 63% in 2Q16. In less positive news, Zafgen terminated development of beloranib for the treatment of Prader-Willi Syndrome and obesity in July and GI Dynamics’ EndoBarrier device saw >100% YOY revenue decline in 2Q16.

Overall Industry Performance

• Diabetes revenue across drugs (excluding generics) and devices totaled an estimated ~$11.0 billion in 1Q16 and ~$12.2 billion in 2Q16, rising 3% year-over-year (YOY) in the first quarter and rising 6% YOY in the second quarter. This uptick in the industry occurs against an easy comparison, with 2015 having been the most challenging year for the diabetes field since we began tracking its performance in 2004 – 1Q15 growth was just 2% and 2Q15 diabetes revenue fell 1%, declining for the first time in our financial model. Overall, the field saw a slight improvement in performance in the first half of 2016; we look forward to seeing whether or not this trend is sustained through the rest of the year and beyond, though we expect pricing pressure to continue to pose significant challenges, particularly in the insulin market.   
• Newer drugs and technologies were bright spots in the field:
  
  • GLP-1 agonists accounted for 38% and 36% of industry growth in 1Q16 and 2Q16 (class growth was 29% in 2Q16, reaching $1.2 billion).
  • SGLT-2 inhibitors accounted for 34% and 31% of the overall growth of the industry in 1Q16 and 2Q16, respectively (the class grew 49% YOY to $715 million in 2Q16).
  • DPP-4 inhibitors contributed 10% and 22% to industry growth in 1Q16 and 2Q16 (class sales rose 7% to $2.6 billion in 2Q16).
  • CGM (Dexcom + estimated Medtronic) contributed 11% and 7% to overall industry growth in the first two quarters of the year (revenue rose 32% to ~$228 million in 2Q16).
  • Insulin pumps (excluding J&J and Roche; see explanation below) contributed 7% and 3% of industry growth in 1Q16 and 2Q16 (with total sales rising 6% to ~$458 million in 2Q16).
• On the other hand, the insulin and blood-glucose monitoring (BGM) markets saw continued pressure. The very large insulin market fell 2% YOY in 1Q16 to ~$5.0 billion and was flat in 2Q16 at ~$5.4 billion. Reported revenue from the three public BGM companies fell 13% in 1Q16 to ~$1.1 billion and fell 2% in 2Q16 to ~$1.3 billion.

Figure 1: Overall Quarterly Market Revenue (1Q12-2Q16) – Diabetes Drugs and Devices

• On the drug side, the growth of GLP-1 agonists and SGLT-2 inhibitors, coupled with the challenges of the basal and rapid-acting insulin markets, suggests that patients and providers may be embracing alternatives to insulin for glucose control and potentially delaying insulin initiation as a result. The potent combination of basal and postprandial glucose control, weight loss, low hypoglycemia risk, and potential cardioprotective benefits have won these agents a preferred position within the diabetes armamentarium for many prescribers. We expect these classes will only continue to grow, especially with new entrants to the GLP-1 agonist class and if cardioprotective class effects are demonstrated. While basal insulins could receive a boost with the entry of GLP-1 agonist/basal insulin fixed-ratio combinations at the end of this year, we are concerned that payers may not fund “combination use” of drugs like these (that may be unfounded of course!). Rapid-acting insulin analogs may be strengthened a bit with next-generation products such as Novo Nordisk’s Faster aspart (particularly if these improved insulins gain popularity for use in pumps or closed-loop systems) although overall we look for that to be a better but probably not standout product overall.
  
  • Looking at the industry’s drug pipeline, there are over 93 diabetes drugs in development including preclinical (21 drugs), phase 1 (18 drugs), phase 2 (20 drugs), phase 3 (15 drugs), filed for regulatory approval (8 drugs), suspended (3 drugs), and discontinued (8 drugs). A breakdown of these drugs by class (insulins, DPP-4 inhibitors, GLP-1 agonists, SLGT-2 inhibitors, TZDs, and glucagon) and phase can be found below – if you’d like to get a copy of the pipeline, please let us know.
• In diabetes technology, pumps are a mature category on the cusp of step-function innovation (automated insulin delivery – see our story about Medtronic’s 670G here), while CGM is still in its infancy ($0.2 billion in sales) but punching way above its weight (driving 7% of overall industry growth in 2Q16. Both of these are small categories relative to the overall field, but the potential is very high for them to expand in the coming years as next-generation products come out: a slew of automated insulin delivery (AID) systems (see our landscape) and improved CGM devices from Abbott (re-entering the US with FreeStyle Libre Pro this month and FreeStyle Libre soon [under FDA review]), Dexcom (G5 Android, new inserter/transmitter, G6, Verily partnership), and Medtronic (Guardian Connect, next-gen iPro devices). There is more runway for CGM to expand – given the potential in type 2 – and it will interesting to see how long it takes and how quickly its counterpart AID is adopted. What will the adoption curve for these technologies look like in the year to come? Will CGM have overtaken BGM sales in five years? 
• Total sales for all major players on the obesity drug market rose 78% YOY to $70 million in 1Q16 and 117% YOY to $91 million in 2Q16, but this growth was driven exclusively by Novo Nordisk’s Saxenda (liraglutide 3.0 mg). Without Saxenda, sales declined 13% YOY to $34 million in 1Q16 and 20% YOY to $34 million in 2Q16. By value, Saxenda held 51% of the market in 1Q16 and an even stronger 63% of the market in 2Q16. Qsymia, Contrave, and Belviq held 18%, 20%, and 11% market share in 1Q16, respectively, and 14%, 13%, and 10% in 2Q16, respectively. These smaller proportions in 2Q16 reflect lost market share to Saxenda, which is performing remarkably well despite major challenges in the obesity drug arena – namely resistance from patients and HCPs to implement pharmacological, non-lifestyle interventions for weight loss.

