Lilly 3Q18 – Diabetes portfolio +18% YOY to $2.4B driven by Trulicity (+55% YOY to $816M), Basaglar (+38% YOY to ~$404M), Jardiance (+31% YOY to ~$506M); Pooled basal + rapid-acting markets; Enthusiasm for phase 1-ready oral GLP-1 – November 6, 2018

Executive Highlights

  • Lilly reported 3Q18 financial results this morning in a call led by CEO Mr. Dave Ricks (press release, presentation slides, webcast). Sales of major diabetes products rose 18% YOY to $2.4 billion, from $2.0 billion in 3Q17, also dipping 3% sequentially from $2.4 billion in 2Q18. Growth was driven by Trulicity (71%), Basaglar (14%), Jardiance (10%), and Humulin (5%), and management emphasized sustained 30% volume growth in Q3 – wow! Additionally, Lilly refined (positively) its 2018 guidance from $24.0-$24.5 billion in 2Q18 to $24.3-$24.5 billion in 3Q18.

  • Trulicity posted stellar 55% YOY growth to $816 million in 3Q18, rising from $523 million in 3Q17. Sequential growth was somewhat low but still positive at 5%, building on 15% sequential growth to $780 million in 2Q18. For the first time, Trulicity leads the US market by volume with 43% of total prescriptions, but the real excitement surrounded positive topline REWIND results. While we’ll have to wait for the ADA 2019 presentation, Trulicity seems poised to become the first GLP-1 agonist indicated for MACE risk reduction in patients with type 2 diabetes and CV risk, not just established CV disease – what a win for the class and for patients!

  • Lilly’s share of Jardiance sales rose 31% YOY and 13% sequentially to $167 million, against $127 million in 3Q17 and $147 million in 2Q18. Assuming Lilly collects ~33% of total revenue (based on previously reported figures), this puts total Jardiance revenue at $506 million for 3Q18, growing from $385 million in 3Q17 and $445 million in 2Q18. And with positioning on CVS Health regained for 2019 (Jardiance was excluded for Invokana in 2018), management is optimistic about the SGLT-2’s growth potential in coming quarter. Also encouragingly, class growth is on the rise (~10% YOY in 3Q18, up from ~7% YOY in 2Q18), following widespread 2Q18 commentary on slowed underlying growth.

  • Basaglar growth slowed to 38% YOY and 0% sequentially; Lilly’s share of the BI-partnered biosimilar glargine totaled $201 million in 3Q18, vs. $146 million in 3Q17. We estimate total revenue, including BI’s share, at $402 million (this is highly speculative). Performance was balanced in the US (+37% YOY to $157 million) and OUS (+44% YOY to $44 million), but management attributed the lackluster showing to a confluence of Medicare Part D gains and policy changes that increase the coverage gap discount levied on manufacturers. 

    • Pooled basal insulin revenue totaled $2.4 billion in 3Q18, reflecting a 7% YOY decline and 2% sequential drop from $2.6 billion in 3Q17 and $2.5 billion in 2Q18. Based on estimated total Basaglar revenue, we calculated value shares of 43% for Sanofi’s Lantus, 17% for Basaglar, 16% for Novo Nordisk’s Levemir, 14% for Novo Nordisk’s Tresiba, and 10% for Sanofi’s Toujeo. In the US, Lantus remains the volume leader (39% of TRx), following by Levemir (22%), Tresiba (14%), Basaglar (11%), and Toujeo (~9%), when Medicaid prescriptions are excluded.

  • Humalog sales dropped 5% YOY and 14% sequentially to $665 million, following revenue of $696 million in 3Q17 and $770 million in 2Q18. This was driven by a weaker US performance (-12% YOY and -21% sequentially to $366 million), and management cited an overall 14% price decline vs. 3Q17 driven by (i) segment mix (i.e. higher Medicaid utilization); (ii) patient affordability programs; and (iii) unfavorable adjustments. That said, we consider both greater utilization in Medicaid and higher investment in affordability programs to be wins for patients.

  • Pooled sales of rapid-acting insulins (NovoLog, Humalog, Apidra, Fiasp, and Admelog) totaled $1.5 billion in 3Q18, a 7% YOY and 10% sequential decline from $1.6 billion in 3Q17 and $1.7 billion 2Q18. Novo Nordisk captured 47% of this revenue ($692 million from NovoLog, $26 million from Fiasp), Lilly 44% ($665 million from Humalog), and Sanofi the remaining 9% ($99 million from Apidra, $30 million from Admelog); volume shares are more difficult to estimate. Notably, only newcomers Fiasp (+20% sequentially) and Admelog (est. +78% sequentially) posted any growth, YOY or sequential, reflecting continued, strong pricing pressure especially in the US.

  • We will be back to readers on pooled sales of GLP-1 and basal insulin.

  • Tradjenta revenue dropped 12% YOY and 4% sequentially to $136 million (Lilly’s share), from $153 million in 3Q17 and $142 million in 2Q18. We estimate total Tradjenta revenue, including partner BI’s share, at $378 million worldwide for 3Q18, assuming Lilly still claims ~36% of sales. CARMELINA results have been submitted to FDA for a label update (reflecting resounding CV safety), and topline results for the CAROLINA CVOT comparing Tradjenta to sulfonylurea glimepiride are now expected in 2019 (hopefully early on, given an internal readout is anticipated in 2018).

  • In this pipeline, management highlighted the recently-acquired (from Chugai Pharmaceuticals) OWL833, a non-peptide oral GLP-1 agonist expected to enter phase 1 “soon.” Management emphasized that, as a small molecule, OWL833 should offer more flexible dosing and greater bioavailability than peptide-based oral GLP-1s, which they hope will make for an improved patient experience vs. Novo Nordisk’s phase 3 oral semaglutide. We’ll have to wait-and-see on the first in-human studies, and we also note that OWL833 joins Novo Nordisk’s new phase 1 oral GLP-1 agonist, OG2023SC. Also in 3Q18, we saw very strong phase 2b data for GLP-1/GIP dual agonist tirzepatide and a positive phase 3 readout on URLi (ultra-rapid insulin lispro).

Lilly reported 3Q18 financial results this morning in a call led by CEO Mr. Dave Ricks. See the press release, presentation slides, and listen to the call. Below, you’ll find highlights on pooled basal and rapid-acting insulin sales in 3Q18, each of Lilly’s diabetes products, and the company’s diabetes-related pipeline, plus a transcript of diabetes Q&A from the call.