Figure 2: Overall Quarterly Market Revenue (1Q12-2Q16) – Obesity

Diabetes Drugs

Insulins

• The overall insulin market (basal, rapid-acting, and human insulins) fell 2% YOY in 1Q16 to $5.0 billion and remained flat in 2Q16 at $5.4 billion. The insulin analog market (including basal and rapid-acting) similarly fell 2% YOY in 1Q16 to $3.8 billion and remained flat in 2Q16 at $4.2 billion. Sequentially, the insulin analog market fell 13% in 1Q16 and recovered with 10% growth in 2Q16. For comparison, the overall insulin market (including human insulins) fell 5% in 2015 and the insulin analog market fell 6%, as reported in our previous earnings roundup. Of the three major insulin manufacturers, Lilly experienced insulin revenue growth in 2Q16 (8%), Novo Nordisk’s sales were flat, and Sanofi continued to face revenue declines (-4%). By class, basal insulin and human insulin sales both remained essentially flat in both 1Q16 and 2Q16. Rapid-acting insulin sales fell 7% in 1Q16, but were flat in 2Q16.The basal insulin analog market remained flat YOY in both 1Q16 and 2Q16, totaling $2.4 billion and $2.6 billion respectively. By value, Sanofi’s Lantus (insulin glargine) led the basal insulin market with 63% of sales in 2Q16 and Novo Nordisk’s Levemir (insulin detemir) held 25% market share. Next-generation basal insulins, Sanofi’s Toujeo (U300 insulin glargine) and Novo Nordisk’s Tresiba (insulin degludec), each held 6% of the market. Lilly/BI’s Basaglar (biosimilar insulin glargine) held <0.1% - but this one is on the move.
• The basal insulin market’s sluggish performance is largely attributable to continually declining revenues for Sanofi’s flagship product Lantus. Revenue for Lantus fell 11% YOY in constant currencies (12% as reported) to €1.4 billion (~$1.5 billion) in 1Q16 and fell 14% YOY as reported (11% in constant currencies) to €1.5 billion (~$1.7 billion) in 2Q16. As in previous quarters, Lantus’ sales decline was largely driven by US performance. US sales fell 18% YOY as reported in both 1Q16 and 2Q16 to €896 million (~$1 billion). Lantus sales also experienced YOY declines in the first half of 2016 in Europe and the rest of the world, but increased in emerging markets by 5% YOY operationally in 2Q16.
• Lantus continues to lead the basal insulin market in terms of new-to-brand prescriptions (NBRx) as well, while next-generation basal insulins appear to be largely drawing market share from Levemir in the first half of 2016. As of June 2016, Lantus led the market with 45% NBRx, followed by Levemir at 20%, Toujeo at 15%, and Tresiba at 11%. For comparison, Levemir held ~28% NBRx as of January 2016 and 22% as of April 2016 and Tresiba held 8% NBRx as of April 2016. Novo Nordisk management has acknowledged that Tresiba has cannibalized Levemir’s sales to a greater extent than the company anticipated. Toujeo’s NBRx share decreased slightly from April 2016 (16%) to June 2016 (15%) – while it’s still the early days, this suggests that perhaps a growing portion of patients and providers are trying out Tresiba now that it’s (finally!) launched in the US. Old faithful Lantus’ NBRx share has held steady in 2016 thus far, despite its patent expiry and declining revenue, though we expect we’ll see a greater impact in 2017 now that both CVS Health and UnitedHealthcare have announced that Lantus will be excluded from their formularies in favor of Basaglar (more on this below).
• During its 2Q16 update Novo Nordisk predicted that there will be a “bifurcation” of the basal insulin market in the next 12-18 months. In Q&A, Novo Nordisk explained that it sees Lantus, Basaglar, and Levemir (and to a certain extent Toujeo, though it has a slightly different action profile) in one category of basal insulins, whereas it envisions Tresiba in a category of its own because of its unique hypoglycemia reduction benefit, as demonstrated in the SWITCH trials. In fact, the company predicts that Tresiba will become the “clear leading product in the basal insulin segment” in two to three years. Time will tell whether this prediction is correct: Novo Nordisk recently submitted an application to the FDA to include the SWITCH data on Tresiba’s label, which, if approved, should help differentiate it from the rest of the market. We are thrilled to see so much discussion surrounding minimizing hypo risk, and very excited to see clinical trials using (and great acknowledgement of) outcomes beyond A1c. That said, we’re skeptical whether Novo Nordisk will be able to sufficiently differentiate Tresiba from Toujeo, given their similarly longer profiles of action. In particular, we wonder if primary care physicians and other non-endocrinologists may be more comfortable prescribing Toujeo given their familiarity with the Sanofi brand through years of Lantus experience.
• The basal insulin field is increasingly influenced by formulary exclusions wielded by PBMs. Two of the largest PBMs, CVS Health and UnitedHealthcare, recently announced that they will exclude Sanofi’s Lantus (insulin glargine) and Toujeo (U300 insulin glargine) from their 2017 formularies in favor of Lilly/BI’s biosimilar insulin glargine Basaglar, which is slated to launch in the US on December 15, 2016. Express Scripts’ formulary list, by contrast, includes both Lantus and Basaglar. Lantus has faced increasing competitive and pricing pressures since its patent expiry in early 2015 so, with the upcoming launch of a biosimilar option, CVS Health’s and United Healthcare’s formulary exclusions are not altogether surprising. That said, these decisions are potentially very disruptive to many patients as Lantus remains the unequivocal market leader within the basal insulin market, both in terms of value as measured by sales (64% as of 1Q16) and volume as measured by new-to-branch prescription (NBRx) market share (45% as of 1Q16). Unlike generics, biosimilar insulins are a new beast and some have expressed concerns about safety and quality control for these products, given the complexity and delicacy of the insulin manufacturing process – it remains to be seen how providers and patients will understand biosimilars. Formularies have typically not excluded any of the basal insulins, though previous formulary exclusions have significantly shifted the rapid-acting insulin landscape (with Express Scripts’ favoring Lilly’s Humalog [insulin lispro] and CVS Health favoring Novo Nordisk’s NovoLog [insulin aspart], to name a few examples). The inclusion of Basaglar on the formularies of all three large PBMs is a big win for Lilly/BI as the product looks to break into the US basal insulin market in an increasingly difficult pricing environment. Consistent with this, during its 2Q16 update, Novo Nordisk forecasted that the basal insulin space will eventually become a “three-player market” consisting of Novo Nordisk, Sanofi, and Lilly. Merck’s recently-submitted biosimilar insulin glargine may further shift the basal insulin market in coming quarters. An approval decision for the product, MK-1293, is expected in 2Q17, assuming a standard 12-month review. We look forward to seeing how the basal insulin market will be impacted by the emergence of this new competitor.
• Newer, long-acting GLP-1 agonists may pose a challenge to the basal insulin market. Drugs such as Lilly’s Trulicity (dulaglutide), GSK’s Tanzeum/Eperzan (albiglutide), and AZ’s Bydureon (exenatide) have balanced effects on fasting and postprandial glucose and may thus become an alternative for patients hoping to delay or minimize the use of insulin in their diabetes regimen. Trulicity in particular offers great patient convenience and ease-of-use advantages. This trend parallels the challenge in the rapid-acting insulin market from SGLT-2 inhibitors and GLP-1 agonists, non-insulin alternatives with targeted effects on postprandial glucose.
  
  • That said, the enormous excitement surrounding fixed-ratio basal insulin/GLP-1 agonist co-formulations could bolster sales for both drug classes. Novo Nordisk’s IDegLira (insulin degludec/liraglutide) is on the market in Europe (under the trade name Xultophy) and was recently submitted for approval in the US (decision expected in December 2016). Sanofi’s LixiLan/iGlarLixi (lixisenatide/insulin glargine) is also filed in the US, with a decision expected in November 2016, narrowly ahead of IDegLira as the expected first-to-market basal insulin/GLP-1 agonist co-formulation in the US. A decision for LixiLan’s approval in the EU is expected in 2017.                                  

Figure 3: Basal Insulin Market (1Q06-2Q16)

^{*Note that the Tresiba revenues here represent the entire Tresiba/Xultophy/Ryzodeg portfolio. Tresiba-only sales for 2Q16 were DKK 903 million (~$137 million).}

• The rapid-acting insulin analog market fell 7% YOY in 1Q16 to $1.4 billion and remained flat YOY in 2Q16 at $1.5 billion. All three of the major rapid-acting insulins (Lilly’s Humalog [insulin lispro], Novo Nordisk’s NovoLog/NovoRapid [insulin aspart], and Sanofi’s Apidra [insulin glulisine]), experienced sequential increases (16%, 6%, and 9% respectively, as reported) in 2Q16, a refreshing uptick after universal YOY declines in 1Q16. By value, NovoLog remains in the lead with 48% of sales in the rapid-acting insulin market. Humalog revenue was close behind with 45% of the market, and Apidra sales accounted for 7%. MannKind’s inhaled insulin Afrezza (formerly partnered with Sanofi) has not reported sales for 1Q16 or 2Q16, so it is not included in these calculations. Given its flat $2 million sales throughout 2015, it likely had a negligible impact on whole-market trends.
• The rapid-acting insulin market has been slowing in recent quarters, likely due to competition from the increasing popularity of the GLP-1 agonist class as an option for basal insulin therapy intensification. Time will tell whether this trend continues; we look forward to learning whether the launch of new products like Novo Nordisk’s Faster aspart, Lilly/Adocia’s BioChaperone Lispro, and Sanofi’s biosimilar insulin lispro can revitalize the rapid-acting insulin market in future quarters, though based on data at ADA 2016 it appears that these upcoming products offer only modest improvements over current rapid-acting insulin analogs. 

Figure 4: Rapid-Acting Insulin Market (1Q06-2Q16)

• Sales of human insulin from all three insulin manufacturers were flat in 1Q16 at $794 million and rose 1% YOY in 2Q16 to $775 million. Lilly’s Humulin experienced growth in 1Q16 (13%) and 2Q16 (5%), while Novo Nordisk’s human insulin experienced revenue declines in both quarters (-6% and -4%, as reported). Sanofi’s Insuman sales remained flat in 1Q16 and 2Q16.
• We have accounted for 39 insulin drugs, each of which are in different phases of development. Table 1 includes a detailed overview of the companies we are aware of with insulin drugs in development, including but not limited to Adocia, Biocon, Biodel, MannKind, Merck, Novo Nordisk, Oramed, and Sanofi. We acknowledge that this list may be incomplete. We will continuously update the table as timelines change.
  
  • Preclinical: 11
  • Phase 1: 7 drugs
  • Phase 2: 11 drugs
  • Phase 3: 4 drugs
  • Undergoing regulatory approval: 3 drugs
  • Suspended: 1 drug
  • Discontinued: 2 drugs

DPP-4 Inhibitors

• The DPP-4 inhibitor class grew 3% YOY in 1Q16 and 8% YOY in 2Q16, reaching $2.1 billion and $2.6 billion in sales, respectively. The class accounted for 10% of the overall growth of the diabetes drug and device industry in 1Q16 and 22% of the overall growth in 2Q16. For context, sales of DPP-4 inhibitors have been fluctuating for several quarters and seemed to be tapering off at the end of 2015, showing 9% YOY decline in 4Q15 and 1% YOY decline for 2015 overall. Their performance in the first half of this year thus occurs against an easy comparison, although the growth may indicate that DPP-4 inhibitors will continue to persevere despite increasing competition from SGLT-2 inhibitors and GLP-1 agonists. These latter drug classes have shown greater potency and have demonstrated more effective glucose lowering vs. DPP-4 inhibitors. On the other hand, patient/provider familiarity with DPP-4 inhibitors combined with their cleaner side effect and tolerability profile will most likely lead to their continued use in diabetes care, especially for older patients or those with renal impairment (though we expect SGLT-2 inhibitors will gain popularity among the latter population as well if their potential renal-protective benefits are substantiated). We’ve heard conflicting commentary from experts on the future of DPP-4 inhibitors – some emphasize their safety and efficacy in elderly patients or individuals with impaired kidney function, others highlight their easy dose administration, and still others point to their neutral CV effects in contrast to the potential cardioprotective benefits of SGLT-2 inhibitors or GLP-1 agonists. Ultimately, we think the revenue growth in 1Q16 and even greater growth in 2Q16 signal that DPP-4 inhibitors will remain a major player in diabetes care into the foreseeable future.

Figure 5: DPP-4 Inhibitor Market (1Q07-2Q16)

• By our estimates, Merck’s Januvia franchise (sitagliptin) continued to lead the class by value, capturing 62% of the market in 1Q16 and 63% in 2Q16. This represents a large majority of market share, followed by Lilly/BI’s Tradjenta (linagliptin) in a distant second with ~12% of the market in 1Q16 and ~13% in 2Q16. Novartis’ Galvus (vildagliptin) held 13% in 1Q16 and 12% in 2Q16; AZ’s Onglyza (saxagliptin) held 9% in 1Q16 and 7% in 2Q16; lastly, Takeda’s Nesina (alogliptin) held 4% in 1Q16 and 5% in 2Q16. Importantly, these percentages are based on our estimates of total Tradjenta franchise sales and are speculation only, since only Lilly’s portion of revenue is reported publicly. We estimate Lilly’s share of revenue at ~36% based on Lilly’s reported Tradjenta franchise sales for 2015 ($357 million) and global net sales for the franchise in 2015 (€909 million, or ~$1 billion) as reported in BI’s recent diabetes update. Most major DPP-4 inhibitor products experienced YOY growth in 2Q16, with the exception of AZ’s Onglyza, which fell 8% YOY as reported.
• Interestingly, different companies are prioritizing DPP-4 inhibitors differently within their diabetes portfolios, which reflects some of the conflicting views we’ve heard on the future of this drug class. During AZ’s 2Q16 update, management acknowledged competition from newer drug classes and the fact that Onglyza is facing significant pressure from SGLT-2 inhibitors. As such, AZ management discussed plans to position its SGLT-2 inhibitor Farxiga (dapagliflozin) as the preferred option for patients switching from DPP-4 inhibitor to SGLT-2 inhibitor therapy – commentary that suggested to us that AZ is willing to cannibalize some Onglyza sales to support Farxiga and is perhaps inclined to promote SGLT-2 inhibitor products over its DPP-4 inhibitor franchise. Meanwhile, other companies stood firm in the strength of their DPP-4 inhibitor agents during the 2Q16 earnings season. According to Novartis management, the company will continue to target Galvus to key segments of the patient population, including the elderly and individuals with renal impairment. Lilly management called attention to phase 3 data showing that Tradjenta effectively lowers A1c in type 2 diabetes patients at risk for kidney dysfunction, which indicates that the company is adopting a similar approach to Novartis in pushing DPP-4 inhibitors within key pockets of the diabetes patient population.
• We have accounted for five DPP-4 inhibitors, each of which are in different phases of development. Table 2 includes a detailed overview of the companies we are aware of with DPP-4 inhibitors in development, though we acknowledge that this list may be incomplete. We will continuously update the table as timelines change.
  