3Q18 Financial Results for Lilly’s Major Diabetes Products


2Q18 Revenue (millions)

Year-Over-Year Reported (Operational) Growth

Sequential Reported Growth

Share of Total Diabetes Revenue

Share of Growth



+55% (+55%)






-5% (-3%)






+7% (+8%)






+38% (+38%)






+31% (+33%)






-12% (-11%)






+1% (+1%)




Total Diabetes






Table of Contents 

Pooled Market Highlights

1. Pooled Basal Insulin Market Falls 7% YOY and 2% Sequentially to $2.4 Billion; Basaglar Overtakes Levemir For Second-Place in Value Share; Next-Gen Basals Combine for One-Quarter of Basal Insulin Revenue

The basal insulin market fell 7% YOY to $2.4 billion in 3Q18, from a base of $2.6 billion in 3Q17. Sequentially, pooled sales fell 2% against a 1% sequential rise to $2.5 billion in 2Q18. This analysis includes Novo Nordisk’s Tresiba and Levemir, Sanofi’s Toujeo and Lantus, and Lilly/BI’s Basaglar (biosimilar Lantus). We’ve estimated Basaglar sales by doubling Lilly’s reported product revenue ($201 million in 3Q18), since BI’s share is not publicly disclosed and we assume that worldwide revenue is split 50/50 between the two companies (this is more likely an underestimate given Lilly collects ~33% of Jardiance and ~36% of Tradjenta revenue). Despite its recent freefall following loss of exclusivity, Lantus remains the market leader in terms of value, with $1 billion in 3Q18 sale (-20% YOY but +1% sequentially) reflecting 43% of pooled revenue (up slightly from 42% in 2Q18). Notably, Basaglar (based on our estimate) now holds second place in the pooled basal insulin market for the first time, overtaking Levemir in 3Q18 with 17% value share and $402 million in sales (+38% YOY). Levemir fell 19% to $398 million in 3Q18 but had held this second-place position since it entered the market in the mid-200s, consistently in the shadow of Sanofi’s Lantus; however, it has conceded the #2 position following a steady downward trend in place since mid-2015. Rounding out the basal insulin market are the next-gen basal insulins Tresiba and Toujeo. Tresiba claimed 14% of the market by value ($335 million, +23% YOY) while Toujeo held 10% ($249 million, +9% YOY), and we’re encouraged to see this class now comprise ~one-quarter of total basal insulin revenue.

Pooled Basal Insulin Sales (1Q05-3Q18)

  • In terms of volume in US markets, Lantus remains in the pole position with 39% of total basal insulin prescriptions, followed by Levemir at 22%, Tresiba at 14%, Basaglar at 11%, and Toujeo at ~9%. Novo Nordisk provided this US volume data in its 3Q18 presentation (slide 10); however, we do note this data excludes basal insulin prescriptions through Medicaid, which accounts for a substantial ~12% of the total retail basal insulin market. In terms of overall trends, the past two years have seen Lantus’ volume share drop from ~60% to its current position of 39%; much of this volume has been picked up by Tresiba and Basaglar, while Toujeo has essentially remained flat at ~9%. Contributing to this phenomenon is that the 2018 Medicare Part D formulary prefers Tresiba, Levemir, and Basaglar (meaning a lower patient copay) over Toujeo and Lantus – we often hear “Medicare doesn’t negotiate” and presumably that is not possibly true. Novo Nordisk leaves ~6% of the market unaccounted for on its slide, which we assume reflects human/regular insulin; this figure has not meaningfully changed in the past few years to our knowledge (in 3Q17, ~7% was unaccounted for).

  • Merck/Samsung’s decision to terminate commercialization of biosimilar insulin glargine Lusduna leaves only one biosimilar insulin glargine on the horizon. Merck/Samsung cited anticipated pricing pressure and cost of production in explaining their decision to withdraw the biosimilar candidate, which had already been approved in the EU and tentatively approved by FDA, pending solution of Sanofi’s patent infringement lawsuit. Pricing pressure certainly remains a significant headwind for the entire basal insulin class, and this decision reinforces just how intense these barriers can be to more robust market entry. With the exit of Lusduna, Mylan/Biocon partnered Semglee is the closest biosimilar insulin glargine to the market, and is on track for a 4Q18 EU launch and a 2020 rollout in the US, pending approval.

2. Rapid-Acting Insulin Sales Fall 7% YOY to $1.5 Billion as All Established Products Experience YOY and Sequential Decline; First-Ever Biosimilar Admelog Enters Market

The rapid-acting insulin market (including Novo Nordisk’s NovoLog and Fiasp, Sanofi’s Apidra and Admelog, and Lilly’s Humalog) declined 7% YOY and 10% sequentially to $1.5 billion against challenging comparisons of $1.6 billion (+8% YOY) and $1.7 billion (+6% YOY, -3% sequentially) in 3Q17 and 2Q18, respectively. By our calculations, Novo Nordisk captured a market-leading 47% share by value with $718 million in sales, $692 million of which were from leader NovoLog (-12% YOY, -7% sequentially). The remaining $26 million came from Fiasp (+20% sequentially). Lilly followed with 44% value share, solely from $665 in Humalog sales, and Sanofi collected the remaining 9% with $129 million in rapid-acting revenue ($99 million from Apidra, $30 million from first-to-market biosimilar Admelog – more on this below). It’s quite a bit trickier to decipher volume share, even in the US alone; Lilly’s 3Q18 earnings presentation (slide 34) shows that total US prescriptions grew at a rate of ~1% in 3Q18, with Humalog capturing ~48% of the US market as of September 30 (our understanding is that Humalog has a slight edge in the US vs. OUS). Based on value share within the market, we can reasonably conclude that the vast majority of the remaining 52% of prescriptions are ending up with Novo Nordisk. Given pricing pressure, volume shares likely closely follow value shares overall. Despite the small but consistent growth in total prescriptions for mealtime insulins over the past three years (also slide 34), the overall value of the mealtime insulin market has plateaued, as evidenced by the graph below. As we understand it, this can almost entirely be attributed to pricing pressure, particularly in the US – manufacturers have not been shy about identifying this competitive force as the primary headwind for mealtime insulin growth. To our understanding, pricing pressure is more intense here than in any other diabetes drug class, as rapid-acting insulins are seen as more interchangeable by payers than any other class – leading to intense negotiations and high rebates in order to gain formulary access. This is a disheartening reality, as we’ve heard many stories of patients unwillingly switched between NovoLog and Humalog – the fact is that, for a variety of reasons, patients can prefer and even have better outcomes on one vs. the other; we’ll never forget this stirring blog post from an individual who couldn’t get UnitedHealthcare to cover MannKind’s inhalable insulin Afrezza even though he benefited enormously from the product over NovoLog. Indeed, MannKind CEO Dr. Michael Castagna alluded to the stiff competition preventing Afrezza from significantly penetrating the mealtime market in the company’s 3Q18 call: “Keep in mind, our competition gives away $0.50 to $0.70 of every dollar spent in rebates to payers to ensure patient and provider access is difficult for new entrants like us.” While we don’t think these rebates are high to keep out new entrants, they do reflect the extremely convoluted system that determines drug access and pricing in the US – a system that is increasingly problematic for patients.

Pooled Rapid-Acting Insulin Sales (1Q06-3Q18)

  • The serious pricing pressure experienced by mealtime insulins is reflected in Humalog and NovoLog Sales by geography in 3Q18. For NovoLog, North American revenue fell 19% YOY as reported (-21% operationally) to $352 million (DKK 2.3 billion) while international sales dropped a more minor 3% YOY (0% operationally) to $340 million (DKK 2.2 billion). Humalog by comparison dove 12% YOY and 21% sequentially in the US to $365 million, compared to 6% YOY growth and 2% sequential decline OUS to $299 million. Overall, OUS sales continue to balance weaker US performances, though ex-US territories are all being impacted by similar effects, to a lesser degree.  