  • Preclinical: 1 drug
  • Phase 1: 1 drug
  • Phase 2: 0 drugs
  • Phase 3: 0 drugs
  • Undergoing regulatory approval: 2 drugs
  • Suspended: 0 drugs
  • Discontinued: 1 drug

GLP-1 Agonists

• The GLP-1 agonist class grew substantially in both 1Q16 and 2Q16 at 26% YOY and 29% YOY, respectively. These YOY growth rates are the highest for the class since 2013 and represent a continued rebound from when GLP-1 agonist sales slowed in mid-2014 to mid-2015. Overall GLP-1 agonist class revenue totaled $1.1 billion in 1Q16 and $1.2 billion in 2Q16. GLP-1 agonist revenue alone accounted for over 10% of total diabetes drug and device revenue in 1Q16 and 2Q16 and accounted for an estimated 38% and 36% of industry growth in 1Q16 and 2Q16. As expected, this market is growing substantially, and we believe it will surpass $5 billion for the full year 2016. By product, Lilly’s Trulicity (dulaglutide) drove a very impressive 51% and 55% of growth of the GLP-1 agonist class in 1Q16 and 2Q16, respectively. Novo Nordisk’s market-leading Victoza (liraglutide) continued to drive revenue growth as well, accounting for 32% and 30% of growth in 1Q16 and 2Q16, respectively. Rounding out the numbers, GSK’s Tanzeum (albiglutide) accounted for 12% and 8% of growth and AZ’s Bydureon (exenatide once-weekly) accounted for 5% and 6% of growth.
• Victoza retained the large majority of market share by value, accounting for 64% of GLP-1 agonist revenue in 1Q16 and 61% in 2Q16. That said, Lilly’s very patient-friendly, once-weekly contender Trulicity (dulaglutide) continued to make inroads in market share in the first half of 2016. In 1Q16 and 2Q16, Lilly’s Trulicity (dulaglutide) held 13% and 16% of the market share by value within the class, compared to 6% for full year 2015. Market share held steady throughout 1Q16 and 2Q16 for AZ’s Bydureon (exenatide once-weekly; 13%), AZ’s Byetta (exenatide twice-daily; 6%), GSK’s Tanzeum (albiglutide; 3%), and Sanofi’s Adlyxin/Lyxumia (lixisenatide; 1%). We expect sales of Adlyxin may increase in coming quarters, thanks to its recent approval in the US, which is by far the largest market for GLP-1 agonists.

Figure 6: GLP-1 Agonist Sales (1Q06-2Q16)

• Total prescription (TRx) volume growth of the GLP-1 agonist class reached ~32% in the US in 2Q16, according to Novo Nordisk’s 2Q16 update. Mirroring the revenue story, Victoza continues to lose TRx share, primarily to Trulicity. According to Novo Nordisk’s presentation slides, Victoza’s US TRx share declined to 52% in June 2016 (down from 54% in March of this year), while Trulicity’s TRx share increased to 20% as of June (up from 16% in March). The slides also report AZ’s exenatide franchise (Bydureon and Byetta) at 20% TRx share and GSK’s Tanzeum at 9%. In the US, Victoza held 59% market share by value in May 2016, down from 62% as reported by the company for February 2016 and 68% as reported by the company for May 2015. This trend holds true in almost all regions except Japan, where Victoza’s market share by value increased between May 2015 and May 2016.
• The positive LEADER results may have slightly boosted sales for both Victoza and the entire GLP-1 agonist class in 2Q16, though most of the financial quarter was over by the time full LEADER data were announced. We expect that a label update for Victoza reflecting cardioprotective benefit would result in further upswing and that the Victoza franchise will do well in future quarters as the market expands. Management forecasted that in the future the GLP-1 market will be split between Novo Nordisk and Lilly – while it’s too early to count out the other existing players, indeed, relative newcomer Trulicity has experienced very strong sales growth since its arrival on the market in 4Q14, garnering praise for the convenience of its IDEO-designed once-weekly dosing and extremely patient-friendly delivery pen. That said, we wouldn’t dismiss the other players down the pipeline at this point.
• We see additional market potential for GLP-1 agonists in multiple upcoming products. We expect Novo Nordisk’s share of the GLP-1 agonist market will return to faster growth with the launch of its once-weekly injectable semaglutide and its oral formulation of semaglutide – both promise to be quite potent and we were especially excited about the positive SUSTAIN 6 cardiovascular outcomes data for the injectable and the phase 2 data for the oral presented at EASD 2016. That said, formidable contenders for GLP-1 market share from other companies are on the horizon as well. We also believe that Intarcia could meaningfully expand the GLP-1 agonist market with its ITCA 650 implantable exenatide mini-pump. Future quarters will reveal how ITCA 650 and Novo Nordisk’s injectable and oral semaglutide affect the competitive landscape for GLP-1 agonists. Intarcia has suggested that ITCA 650 will be competitively priced, at a per-day discount to current GLP-1 agonist options, which will be a huge draw for many.
• We have accounted for 27 GLP-1 agonists, each of which are in different phases of development. Table 3 includes a detailed overview of the companies we are aware of with GLP-1 agonists in development, including but not limited to Lilly, Novo Nordisk, Transition Therapeutics, and Zosano. We acknowledge that this list may be incomplete. We will continuously update the table as timelines change.
  
  • Preclinical: 7 drugs
  • Phase 1: 8 drugs
  • Phase 2: 2 drugs
  • Phase 3: 4 drugs
  • Undergoing regulatory approval: 1 drug
  • Suspended: 0 drugs
  • Discontinued: 5 drugs

SGLT-2 Inhibitors

• The SGLT-2 inhibitor class experienced impressive YOY growth of 47% and 49% in 1Q16 and 2Q16, reaching $606 million and $715 million in total revenue by our estimates. Importantly, as referenced, these percentages are based on our estimates for total Jardiance revenue –as only Lilly’s share of revenue for Jardiance is publicly reported, these estimates are speculation only. We estimate Lilly’s share of Jardiance revenue at ~33%, based on a comparison between Lilly’s portion of full-year 2015 Jardiance revenue ($60 million) and total global net sales for 2015 from BI’s recent diabetes update (€165 million, or ~$183 million). We estimate SGLT-2 inhibitors accounted for 34% and 31% of the overall growth of the industry in 1Q16 and 2Q16, respectively.
• By our estimates, J&J’s Invokana (canagliflozin) franchise continued to dominate the class with 54% of the total market share by value in both 1Q16 and 2Q16. The vast majority (~90%) of Invokana’s revenue is derived from the large US market in particular, by virtue of its first-to-market position there. AZ’s Farxiga (dapagliflozin) held 27% and 29% of the SGLT-2 inhibitor market share in 1Q16 and 2Q16, respectively and Lilly/BI’s Jardiance (empagliflozin) franchise captured 19% and 17% of the market in 1Q16 and 2Q16. In its 1Q16 update, AZ attributed US gains in market share to the achievement of preferred status with a major health plan. This could refer to CVS Caremark’s formulary, which granted preferred status to Farxiga and Jardiance over Invokana for 2016. That said, management acknowledged the “competitive landscape” and the fact that Farxiga’s Medicare Part D formulary access could be stronger. We expect that much of the formulary challenges stem from Farxiga’s second-to-market status in the US – it was approved in the US in 1Q14, almost a full year after Invokana’s approval in March 2013.
• AZ management shared in its 2Q16 update that Farxiga leads the SGLT-2 inhibitor class globally by volume with 42% market share as measured by patient days on therapy (POTD). Based on AZ’s presentation slides, it appears that Invokana holds 29% market share in terms of POTD and Jardiance holds 19%, with 10% share going to “other SGLT-2s” (presumably including products such as Astellas’ Suglat [ipragliflozin], available only in Japan and South Korea). In the US in particular, management highlighted Farxiga’s >90% covered status for commercial plans – this is a big deal. Outside the US, Forxiga holds 60% market share in terms of POTD.
• In its 2Q16 update, Lilly management noted that growth of Jardiance within the SGLT-2 inhibitor class was lower than expected and attributed this to flattening new patient prescriptions for the SGLT-2 inhibitor class as a whole. Though Jardiance’s share of new therapy starts among endocrinologists exceeds 35% according to the most recent data, management commented that this number has started to plateau, fluctuating just above 30% since 4Q15. Share of new therapy starts among primary care physicians has also flattened at just above 20% since 4Q15. While growth in total prescription volume for the SGLT-2 inhibitor class remains strong at 30%, management shared that the number of new patients initiating SGLT-2 inhibitor therapy has essentially flat-lined. Management suggested that recent FDA and EMA warnings for the SGLT-2 inhibitor class may be contributing to the below-expected growth, namely the DKA warning that is affecting all products in the SGLT-2 inhibitor class. However, the company also emphasized that the strengthened bone fracture and kidney function warnings for some SGLT-2 inhibitors do not apply to Jardiance and that an updated label for Jardiance incorporating results from the EMPA-REG OUTCOME trial would likely serve as an inflection point to spark an increase in the drug’s total prescription (TRx) share and drive substantial sales growth overall. In particular, management suggested that the incorporation of a cardioprotective benefit into Jardiance’s indication could serve as a catalyst for overall SGLT-2 inhibitor class growth as well. We imagine the unexpected and unprecedented nature of the EMPA-REG OUTCOME results and the lack of consensus on the mechanism of cardioprotective benefit have hindered growth for the Jardiance franchise to some extent – this is reflected to some extent in the FDA’s 90-day delay on the decision to incorporation of EMPA-REG OUTCOME results into the Jardiance label.
• We have accounted for 11 SLGT-2 inhibitors, each of which are in different phases of development. Table 4 includes a detailed overview of the companies we are aware of with SLGT-2 inhibitors in development, including but not limited to AstraZeneca, Ionis, Merck, Novartis, and Pfizer. We acknowledge that this list may be incomplete. We will continuously update the table as timelines change.
  
  • Preclinical: 1 drug
  • Phase 1: 1 drug
  • Phase 2: 2 drugs
  • Phase 3: 5 drugs
  • Undergoing regulatory approval: 2 drugs
  • Suspended: 0 drugs
  • Discontinued: 0 drugs

Figure 7: SGLT-2 Inhibitor Sales (1Q13-2Q16)

TZDs

• The negative sales trajectory for the thiazolidinedione class (TZDs) continued in 1Q16 and 2Q16. This class no longer drives much interest due to its generic status and long list of safety concerns. Takeda’s Actos (pioglitazone; revenue split with Lilly) is the only major brand accruing revenue in the class and Takeda appears to be diminishing its attention to this drug, as evidenced by a steady decline in the reporting on Actos revenue throughout the first half of 2016. As in 4Q15, Takeda only released Japanese revenue for Actos in 1Q16, which declined 32% YOY and 35% sequentially to $13 million (1.5 billion JPY). 2Q16 marked the first time Takeda did not break out sales for Actos. Lilly similarly ceased breaking out Actos sales with its 1Q16 update, likely due to declining revenue for the product over the last few years due to loss of patent exclusivity and safety concerns over bladder cancer, heart failure, and bone fractures.
• The waning popularity and revenue of TZDs continues and this is as expected due to the class going generic. In 1Q16 the IRIS trial demonstrated that treatment with the generic TZD pioglitazone produced a relative risk reduction of 24% for the composite primary endpoint of fatal and non-fatal stroke and MI (HR=0.76; CI: 0.62-0.93, p<0.007). Secondary endpoint analyses presented at ADA in 2Q16 demonstrated a 52% risk reduction for progression to type 2 diabetes and reductions in insulin resistance and fasting plasma glucose (FPG) with pioglitazone in patients with elevated insulin resistance and high cardiovascular risk. Furthermore, we have heard significant optimistic commentary on the TZD class at recent meetings, particularly in relation to its role in Dr. Ralph DeFronzo’s very successful triple therapy trial. We do think there is still significant class overhang given the move to generic status – historically, the weight gain and rest of the side effect profile have been major negatives for consumers though we have heard multiple experts say that if patients can take a lower dose, some of this dissipates. Of course, we would expect the majority of patients to choose generic pioglitazone rather than branded Actos at this point, meaning that the downward trend in Actos revenue is unlikely to change course.

Diabetes Technology

BGM

• Pooled global revenue for the “Big Three” BGM companies (Roche, J&J, and Abbott) totaled ~$1.1 billion in 1Q16 (down 13% YOY) and ~$1.3 billion 2Q16 (down 2% YOY) Both performances come against easy comparisons to the prior year, as revenue declined 8% in 1Q15 and 11% in 2Q15. As the chart below shows, the field is still on a generally downward trend, with J&J faring the worst over the past four years (30% lower sales in 2Q16 vs. 2Q12), Abbott faring the best over that time (-14%), and Roche in the middle (-23%). We are now reporting Big Three BGM sales as Bayer’s former BGM business is now private under Ascensia Diabetes Care.
  
  • We’d note that these revenue estimates (and those below) report each company’s total Diabetes Care business, which all include a fraction of non-BGM revenue. J&J and Roche have insulin delivery and Abbott has FreeStyle Libre outside of the US. Since BGM is most of these companies’ business – and none give non-BGM estimates – we have chosen to include their entire Diabetes Care business’ in this section. As a result, the pump section excludes J&J’s and Roche’s insulin delivery sales, and the CGM section excludes sales of Abbott’s FreeStyle Libre.

Figure 8: Pooled Roche, Abbott, and J&J Quarterly Sales (1Q11 – 2Q16)

* Includes non-BGM revenue for Abbott (CGM, FreeStyle Libre), J&J (pumps), and Roche (pumps).