  • Only new-to-market mealtime insulins Admelog (estimated +78% sequentially to $30 million) and Fiasp (+20% sequentially to $26 million) grew. NovoLog (-12% YOY, -7% sequentially), Humalog (-5% YOY, -14% sequentially), and Apidra (-5% YOY, -8% sequentially) all fell YOY and sequentially as reported. When Novo Nordisk was asked about this decline in its 3Q18 Q&A, management noted that the short-acting market, in general, is a very stable market from both a market share and competitive dynamics point of perspective (more on this below), downplaying any potential concern about the long-term size of the market. To this end, we’ll be interested to see how Fiasp and Admelog affect the market moving forward. When Fiasp was launched in February we heard from SVP of Commercial Mr. David Moore that reimbursement would be built up slowly over 12 months, and our sense is that Novo Nordisk is slowly accumulating coverage before firing full-force on Fiasp promotion. As for Admelog, Sanofi has previously been very clear that the first-ever biosimilar mealtime is not expected to be a real source of revenue for the company’s diabetes portfolio until 2019 – after it has had a chance to build reimbursement – though management was certainly excited about the product during both prepared remarks and Q&A. In 3Q18, 97% of Admelog’s revenue ($29 million) came from the US, mostly from the Managed Medicaid channel, which was part of Sanofi’s launch strategy. This more or less reflects the general sense we get from other manufacturers in the rapid-acting insulin market: President of Lilly Diabetes and Lilly USA Mr. Enrique Conterno predicted in the company’s 2Q18 call that Admelog would likely have its biggest impact in the Medicaid channel but also maintained that Admelog would not pose a significant commercial threat (more on this below). The rest of Admelog’s 3Q18 US revenue was from cash channels, presumably through Sanofi’s Insulin VALyou Savings Program, which launched concurrently with Admelog in 2Q18 and caps out-of-pocket expenses at $99/vial or $149/pen pack in the US. The program was expanded the day after Sanofi’s 3Q18 call to include all Sanofi insulins.

    • Despite this being the first quarter in which sales for first-to-market biosimilar mealtime insulin Admelog were released, neither Novo Nordisk nor Lilly identified (openly) the product as a significant commercial threat. Given Lilly CEO Mr. Dave Rick’s comments from JPM 2018 that both Humalog and NovoLog prices are already at the floor (we aren’t sure what this means - possibly in terms of being able to turn a profit though we doubt it), we wonder what impact Admelog will realistically have on pricing pressure and patient affordability. Moreover, Admelog is the first biosimilar mealtime insulin to reach the market, and we have heard multiple times that multiple biosimilars are required to meaningfully reduce prices. On the other hand, perhaps Novo Nordisk’s and Lilly’s silence speaks to the former’s claim that the short-acting market, in general, is quite stable from both a market share and competitive dynamics point of view (see above). Considering that Fiasp is priced at parity to NovoLog in most major markets, including the US, both Novo Nordisk’s and Lilly’s comments on NovoLog and Humalog would presumably extend to Fiasp as well. (That said, although technically they are the same list price, a number of US payers like Aetna do not recognize the lower price and require exhausting prior authorizations. Ultimately, we do think Admelog will shake things up in the rapid-acting insulin segment over the next few years: If costs for Sanofi’s biosimilar are lower than the insulin analogs, it can presumably offer a better deal to PBMs and payers than Lilly and Novo Nordisk can on Humalog and NovoLog, below the “floor” they’re at now. To be sure, we’ll be on the lookout for any comments on this in the future.

Financial Highlights

3. Overall Diabetes Portfolio Rises 18% YOY to $2.4 Billion, Driven by Trulicity, Basaglar, Jardiance; Prescription Volume for Diabetes Products Continues 30% YOY Growth; Updated Financial Guidance

Lilly’s overall diabetes portfolio surged 18% YOY to $2.4 billion against a challenging comparison of 39% YOY growth to $2.0 billion in 3Q17 and dipped 3% sequentially from $2.4 billion in 2Q18. This represents slightly slowed YOY growth vs. +38% in 1Q18 and +29% in 2Q18 but is still a very impressive performance driven by human insulin Humulin (5% share of growth), SGLT-2 inhibitor Jardiance (10%), biosimilar basal insulin Basaglar (14% share of growth), and superstar GLP-1 agonist Trulicity (71% share of growth). By geography, Lilly’s overall diabetes portfolio OUS climbed 17% YOY and fell 1% sequentially to $767 million in 3Q18. As usual, the company’s revenue was driven by US sales (68% share of total revenue), which were up 18% YOY but down 4% sequentially to $1.6 billion in 3Q18. Management was particularly optimistic about continued, 30% US volume growth by its diabetes portfolio in 3Q18 (sustained from 31% in 2Q18 and 30% in 1Q18), noting that a volume-based growth strategy will be critical to sustained performance in the face of rising pricing pressure. Indeed, as noted by one analyst during Q&A, the relatively uninspiring sequential performance of Lilly’s diabetes portfolio (both US and OUS) was driven by a surprising drop in Humalog (-14% sequentially) and a flat performance from Basaglar, two products particularly vulnerable to pricing pressure. In response to these concerns, management specifically cited high rebates for Basaglar within Medicare Part D and changes in Humalog’s segment mix (more on both of these below). Of note, sequential growth in 3Q18 was only experienced by Trulicity (+5%), Jardiance (+13%), and Lilly’s Glucagon (+45%), and both Jardiance and Trulicity promise to be continued bright spots moving forward: Jardiance now holds half of all new-to-brand SGLT-2 prescriptions at Lilly and 41% of the total US market by volume, while Trulicity’s stellar performance to date should only be bolstered by newly-announced positive results from the REWIND CVOT. All in all, Lilly’s diabetes business is continuing to grow steadily from an already-high base (as a reminder, the company recently surpassed Merck to because the second-largest diabetes company), and Lilly’s overall financial growth is unparalleled across the major players in the diabetes industry. As a quick comparison, Novo Nordisk’s diabetes portfolio grew 5% in 3Q18 (after falling 2% YOY in 2Q18 and 5% YOY in 1Q18), and Sanofi’s diabetes portfolio continued its steady decline with an 11% YOY drop in 3Q18. AZ diabetes grew 10% YOY in 2Q18 and 6% YOY in 1Q18, though both of these increases were driven solely by SGLT-2 inhibitor Farxiga and AZ diabetes remains <$700 million quarterly.

  • Lilly narrowed its 2018 guidance from $24.0-$24.5 billion in 2Q18 to $24.3-$24.5 billion in 3Q18. Positively, the increase on the low end of the revenue range was due to “strong performance across the pharmaceutical portfolio, particularly in diabetes.” Indeed, Lilly’s diabetes portfolio has been the company’s key growth driver throughout 2018. Since the initial announcement in December 2017, Lilly has bumped its guidance up twice: first from $23-$23.5 billion to $23.7-$24.3 billion in 1Q18, then to $24.0-$24.5 billion in 2Q18 (also explicitly on the strength of its diabetes portfolio). Moving forward, management expects diabetes revenue growth to continue to be driven by new products including Trulicity, Basaglar, and Jardiance.