• Declines in the US have driven the weakness – combined sales of ~$297 million in 1Q16 and ~$347 million in 2Q16 fell 28% and 16% YOY. YOY sales in the US have now fallen in 15 of the past 16 quarters (1Q15 being the lone exception), and the rest of 2016 doesn’t look much better given the arrival of the second round of CMS price cuts. 1H16’s US revenue of $644 million is the first in our financial model to land under $800 million for a half year – we wonder at what point these diminishing revenues will begin to negatively impact innovation.
• The Big Three fared better internationally in 1H16 – combined sales of ~$821 million in 1Q16 declined 6%, while sales of ~$979 million in 2Q16 grew 4% YOY. Both came against easy YOY comparisons, as sales declined 13% in 1Q15 and 2Q15. 2Q16 represented the first positive international growth since 3Q14’s 1% growth, and the largest since 3Q13 (+6%). All three companies saw improved growth OUS in 2Q16, led by Abbott’s 14% growth (clearly buoyed by uptake of Freestyle Libre). In the prior quarter (1Q16), Abbott was the only one of the Big Three to see positive OUS growth.
• Roche had the weakest 1H16 of the Big Three in terms of growth, with sales falling 9% YOY relative to 1H15. North America has been particularly challenging for Roche, with revenue plummeting to an all-time low of ~$48 million in 1Q16 (down 49% as reported YOY) and down a further 5% YOY in 2Q16 (~$97 million). Sales outside North America dropped a more modest 4% YOY in 1Q16 (~$398 million) and grew 2% YOY in 2Q16 (~$475 million). Challenging environment aside, we’re glad to see Roche still investing in the pipeline. The Accu-Chek Guide BGM system recently launched in Denmark, Switzerland, and Australia and received FDA clearance in September (US launch in 2017). In addition, this past April, Roche and mySugr announced a global partnership to directly integrate the Accu-Chek Connect meter into the world’s most popular diabetes app, mySugr Logbook (750,000+ users in 51 countries) – the partnership positions Roche to sell more strips to high frequency testers and leverage external innovation from an app expert.
• J&J LifeScan sales of ~$429 million in 1Q16 and ~$471 million in 2Q16 fell 11% and 2%, respectively. International performance was stronger than in the US, falling 9% YOY in 1Q16 and rising 5% YOY in 2Q16 vs. -15% and -17% stateside. Still, global sales have now fallen in 16 of the last 17 quarters. In February, J&J launched the Bluetooth-enabled OneTouch Verio Flex BGM in the US, but no sales metrics have been shared. A month later, the company partnered with WellDoc to integrate BlueStar software into the OneTouch Verio Flex BGM and Reveal App.
• Abbott sales of ~$243 million in 1Q16 and ~$283 million in 2Q16 declined 9% and grew 2% YOY, respectively. The company’s sales have been the steadiest of the Big Three over the past few years, and we expect growth to pick up as FreeStyle Libre continues to succeed in Europe (125,000+ users as of July’s 2Q16 call, roughly backing out to revenue of $45 million), and comes to the US initially through the blinded FreeStyle Libre Pro (approved last week). The real-time (consumer) version of FreeStyle Libre was submitted to the FDA in 3Q16, and based on July’s call, approval is expected in 1Q17. Abbott is rightly investing in FreeStyle Libre innovation over its traditional BGM business, and given the trends, we think this is a very prudent strategic decision.

CGM

• By our estimates, the worldwide CGM market (Dexcom + estimated Medtronic) grew to ~$215 million in 1Q16 (up 42% YOY) and ~$228 million in 2Q16 (up 32% YOY). The latter was an all-time record in our model, just eclipsing the previous record in 4Q15 (~$226 million). The field saw the typical Q4-Q1 seasonal decline in the US, but sales rebounded in Q2. This trajectory puts CGM on track to nearly hit $1 billion in sales in 2016 – wow! Dexcom’s performance was again the story in CGM, with record-high sales in 2Q16 ($137 million) contributing most of the category’s growth (81%). Dexcom management estimated current US CGM penetration in type 1s at 15%-20%, and believes Dexcom has a 70% market share. [Note: Abbott’s FreeStyle Libre and Senseonics’ Eversense are not included in these estimates, as neither company has shared specific sales figures yet.]

Figure 9: Pooled CGM Revenue, Estimated (1Q12 – 2Q16)

• We estimate that ~75% of CGM sales came from the US in 2Q16 (consistent with 2Q15), and the US drove ~77% of the field’s growth. This likely reflects: (i) Dexcom’s sales are still heavily US-weighted (~85%) and are driving the majority of the category’s growth; (ii) there is less reimbursement for CGM in Europe than in the US (though hopefully changing with the positive decision in Germany and potentially France and the UK coming up); and (iii) FreeStyle Libre is potentially cannibalizing some CGM sales in Europe (now 125,000+ users). 
• Dexcom CGM sales totaled $116 million in 1Q16 (up 60% YOY) and a record-high $137 million in 2Q16 (up 47% YOY). This marked 15 straight quarters of 47%+ YOY growth (since the launch of G4 in late 2012), though the law of large numbers is kicking in: this was Dexcom’s lowest quarterly YOY growth since 3Q12. Still, Dexcom estimates it increased US CGM penetration in type 1s by two percentage points in 1H16, equating to ~30,000 new patient adds (assuming ~1.5 million US type 1s) – that would put its installed base in the US at ~140,000. Over 60% of new patient adds in 2Q16 were MDI users, flipping Dexcom’s historic 60% pump/40% MDI split – this will be an important metric to follow as CGM moves to a broader non-pumper population, particularly as the DIaMonD study results are published and trickle to payers. Dexcom’s 2016 sales are expected in the range of $550 million-$575 million (37%-43% YOY growth), which some analysts felt was conservative. We see upside to exceed the guidance with a slew of pipeline launches coming in the US. Read our Dexcom 2Q16 report.
• We estimate Medtronic’s worldwide CGM sales totaled ~$99 million in 1Q16 (up 25% YOY) and ~$90 million in 2Q16 (up 13% YOY). [These are estimates and could be off in magnitude and direction. Medtronic confirmed with us late last year that roughly 20% of sales now come from CGM. The mix likely varies from quarter to quarter.] Assuming Medtronic’s geographic CGM sales split is broadly consistent with overall sales, US CGM revenue was an estimated ~$59 million in 1Q16 (up 18% YOY) and ~$53 million in 2Q16 (up 7% YOY) compared to international revenue of ~$41 million in 1Q16 (up 37% YOY) and $38 million in 2Q16 (up 23% YOY). The 640G/Enlite Enhanced are driving strong growth outside the US. Medtronic’s key near-term CGM launches include Guardian Connect in Europe in ~November 2016-January 2017 (standalone, Bluetooth-enabled mobile CGM targeted at MDIs) and the Enlite 3 CGM sensor in the US by April 2017 (in the just-approved MiniMed 670G hybrid closed loop and standalone Guardian Connect). Read our Medtronic 2Q16 report.
• Abbott has not officially reported sales for FreeStyle Libre, but shared there are 125,000+ users in Europe. EU Libre sales back out to roughly $45 million in 2Q16 if all patients are using two sensors per month (~120 euros per month). [We have not included any estimates for FreeStyle Libre sales in the CGM category sales noted above.] This is very impressive uptake less than two years into the launch, putting quarterly Libre sales at nearly half what Dexcom and Medtronic are doing – and of course, Libre is less expensive. We expect growth to pick up as FreeStyle Libre continues to succeed in Europe and comes to the US initially through the blinded FreeStyle Libre Pro (approved last week). The real-time consumer version of FreeStyle Libre was submitted to the FDA in 3Q16, and based on the July call, approval is expected in 1Q17.  Read our Abbott 2Q16 report.

Estimated CGM Market Shares by Sales (2Q16)

• Dexcom had an estimated ~69% share of the US CGM market by sales in 2Q16, up from ~62% in 2Q15 and 1Q16. This seems in the ballpark based on Dexcom’s 76% share of the diabetes market research company dQ&A patient panel as of 4Q15 (contact CEO Richard Wood) and 63% share of the T1D Exchange registry (per a presentation at Dexcom’s ATTD 2016 symposium). Dexcom has steadily grown its market share with a slew of products leveraging the G4 Platinum sensor (Share, G5, t:slim G4, Animas Vibe), though things will become more competitive soon when: (i) Medtronic launches Enlite 3 in the US (MiniMed 670G and Guardian Connect; by April 2017); and (ii) Abbott launches the consumer version of FreeStyle Libre (expected in 1Q17 per its 2Q16 call; it was submitted to the FDA in 3Q16).
• We estimate that Medtronic held ~67% of the international CGM market by sales in 2Q16. This was down slightly from ~70% in 2Q15 but flat with 1Q16 (67%). Medtronic has seen strong adoption of the MiniMed 640G/Enlite Enhanced in Europe, and assuming Guardian Connect is launched as expected in ~November 2016-January 2017, its international market share could expand.

Insulin Pumps

• We estimate the insulin pump market (excluding J&J and Roche) grew ~9% YOY in 1Q16 (~$483 million) and ~6% in 2Q16 (~$458 million). These were the two strongest quarters of YOY pump growth since late 2014. On the strength of Medtronic’s MiniMed 640G, international sales drove most of the industry’s gains in 1H16: the field grew ~20% (1Q16) and 13% (2Q16) YOY outside the US vs. 4% and 2% YOY in the US. This reverses the trend we saw in 2015, where insulin pump sales declined 7% outside the US for the year.
  
  • These pump industry figures ONLY include sales from Medtronic, Insulet, Tandem, and Cellnovo, departing from our previous roundups that also loosely estimated Roche and J&J sales at roughly $50 million each. We felt this quarter – and moving forward – that excluding J&J and Roche provides the best and most accurate picture of the pump and BGM fields, unless Roche or J&J choose to breakout the revenue. Our previous estimates of their insulin delivery sales were highly estimated and we feel more comfortable excluding them from this section of the report, particularly because most of their business is from BGM. Moving forward, we are reporting total J&J and Roche Diabetes Care sales in BGM section, even though they include some pump revenue. Please let us know if you disagree!
  • As we did in our last roundup, we are assuming ~80% of Medtronic Diabetes’ sales come from pumps. The company confirmed this estimate with us late last year, though the mix likely varies from quarter to quarter.
• The pump industry saw higher YOY growth in 1H16, particularly on momentum from Medtronic outside the US and Tandem and Insulet in the US. Will growth accelerate on the heels of upcoming near-term products? Medtronic’s MiniMed 630G launched in September (first new Medtronic pump in three years in the US) and the MiniMed 670G was approved last week (launching in Spring 2017). Meanwhile, Tandem’s t:slim X2 will launch in 4Q16.
  
  • We’d note that Dexcom + estimated Medtronic CGM sales in 2Q16 ($228 million) were half as big as the very mature insulin pump field (excluding J&J and Roche). CGM is obviously growing faster from a lower base, but it does make us wonder – will it overtake pumps in the next couple of years? Will it ever overtake BGM sales? Certainly there is more potential for CGM technology long-term, given the runway in type 2. Automated insulin delivery could provide a step-function change in the pump market near-term, though it’s hard to know how many MDI patients will make the leap. Will AID dramatically expand the market, or appeal mostly to current pumpers in the first generation? We’ll get a first look in Spring 2017 when the 670G is expected to launch in the US (see our latest AID competitive landscape for the full cadence of launches we expect to see).

Figure 10: Pooled Insulin Pump Revenue (Estimated Medtronic + Tandem + Insulet + Cellnovo; 2Q12 – 2Q16)

• Medtronic’s estimated worldwide insulin pump sales were ~$397 million in 1Q16 (~2% growth YOY) and ~$362 million in 2Q16 (~1% decline YOY). [Medtronic confirmed with us late last year that roughly 80% of its sales now come from pumps, with the remaining 20% from CGM. This could vary from quarter to quarter.] The international business is really driving growth for the company on continued strong adoption of the MiniMed 640G in Europe and Australia: we estimate the company’s OUS pump sales grew 12% YOY in 1Q16 and 8% YOY in 2Q16. Meanwhile, the US business is hurting: estimated pump sales were down 4% YOY in 1Q16 and declined 6% YOY in 2Q16 (the first quarters of negative YOY pump growth in the US since 2013). The US business should see an uptick in the upcoming quarters with the September launch of the MiniMed 630G (threshold suspend on the new pump platform), and of course, the long-awaited 670G (hybrid closed loop) in Spring 2017. Read our Medtronic 2Q16 report and our coverage of the last week’s earlier-than-expected 670G approval.
• Insulet’s OmniPod is hitting its stride under the new management team: 1Q16 sales totaled $66 million (up 54% YOY) and 2Q16 sales hit a record-high $73 million (up 39% YOY). The YOY comparisons were on the easier side, though absolute sales were strong across the board. US OmniPod revenue hit an all-time high of $56 million in 2Q16, a 24% YOY increase and a strong 11% sequential gain from 1Q16. International OmniPod sales also reached a record high in 2Q16 ($16.6 million), more than doubling YOY on an easy comparison to low sales in 2Q15. Management raised 2016 guidance to “upper teens” US growth and 65% international growth. Insulet is on track to grow its US and worldwide installed bases 20% in 2016, which would bring the global base from 85,000 at the end of 2015 to just over 100,000 at the end of 2016. In the pipeline, at ADA 2017, Insulet will “debut” its next-gen Bluetooth-enabled PDM, Bluetooth-enabled pod (confirmed for the first time), and mobile app to display key pump information and integrate with Dexcom’s G5 app. It’s not clear when it will be submitted to the FDA, but more color is expected at Insulet’s investor day in November. The company also started the first on-body trials of the OmniPod Horizon Automated Glucose Control System in mid-September. The 1Q16 call expected a market launch in late 2018. Read our Insulet 2Q16 report.
• Tandem’s sales totaled $20 million in 1Q16 (up 63% YOY) and $23 million in 2Q16 (up 46% YOY). The company shipped 4,582 pumps in 2Q16, a 38% YOY increase from 2Q15 and a 13% sequential increase from 1Q16. Demand remains high for the t:slim G4 (launched in September 2015), which accounted for 57% of all pumps shipped in the quarter – clearly this is not pent-up demand nearly a year into this launch. Tandem has now shipped over 42,000 pumps since initial commercialization in 2012. “Approximately half” of Tandem’s new customers are still new to pump therapy, a good sign it is expanding the market. In the wake of UHC’s selecting Medtronic as its “preferred” provider of insulin pumps, management lowered guidance for full-year revenue from $108-$115 million (as estimated in 1Q16) to $105-$110 million. On the plus side, Tandem will launch a new pump platform, t:slim X2, in 4Q16 – notably, it will enable online software “upgrade-ability” to Dexcom G5 and G6 CGM integration and support future upgrades to automated insulin delivery. We expect high demand for this pump from tech-savvy patients. Read our Tandem 2Q16 report.
• Cellnovo’s sales totaled €435,209 in 1Q16 (up 26% sequentially) and €317,164 in 2Q16 (down 27% sequentially). We look more at the sequential trend rather than YOY growth, given the low sales base in these early days of commercialization. Management attributed the sequential decline in 2Q16 to the decision to control the influx of new patients – only 83 new pumps were shipped in 2Q16, down 43% from the 145 pumps shipped in 1Q16 (452 cumulatively) – as production for the consumable element of the system has neared capacity. The manufacturing expansion/switchover to Flextronics (which began in September 2015) is projected to complete in 2H16. This must go well for Cellnovo, given the cash balance of just over €16 million (roughly enough for three more quarters, barring a raise). Read our Cellnovo 2Q16 report.
• We’re not sure of J&J Animas pump sales or growth. For context, combined LifeScan/Animas sales declined 11% in 1Q16 and declined 5% in 2Q16, though most of that is likely from challenges in BGM. In the pipeline, the company’s hypoglycemia-hyperglycemia minimizer is expected to enter a pivotal study in 4Q16. Separately, we’ve learned from J&J that the Animas Vibe with G5 integration is under FDA review (submitted in June). Read our J&J 2Q16 report.
• We’re not sure of Roche’s pump sales. For context, worldwide Roche Diabetes sales declined 11% YOY in 1Q16 and grew a modest 1% in 2Q16. Unlike updates in 2015, the first two quarterly updates in 2016 did not mention insulin pumps as a growth driver. However, the 1Q16 update did allude to a pump slated for launch “in 2017.” We are not sure what management was referring to: A sensor augmented pump with the new CGM? The Solo MicroPump (Roche’s patch pump acquired from Medingo in 2010)? Automated insulin delivery? Read our Roche 2Q16 report.