4. Trulicity Continues Skyward: 55% YOY and 5% Sequential Growth to $816 Million; Newly US Volume Leader at 43% TRx; Excitement Abounds Over Positive REWIND Results

Sales of GLP-1 agonist Trulicity (dulaglutide) continued to climb, rising 55% YOY to $816 million in 3Q18 from a base of $523 million in 3Q17. Sequentially, revenue rose a more modest 5%, albeit against a tougher comparison of 15% sequential growth to $780 million in 2Q18. Buzz around Trulicity was palpable on today’s call, reflecting excitement surrounding positive REWIND results (much more on this below), the GLP-1 agonist’s strong 3Q18 performance, and potential for sustained growth in coming years. Within Lilly’s overall diabetes portfolio, Trulicity claimed 35% of total revenue (up from 32% in 2Q18 – we only expect this number to increase) and drove an impressive 71% of growth in Lilly diabetes. In fact, this share of growth mark is the highest we’ve recorded for Trulicity since we first started calculating the metric in 1Q16 – yet another encouraging sign of Trulicity’s strong performance this quarter (though, of course, this number also depends on the performance of other products). Trulicity is now Lilly’s leading diabetes product by a fair revenue margin; in 2Q18, it inched past previous leader Humalog by ~$10 million, while this quarter the gap has widened to ~$150 million, partially due to a drop in Humalog sales ($816 million vs. $665 million).

By geography, Trulicity again posted strong numbers within the US, where sales grew 56% YOY and 6% sequentially to $646 million from bases of $413 million and $613 million in 3Q17 and 2Q18, respectively. Management highlighted that Trulicity’s uptake in the US is driven by higher demand for the individual product in addition to strong underlying growth (~28% YOY in September 2018) lifting the entire GLP-1 class (slide 26). On this note, Lilly reported that Trulicity has become market leader by volume in the US, holding ~43% of total prescriptions (slide 26 and reproduced below); Novo Nordisk reported the same data in its 3Q18 update. Internationally, sales rose 48% YOY and 2% sequentially to $170 million from bases of $115 million and $167 million in 3Q17 and 2Q18, respectively. Lilly expressed optimism about Trulicity’s potential abroad, noting strong growth patterns for the GLP-1 class internationally (“mid-20s [YOY]” growth) and the possibility for Trulicity to capitalize on the sizable basal insulin market in these areas by positioning itself as a first-line injectable. We view this as a tremendous opportunity for Trulicity/GLP-1s and patients alike and would love to see more patients presented with the option of GLP-1s as a first-line injectable ahead of basal insulins. We’ll be back with a pooled analysis of GLP-1s after AZ reports on Thursday.

Trulicity Sales (4Q14-3Q18)

  • Unsurprisingly, the recent announcement of topline REWIND results, demonstrating superiority of Trulicity over placebo in reducing risk of three-point MACE, was heavily featured on today’s call. By our count, REWIND was directly mentioned 19 times over the course of the ~one hour call, and for good reason: The trial will have enormous implications for patients, Lilly’s commercial prospects, and how GLP-1s are prescribed as a class. REWIND was consistently described as “precedent-setting” by management, as they underscored that the trial (i) enrolled the broadest population of type 2 diabetes patients to-date in a GLP-1 CVOT (69% had no established CVD), (ii) had the longest median follow up of any GLP-1 CVOT (5.4 years), and (iii) enrolled patients with a low baseline A1c (average of 7.3%). All of these components support Trulicity’s effectiveness in reducing CV risk in a broad, lower-risk population of patients with diabetes; however, we’ll definitely have to see the full data for REWIND’s primary vs. secondary prevention cohorts before drawing any strong conclusions. On this note, management was not able to provide any details beyond what was released in its announcement this Monday – although many did ask during the Q&A period. Instead, they reiterated that full data from the trial will be presented at ADA 2019 and submission to regulatory agencies will occur broadly in 2019, to hopefully support an updated label/indication by 2020. No speculation was entertained as to what type of label may be granted, although we fully expect Lilly to seek an indication for full three-point MACE reduction in patients with type 2, either with CV risk or established CVD. To be sure, these positive results should serve as a major tailwind for the already-strong Trulicity going forward.

    • Notably, management commented that REWIND results confirm a class-wide cardioprotective effect for GLP-1s. During Q&A, Mr. Conterno noted that REWIND results are “important for the [GLP-1] class in that it confirms what other GLP-1s have shown when it comes to seeing CV benefit.” REWIND is now the third GLP-1 CVOT to demonstrate superiority on three-point MACE (joining LEADER and SUSTAIN-6 for Novo Nordisk’s liraglutide and semaglutide, respectively, and HARMONY-Outcomes for GSK’s albiglutide). Meanwhile, EXSCEL (exenatide) is widely considered a positive trial despite a narrow miss on statistical superiority, and ELIXA (lixisenatide) is somewhat of an outlier because of its abnormally high-risk population (recent ACS patients) and the short-acting nature of lixisenatide (half-life of ~3 hours). We’ve seen academic thought leaders strongly coalesce around the idea of GLP-1s being universally cardioprotective and find it reassuring to see industry think likewise. Nevertheless, this isn’t to say that Lilly sees all GLP-1s as homogenous: In a light-hearted moment, Mr. Conterno remarked, “I cannot resist, but I heard on a Novo Nordisk call that they also mentioned the GLP-1s were not all the same, and I think we agree with them,” possibly touting Trulicity’s potential CV indication as superior to Novo Nordisk’s Victoza’s, which is limited to patients with established cardiovascular disease. Of course, this is more of a function of how REWIND vs. LEADER were designed and conducted rather than any difference in the molecules themselves, though we note that SUSTAIN-7 results demonstrate convincing superiority of Novo Nordisk’s Ozempic (semaglutide) over Trulicity on both A1c reduction and weight loss. In our view, however, comparing molecules within the class isn’t the most productive conversation to have, given that only ~1.3 million Americans are currently filling a GLP-1 agonist prescription and they continue to comprise <10% of second-line diabetes prescriptions. We’re simply ecstatic to see so many innovative agents in the GLP-1 class and hope to see improved access for patients as these products mature on the market.