Obesity

• Total sales for all major players on the obesity drug market rose 78% YOY in 1Q16 and 117% YOY in 2Q16, but this growth was driven exclusively by Novo Nordisk’s Saxenda (liraglutide 3.0 mg). Without Saxenda, sales declined 13% YOY in 1Q16 and 20% YOY in 2Q16. Revenue in 1Q16 was $34 million for Vivus’ Qsymia (phentermine/topiramate extended-release), Arena/Eisai’s Belviq (lorcaserin), and Orexigen’s Contrave (naltrexone/bupropion extended-release), but increased to $70 million when adding Saxenda sales. Similarly, revenue from Qsymia, Belviq, and Contrave was $34 million in 2Q16 but increased to $91 million when adding Saxenda sales. Individually, Qsymia sales fell 2% YOY in 1Q16 (the product’s first-ever YOY drop) and 9% YOY in 2Q16, Belviq sales fell an astounding 49% YOY in 1Q16 and 24% YOY in 2Q16, and Contrave sales grew 22% YOY in 1Q16 but fell 23% YOY in 2Q16. Novo Nordisk only started breaking out sales for Saxenda in 1Q16, and revenue grew 55% sequentially to DKK 376 million ($57 million) in 2Q16.

Figure 11: Total Obesity Market Sales (1Q13-2Q16)

• By value, Saxenda held 51% of the market in 1Q16 and an even stronger 63% of the market in 2Q16. Qsymia, Contrave, and Belviq held 18%, 20%, and 11% market share in 1Q16, respectively, and 14%, 13%, and 10% in 2Q16, respectively. These smaller proportions in 2Q16 reflect lost market share to Saxenda, which is performing remarkably well despite major challenges in the obesity drug arena – namely resistance from patients and HCPs to implement pharmacological, non-lifestyle interventions for weight loss. During Novo Nordisk’s 2Q16 update, management expressed confidence in Saxenda’s continued strong performance as the drug is rolled out in more international markets and as promotional efforts for the drug in the US are ramped up. Management also highlighted Saxenda’s rapid growth in total volume of prescriptions (TRx). The product reached 16,000 prescriptions in June 2016, accounting for 10% of the obesity drug market by volume. Also in June 2016, TRx was 62,000 for Contrave (40% of the market by volume), 40,000 for Belviq (26% of the market by volume, marking a significant YOY decline in prescriptions), and 37,000 for Qsymia (24% of the market by volume, marking another YOY decline in prescriptions). Saxenda compensates for its smaller market share by volume with a high list price based on concentration of liraglutide, costing $1,068/month vs. ~$200-300/month for the other agents. Furthermore, despite its smaller TRx volume, it’s rapidly gaining traction as the other obesity drugs experience decreases in number of prescriptions – we fully expect to see greater market share by volume for Saxenda in future quarters.
• Notably, Zafgen terminated development of its obesity drug candidate beloranib in July 2016. The decision followed two thrombosis-related deaths in a phase 3 clinical trial for beloranib, a MetAP2 inhibitor that had previously shown efficacy in treating Prader-Willi Syndrome as well as severe obesity complicated by type 2 diabetes. Without a doubt, this was unfortunate news for the company, and was yet another reminder of the tough market for obesity drugs. Zafgen management announced plans to restructure resources around ZGN-1061, another MetAP2 inhibitor for the treatment of obesity complicated by type 2 diabetes, which recently initiated the multiple ascending dose portion of its phase 1 trial.
• On the obesity device side, sales of GI Dynamics’ EndoBarrier therapy declined >100% YOY in 2Q16, although interim results from the REVISE Diabesity trial show some early promise for an EndoBarrier/liraglutide combination. Management emphasized this positive interim data during the company’s 2Q16 update. While EndoBarrier treatment alone (n=24) was associated with a 1.2% average reduction in A1c and treatment with GLP-1 agonist liraglutide 1.8 mg alone (n=22) was associated with a 1.3% average reduction in A1c, treatment with both EndoBarrier and liraglutide 1.2 mg (n=24) led to a 2.1% average reduction in A1c alongside 11.3% average weight loss. Despite these results, this device must overcome numerous obstacles in quarters to come as reflected by the tremendous YOY revenue drop in 2Q16. Obesity devices have yet to make their mark as a viable therapeutic option. While resistance to weight loss medications is high, patient/provider reluctance to consider an obesity device is even greater.
• We have accounted for 25 obesity drugs, each of which are in different phases of development. Table 6 includes a more detailed overview of the companies we are aware of with obesity drugs in development, including GSK, Intarcia, Rhythm, and ViaCyte. We acknowledge that this list may be incomplete. We will continuously update the table as timelines change.
  
  • Preclinical: 6 drugs
  • Phase 1: 13 drugs
  • Phase 2: 4 drugs
  • Phase 3: 0 drugs
  • Undergoing regulatory approval: 0 drugs
  • Suspended: 1 drug

-- by Adam Brown, Abigail Dove, Helen Gao, Brian Levine, Payal Marathe, Sarah Odeh, Jennifer Zhao, and Kelly Close