5. Jardiance Climbs 31% YOY and 13% Sequentially to $167 million as US Sales Bounce Back; CVS Health Coverage Regained for 2019; Re-accelerating Class Growth Reflects Promising SGLT-2 Tailwinds; No Type 1 Update

Lilly’s share of revenue from BI-partnered SGLT-2 inhibitor Jardiance (empagliflozin) climbed 31% YOY and 13% sequentially to $167 million from $127 million in 3Q17 and $147 million in 2Q18. We assume that Lilly collects ~33% of total Jardiance sales as part of its diabetes alliance with BI (based on reported figures from 2015 of $60 million in Lilly revenue and $183 million in global net sales), leading us to estimate total Jardiance franchise revenue in 3Q18 at $506 million, growing from $385 million in 3Q17 and $445 million in 2Q18. The franchise – including standalone Jardiance, fixed-dose combination Synjardy (empagliflozin/metformin), and fixed-dose DPP-4/SGLT-2 combo Glyxambi (linagliptin/empagliflozin) – drove 10% of the growth in Lilly’s diabetes portfolio, also accounting for 7% of total diabetes sales). All in all, 3Q18 was a strong quarter for Jardiance following a small but concerning 3% sequential loss in 2Q18; while 31% YOY growth is the franchise’s lowest-ever posting, it occurred from a much higher base. By geography, OUS sales rose to $63 million, up 45% YOY from $43 million in 3Q17 and 2% sequentially from $62 million in 2Q18, driven by increased volume. US revenue bounced back after a significant, sequential dip of 10% in 2Q18, growing 24% YOY to $104 million from $84 million in 3Q17 and rising 22% sequentially from $86 million in 2Q18, against an easy comparison. For the US, management underscored increased demand and volume leadership as key growth drivers, noting that Jardiance currently holds 41% of prescription volume within the US SGLT-2 market and 50% of new-to-brand prescriptions – the latter seems particularly high to us. Management exuded optimism for their SGLT-2 inhibitor moving into 2019, as it has been reincluded in “a major account that it was excluded from in 2018” – presumably CVS Health’s national formulary, which excluded Jardiance in favor of J&J’s Invokana (canagliflozin) in 2018 (~25 million lives affected). With this inclusion, coverage for Jardiance starting in January 2019 will be above 90%, according to Mr. Conterno. We’ll be back with a pooled analysis of SGLT-2s after AZ reports on Thursday.

Lilly’s Reported Jardiance Sales (3Q14-3Q18)

  • Re-accelerating SGLT-2 inhibitor class growth reflects promising tailwinds for Jardiance and SGLT-2s alike, according to Mr. Conterno. SGLT-2 class growth in the US (based on total prescriptions) has surged to 9.5% in 3Q18, up from 6.5% in 2Q18 – its lowest point in the past two years according to slide 30. While this is still leagues away from the >25% YOY TRx growth maintained through 2017, it is an encouraging uptick following a relatively dismal outlook from SGLT-2 manufacturers in 2Q18. As Mr. Conterno sees it, there are two main drivers behind this improvement: (i) the recently-published ADA/EASD recommendation to use SGLT-2s (or GLP-1s) as a second-line therapy for patients with type 2 diabetes with established CV disease; and (ii) general consensus on cardioprotection as a class effect of SGLT-2s – partially from the ADA/EASD statement and partially from investment by competitors. On the latter, we imagine Mr. Conterno was alluding to positive results from DECLARE for AZ’s Farxiga (dapagliflozin) and recent FDA approval of Invokana’s CV indication for reducing three-point MACE in those with type 2 and established CVD. Despite Invokana receiving a broader indication than Jardiance’s indication for reducing CV death (both based on very similar data), Lilly management was positive about the class implications of this new indication. For what it’s worth, Mr. Conterno asserted that Jardiance is in the best position to capitalize on SGLT-2 tailwinds.

  • When asked about plans to file Jardiance for a type 1 indication, management remained disappointingly reticent. For context, Lilly presented positive results from the phase 3 EASE program (empagliflozin in type 1 diabetes) at EASD 2018, demonstrating significant improvements in time-in-range, A1c, and weight loss, even without elevated DKA at the lowest dose of empagliflozin (though efficacy was substantially reduced). According to Mr. Conterno, Lilly is “working with BI on the next step and hoping to share more soon.” Sanofi/Lexicon have already filed SGLT-1/2 dual inhibitor sotagliflozin for type 1 with both FDA and EMA, and AZ has filed Farxiga (dapagliflozin) for type 1 with EMA (FDA filing planned for 2H18; we expect an update on Thursday). Given that FDA’s PDUFA date for sotagliflozin is in March 2019 and the company anticipates an early 2019 Advisory Committee meeting, we wouldn’t be surprised if Lilly/BI were biding their time until they have a better sense of how both agencies will respond to these NDAs.

6. Basaglar Growth Slows to 38% YOY, 0% Sequentially at $201 Million; First Quarter Without Triple Digit YOY Increase; Volume Gains in Medicare Part D Offset by Pricing Pressure, Discounts

Basaglar (biosimilar insulin glargine) revenue growth was subdued in 3Q18, climbing 38% YOY to $201 million from a base of $146 million in 3Q17. Moreover, sales were sequentially flat (-<1% – technically a first-ever sequential loss, if miniscule) from a base of $202 million in 2Q18. This comes after triple-digit YOY growth rates for Basaglar in every quarter since this calculation could be made after revenue was first reported in 3Q15, including +261% in 1Q18 and +133% in 2Q18. Of course, Basaglar is now growing from a much higher base, making it much more difficult to sustain this growth, but it remains discouraging to see absolutely no sequential growth from the product. Importantly, revenue of $201 million represents only Lilly’s share of Basaglar sales; unfortunately, partner BI has yet to report Basaglar revenue in any form. Speculating that Lilly and BI split revenue 50/50, our very rough estimate for total global Basaglar revenue in 3Q18 is $402 million. This does not meaningfully affect YOY or sequential growth. Basaglar sales accounted for 9% of total diabetes revenue (slightly up from 8% in 2Q18) and 14% of the growth in Lilly diabetes, which was down significantly from 21% in 2Q18. By geography, Basaglar grew 37% YOY and 1% sequentially to $157 million in the US, from $115 million in 3Q17 and $157 million in 2Q18. OUS revenue grew 44% YOY and fell 3% sequentially to $44 million, from $31 million in 3Q17 and $45 million in 2Q18.

Basaglar Sales (3Q15-3Q18)

  • Management explained that price declines driven by new access for Basaglar in Medicare Part D have offset corresponding volume gains. In particular, Lilly remarked that 3Q18 alone saw a quarterly US price decline of 6% for the entire diabetes portfolio and that 3 percentage points of this decline were due to Basaglar’s access to Medicare Part D, which began in 2018. On this note, the Bipartisan Budget Act of 2018 and subsequent shift of certain costs to manufacturers in covering the Medicare Part D coverage gap continues to serve as a headwind for Basaglar and other diabetes therapies ­– of course, the intended upside is lower costs for patients. According to Lilly, Basaglar and other Lilly diabetes products are “in that [coverage gap] segment strongly” and will be affected by this legislation, which increased the discount that manufacturers must offer for drugs in the coverage gap from 50% to 70%. However, management did not seem fazed by these changes, remarking that “we’ll see a little bit of a headwind there” and otherwise expressing optimism over continued volume gains that should come from Medicare access. For our part, we hold some reservation over the dichotomy between Basaglar’s US volume gains (more on this below) and the relative stagnation in revenue growth – especially after 2Q18, where volume gains were coupled with robust revenue increases.

  • With respect to volume, Basaglar now holds ~17% of the US basal insulin market. This reflects only a slight increase from 2Q18, when Lilly reported a volume share of ~16%, but, as seen below, Basaglar does continues to slowly gain basal prescription share. We note that this statistic does not align with data presented by Novo Nordisk in its 3Q18 update (slide 10), which showed that Basaglar held ~11% of the total basal insulin market in the US at a similar time. However, Novo Nordisk’s data does exclude Medicaid coverage, which represents ~12% of the basal insulin market, and our best guess is that this difference is driving the discrepancy seen. Lilly also highlighted in its presentation slides that, in terms of US new-to-brand prescriptions, Basaglar is the second highest ranked product in the basal insulin market.

7. Humalog Posts $665 Million Following 5% YOY and 14% Sequential Loss; Higher Medicaid Usage and Patient Affordability Programs in US Drive Loss

Sales of mealtime insulin Humalog (insulin lispro) were $665 million in 3Q18, falling 5% YOY from $696 million in 3Q17 and plummeting 14% sequentially from $770 million in 2Q18. US revenue dove 12% YOY and 21% sequentially to $366 million, from $415 million and $465 million in 3Q17 and 2Q18, respectively. OUS, 3Q18 Humalog sales grew 6% YOY to $299 million from $281 million in 3Q17, also falling 2% sequentially from $305 million in 2Q18. In explaining this performance, Mr. Conterno said, “Clearly when it comes to Humalog, we did see a 14% price decline vis-à-vis Q3 of 2017,” seeming to attribute the entire YOY loss to a reduction in realized price. He further went on to attribute eight points to unfavorable segment mix, four points to patient affordability programs, and the remaining two points to an unfavorable adjustment vs. 3Q17. Importantly, the first of these likely signals greater utilization of Humalog within the Medicaid channel – the opposite of which drove US Humalog growth in 2Q18. From our perspective, we’re thrilled to see more patients on Medicaid accessing Humalog, and we wonder if this increased segment mix is at all an effect of Lilly’s new Diabetes Solution Center, which launched August 1 with an aim to connect US patients to their most affordable Lilly insulin. If so, we applaud Lilly’s commitment to prioritizing access and affordability of their insulins, especially in the face of significant pricing pressure within the mealtime insulin class. Similarly, we’re glad to see apparent greater investment into patient affordability programs. Notably, volume growth for Humalog in the US (which management also partially attributed to patient access programs) was identified as offsetting some of Humalog’s losses in both Lilly’s call and press release (page 12). However, we believe this growth is likely minor at best, as total US prescription growth for rapid-acting insulins was ~1% YOY during 3Q18 while Humalog’s volume share of the US market has fallen ~half a percent in the same time, according to slide 34 (we’ve yet to see volume broken down by product for 3Q18). OUS, Lilly cited increased volume slightly offset by lower realized price and unfavorable exchange rates. See above for a full breakdown of the rapid-acting insulin market in 3Q18.

Humalog Sales (1Q11-3Q18)

8. Tradjenta Falls 12% YOY to $136 Million; CARMELINA Data Submitted to FDA for Potential Label Update; CAROLINA Topline Results Expected in 2019

Lilly’s reported revenue for BI-partnered DPP-4 inhibitor Tradjenta (linagliptin) dropped 12% YOY to $136 million from a base of $153 million in 3Q17. Sequentially, revenue also fell 4% from $142 million in 2Q18, when sales climbed 1% sequentially. Tradjenta comprised 6% of total diabetes revenue, identical to its 6% mark in 2Q18. We estimate total Tradjenta franchise revenue, including BI’s portion (not reported publicly, seeing as BI remains a privately-held company) at $378 million worldwide for 3Q18 ($139 million in the US and $236 million OUS), compared to $394 million in 2Q18 and $425 million in 3Q17. These calculations assume that Lilly still collects ~36% of Tradjenta franchise sales in all geographies. Geographically, Tradjenta sales especially struggled within US markets in 3Q18, dropping a significant 27% YOY to $50 million, albeit against a tough comparison of an all-time high of mark of $68 million in 3Q17. Sequentially, sales in the US fell 11% from a somewhat tough comparison of 5% sequential growth to $57 million in 2Q18. Continuing an established trend for Tradjenta, the majority of sales in 3Q18 occurred in OUS markets, where the DPP-4 sold $85 million, growing 1% YOY and 1% sequentially. Total OUS sales for Tradjenta have outpaced US revenue for five-plus years.

  • Lilly is seeking an updated Tradjenta label following release of CARMELINA results and has submitted an sNDA to FDA; this update would reinforce the cardiovascular safety of Tradjenta. As a reminder, CARMELINA results were recently presented at EASD 2018 ahead of publication and demonstrated strong CV safety for linagliptin. Notably, CARMELINA offered very reassuring data on heart failure risk; the endpoint of hospitalization for heart failure trended solidly in favor of linagliptin vs. placebo (HR=0.90, 95% CI: 0.74-1.08). Heart failure continues to be a point of discussion for DPP-4 inhibitors, in light of a complex history involving a small but significant signal for increased risk in the SAVOR CVOT for AZ’s Onglyza (saxagliptin). While these findings are certainly very positive for Tradjenta, we don’t imagine the commercial implications will be huge: As a whole, DPP-4 inhibitors are not experiencing significant growth, and there was little concern (in our observation) that Tradjenta presented a cardiovascular threat.

  • Topline results for the CAROLINA CVOT comparing Tradjenta to sulfonylurea glimepiride are expected in 2019. Management had previously remarked that an internal readout was expected in 2018 and suggested the possibility of a public release by year-end, but to our knowledge this is the first exact timing Lilly has offered on a public release. We’re very excited about this innovative trial, which could be highly influential in shifting prescribing practices away from SUs and toward DPP-4s (and hopefully newer drug classes, as well) if it demonstrates Tradjenta’s superiority to glimepiride on CV outcomes. We’re reminded of a recent quote from Dr. Alice Cheng at Diabetes Canada 2018: “If CAROLINA shows that sulfonylureas have fewer CV events than DPP-4 inhibitors, then I will retire, because nothing will make sense at all.”

Tradjenta Sales (2Q11-3Q18)

Pipeline Highlights

9. High Hopes for Phase 1-Ready, Small Molecule Oral GLP-1 Agonist, Acquired in September from Chugai Pharmaceuticals

During 3Q18, Lilly acquired a phase 1-ready oral GLP-1 agonist from Chugai Pharmaceuticals, and management expressed serious confidence in and excitement about the candidate. OWL833 is a non-peptidic small molecule that Lilly intends to investigate for type 2 diabetes; as a reminder, Lilly paid $50 million upfront (also agreeing to undisclosed milestones and royalties) in exchange for worldwide development and commercialization rights for OWL833. Lilly management highlighted the addition of this candidate twice during prepared remarks, characterizing it as a key component of a strengthened early-stage pipeline. The first in-human (phase 1) trials are expected to start soon, though no precise timing has been given yet – we imagine 1H19 is a reasonable estimate. In describing the molecule and its role in Lilly’s portfolio, management emphasized that OWL833 is a non-peptide agonist, which makes it fundamentally different from peptide-based drugs – i.e., less vulnerable to degradation in the stomach and intestine. As such, Lilly anticipates higher bioavailability in humans than is seen with peptide-based oral GLP-1s, a difference they hope will contribute to a better patient experience and outcomes. vTv has taken a similar approach to oral GLP-1 development, advancing a small molecule to phase 2b to become the closest competitor to Novo Nordisk’s expected first-to-market oral semaglutide.

Management also anticipated improvements in how and when dosing would happen, alluding to the fact that Novo Nordisk’s phase 3 oral semaglutide has been tested using six hours of pre-dose fasting and 30 minutes of post-dose fasting to ensure adequate gut absorption and subsequent bioavailability. To be sure, we’ll have to wait and see whether Lilly’s high hopes for OWL833 seem reasonable after the first in-human studies, and we also think a significant emphasis should be placed on weight loss, which is known to meaningfully differ between injectable GLP-1s and is a tremendously important outcome for patients. We certainly appreciate continued investment in GLP-1-based candidates from Lilly. All of this said, we have heard thought leaders assert that the dosing requirements for oral semaglutide actually haven’t been much of a problem for trial participants – and PIONEER results thus far support impressive efficacy on both glucose lowering and weight loss. The protocol entails, typically, taking oral semaglutide upon waking in the morning and avoiding food or drink for only 30 minutes afterward. For many people, this won’t and hasn’t presented a substantial burden, but we also recognize that there are many patients – e.g., with irregular schedules – who may struggle to fit oral semaglutide dosing into their day. For Novo Nordisk’s part, the company is not sitting still on oral GLP-1s: Just last week during the company’s 3Q18 update, management announced a new, phase 1 oral GLP-1 agonist, OG2023SC. The molecule, from the series of compounds that led to semaglutide, has already entered phase 1 (n=72) and carries the possibility of either (i) higher efficacy vs. oral semaglutide or (ii) similar efficacy at a lower cost, through greater bioavailability – OG2023SC uses the same SNAC carrier system as oral semaglutide but is optimized for better oral exposure. Lilly didn’t offer much color on how it’s thinking about the positioning of OWL833 on cost vs. efficacy, so we’ll keep our eyes and ears peeled for more on this going forward. All in all, we’re very excited about where oral GLP-1s are headed and extremely optimistic about the potential of these new candidates to drive more patients toward the strong glucose-lowering, weight loss, and even cardiovascular benefits offered by these molecules.

Lilly Diabetes Pipeline Summary

The table below reflects the latest updates, as far as we are aware, on Lilly’s diabetes pipeline products. Items highlighted in yellow indicate notable changes to the pipeline in 3Q18.




Nasal glucagon


Under review at FDA and EMA on track with 1H18 timing and as per 2Q18 update; Acquired from Locemia; Real-world data presented at ADA 2017

Jardiance (empagliflozin) in type 1 diabetes

Phase 3

Phase 3 data presented at EASD 2018; EASE-2 and EASE-3 completed October 2017 and September 2017, respectively

Jardiance (empagliflozin) in heart failure

Phase 3

EMPEROR HF-Preserved and EMPEROR HF-Reduced initiated March 2017, both expected to complete June 2020; Two EMPERIAL studies initiated March 2018 to investigate effect of Jardiance on exercise capacity in heart failure patients, expected to complete June 2019

Jardiance (empagliflozin) in chronic kidney disease

Phase 3

EMPA-KIDNEY announced June 2017 and slated for November 15, 2018 start in collaboration with University of Oxford and Duke Clinical Research Institute

LY900014 (ultra-rapid-acting insulin lispro)

Phase 3

Topline phase 3 results released October 2018, including PRONTO-T1D and PRONTO-T2D; Phase 2 data presented at ADA 2017 (type 1, type 2)

High-dose dulaglutide (3 mg and 4.5 mg once-weekly)

Phase 3

Phase 3 study (AWARD-11) launched April 2018, expected to complete October 2019; Phase 2 data presented at ADA 2018; Phase 2 trial in people with type 2 on metformin monotherapy completed August 2017

DACRA-042 (dual amylin calcitonin receptor agonist)

Phase 2

Acquired through partnership with KeyBioscience in June 2017; No study timing shared

GIP/GLP-1 dual agonist

Phase 2

Phase 3 SURPASS program to begin by early 2019; Phase 2 data presented at EASD 2018; Phase 1 trial completed June 2017

Soluble glucagon

Phase 1

Not currently listed on company’s pipeline page; Announced in May 2016 R&D update; Candidate is a short-acting, soluble, stable glucagon; Potential use in bihormonal closed loop systems

Basal insulin/dulaglutide fixed-ratio combination

Phase 1

Likely a combination of once-weekly “next-generation basal insulin” and Trulicity to support once-weekly dosing; Added to pipeline in 4Q16

DACRA-089 (dual amylin calcitonin receptor agonist)

Phase 1

Acquired through partnership with KeyBioscience in June 2017; No study timing shared

GLP-1/glucagon dual agonist (once-weekly)

Phase 1

Advanced into phase 1 in 4Q16; Oxyntomodulin analog under development for type 2 diabetes and NASH; First announced in May 2016 R&D update

Automated insulin delivery system

Phase 1

Feasibility study with Dexcom CGM and in-house pump/closed loop algorithm completed February 2018, initiated December 2017

Next-generation basal insulin

Phase 1

Announced in May 2016 R&D update; Potential for combination with Trulicity

Beta cell encapsulation therapy for type 1 diabetes


Lilly enters partnership with Sigilon in April 2018; Sigilon will file IND; Afterward, Lilly will lead in-human trials

Long-acting once-weekly glucagon


Announced in May 2016 R&D update; Potential for co-formulation with Trulicity or with GIP/GLP-1 dual agonist

Oral GLP-1 agonist(s)


Announced in 1Q16, confirmed in May 2016 R&D update; Management reaffirms Lilly’s commitment at JPM 2018 and during 4Q17 call


Q: Could you provide any additional color on REWIND results? Why are they “precedent-setting,” and how can they be interpreted in the context of other GLP-1 CVOTs, particularly with respect to the healthier patient population?

Mr. Enrique Conterno (Head of Lilly Diabetes and Lilly USA, Lilly): REWIND is indeed a precedent-setting trial with respect to its enrolled population. It included a majority of patients without established cardiovascular disease. It had the longest median follow-up – over five years – and a low A1C baseline of 7.3%. I think what we can share at this stage is really what's on the press release. We will not be sharing more information on REWIND until our detailed presentation at ADA next year. We do intend to submit in 2019 for a new indication.

Q: What is your view on the incremental opportunity with REWIND? Any thoughts on how it might change the treatment paradigm? What might the impact be on the overall GLP-1 class?

Mr. Conterno: I don't want to speculate on the particular indication that we will get. Clearly, this is important for the class in that it confirms what other GLP-1s have shown when it comes to CV benefit. And, I'm sorry, I cannot resist, but I heard Novo Nordisk's call. They also mentioned that GLP-1s were not all the same, and I think we agree with them.

Q: On REWIND, June 2019 seems like a long time to wait for the results. You mentioned you plan to file the indication in 2019. When should we expect to see a new Trulicity label?

Mr. Conterno: We are planning to pursue an indication under a submission in 2019 and a standard review. We will be looking at a label update in 2020. Of course, we are working as fast as we can to try to expedite that as much as possible, and we'll be working with the regulatory authorities. Unfortunately, we won't be able to share the REWIND data prior to ADA.  

Q: Has REWIND met your internal expectation for relative risk reduction? And have you historically talked about what kind of incremental sales opportunity it could mean for you?

Mr. Conterno: This is a positive trial. Trulicity is showing superiority in a broad range of patients with Type 2 diabetes.

Commercial Dynamics

Q: Were there any one-time issues related to Humalog or Basaglar that affected this quarter’s results?

Mr. Conterno: First, I think we're very pleased with the overall volume growth when it comes to the diabetes portfolio. When it comes to Humalog, we saw a 14% price decline vis-a-vis 3Q17. We are seeing an unfavorable segment mix which accounts for about eight points of that 14. Patient affordability has also impacted the Insulin portfolio; that's another four points. Then we had an unfavorable adjustment as a comparator to last year's quarter. We do see a lot of volatility with Humalog. It is better to look at this product on a year-to-date basis, which shows mid-single digit erosion on price. The dynamics on Basaglar are slightly different, in that Basaglar only gained Part D access this year this year, so we are comparing on a segment that is highly rebated. That has an impact on price. When we look at 2Q18 versus 3Q18 for Basaglar, there is some price erosion but not the same magnitude that you see year-on-year.

Q: At some point, FDA may deem biosimilar insulins as interchangeable and substitutable with branded products. Do you think this is more likely than not to occur? Could that be something that happens as soon as 2019? If not 2019, when?

Mr. Conterno: We do not believe this is imminent. At best, it's likely a few years away. It will require the appropriate studies for the FDA to provide such a designation.

Q: What are some of Lilly’s tailwinds and headwinds for 2019?

Mr. Joshua Smiley (CFO): A headwind, certainly on the top line, will be the change in the donut hole provision of Medicare Part D – the move from 50% to 70%. As Enrique mentioned, Basaglar is involved in that segment strongly, so we'll see a little bit of a headwind there. However, I think we're really pleased with the volume growth we saw this quarter and expect the main components of that will continue into 2019. On the R&D side, we do have a very robust set of Phase 3 opportunities ahead of us, including tirzepatide, which will be starting in 2019.

Q: Could you talk to the barriers prohibiting broader Jardiance usage? Is it education? Reimbursement? Related to that, could you outline the Tier 2 formular status within Medicare for Jardiance especially? In particular, I’m wondering whether a second brand marketed at a net price rather than the current list price may be helpful in increasing penetration by reducing the burden on the patients and preparing for a world of net pricing, which seems to be potentially upon us?

Mr. Conterno: First, the class right now is growing in high single digits. More importantly, we are seeing new patient starts. We now see that in the 14%, 15% range. It reminds me of bit of the reacceleration of the GLP-1 class back when Trulicity was launched. Importantly, there are new guidelines that have been published by both ADA and EASD. We think those guidelines are a significant improvement. That's an important tailwind. Finally, we have improved access for Jardiance. As you know, we were excluded from one of the major accounts, and we are back as of January 1, 2019. That's going to be very important for access to Jardiance. Jardiance will be covered for at least 90% of patients January of 2019. That's an excellent foundation for us to capitalize on. An additional point, which I emphasized in the GLP-1 class, but it's relevant for now, SGLT2s, is that we have basically additional data and investments from competitors that are confirming the CV benefits of the class. Jardiance has the best opportunity to capitalize on that. You asked about Medicare, and whether we would be launching a second brand with Jardiance. Honestly, that's not where I see the issue for Jardiance when it comes to the use of it. Clearly, we always assess all opportunities for our brands, but we are much more focused on continuing to educate health care professionals to ensure that they understand the full benefit that Jardiance can offer patients. The new guidelines are going to be extremely helpful to do that.

Q: You mentioned the U.S. price hit from gaining access for Basaglar, Taltz, and Humalog. Was there a one-time volume gain from this reflected in the 17% U.S. volume increase?

Mr. Smiley: Yes, if you look at volume for the third quarter in the U.S., it grew 17%, which Dave and I both mentioned was the most significant volume growth we've seen this decade. Basaglar absolutely contributed to that growth. The access that we got in Medicare Part D is contributing to the 17% growth, as are the other elements that we mentioned around patient access and all of the new launches. It is a positive offset there. And you can see in total, even with that negative 6% growth, the U.S. business grew at 11% overall.

Q: What is your exposure to a step up in the donut hole in 2019?

Mr. Smiley: When we look across our portfolio, it's about $200 million of incremental impact in 2019.

Q: Can you talk more about the international growth opportunity for the GLP-1 class overall? When I look at the class, we're looking at about a third of sales relative to the U.S. Frankly, when we look at other large classes, the opportunity seems like it should be quite a bit greater. I would love to know what you see as the opportunity internationally? Or perhaps what the holdbacks are?

Mr. Conterno: There is a significant opportunity, just like in the US, outside of the US for the GLP-1 class. The GLP-1 class is growing in the mid-20s outside of the US in most markets, so very healthy growth. Something unique when we look internationally is the size of the basal insulin market, which represents a significant opportunity as we see GLP-1s becoming first-choice injectables prior to basal insulin. We see that playing out over time.


Q: With oral GLP-1, can you talk a little bit about what you hope to accomplish, especially with the great data on GIP/GLP-1? As we think about Novo Nordisk's oral semaglutide, there are obviously some challenges in terms of dosing. Can you talk about what you're looking for in terms of a go, no-go profile?

Dr. Daniel Skovronsky (CSO): Thank you for the question on our partnership with Chugai on this oral GLP-1 receptor agonist. This is still a pre-Phase 1 molecule, but we’re moving quickly towards human trials. An important comment here is that this is a non-peptidic agonist. So you should think of this very differently from an oral peptide. This is a small molecule approach. As such, we expect it to have in humans – and this is what we'll be testing – a much higher bioavailability than you would see for any kind of oral peptide approach. What does that mean for patients? We hope that will translate into a better experience, certainly in terms of how and when this molecule will be dosed for the people on the trials, and ultimately, in practice. That also should translate into improved outcomes. It's still very early, but we're excited about the potential here. I look forward to getting some human data.  

Q: Can you provide us with any updates on that Phase 3 trial for tirzepatide? We noticed yesterday that the SURPASS 4 was posted to

Mr. Conterno: As part of tirzepatide's Phase 3 program, we are planning to study three maintenance doses at 5, 10 and 15 mg. I think we've learned that we should titrate tirzepatide in smaller increments and over time. As we've shared, we have a titration study that we intend to share next year. Clearly we've learned a lot from the study in terms of optimal titration schedules. We are planning aggressively when it comes to starting this trial and enrollment. You'll be hearing more when it comes to our trial for Type 2 diabetes soon.

Q: Why is Lilly not filing for type 1 diabetes as an insulin adjunct in 2018, given the positive data?

Mr. Conterno: As part of the EASE studies we showed that Jardiance as an adjunct to insulin in patient with type 1 can add glycemic control. We saw this across all doses that we studied. At this point in time, we are working with BI on the regulatory next steps. We hope to be able to share more shortly.  


-- by Ann Carracher, Martin Kurian, Peter Rentzepis, and Kelly Close