Lilly 3Q17 – Strong quarter for diabetes, portfolio up 39% YOY; Trulicity >doubles YOY, drives nearly 50% of growth; Marked growth for Jardiance, Basaglar, Tradjenta; Enthusiasm for diabetes pipeline – October 24, 2017

Lilly provided its 3Q17 earnings update in a call this morning led by CEO Mr. Dave Ricks. In this full report, you’ll find itemized highlights covering the company’s major diabetes products and pipeline, followed by select Q&A. For a very quick take, see table 1 for key financial results and table 2 for an up-to-date pipeline summary. You can also view Lilly’s 3Q17 slide deck, listen to a webcast replay, and read the press release.

Big picture, 3Q17 was an incredibly strong quarter for Lilly Diabetes. As you’ll see in the table below, only Humulin experienced YOY sales decline, with all other products experiencing single-, double-, or triple-digit growth (as expected, the most dramatic growth came from newer products like GLP-1 agonist Trulicity, SGLT-2 inhibitor Jardiance, and biosimilar insulin Basaglar). We heard more discussion than usual on diabetes candidates in development during Q&A. Ultimately, what we gathered was clear enthusiasm from management for next-gen diabetes therapies in the pipeline, which will complement Lilly’s strong existing portfolio of diabetes drugs.

Table 1: 3Q17 Financial Results for Lilly’s Major Diabetes Products

Financial Highlights

1. Lilly’s overall diabetes portfolio grew 39% year-over-year (YOY) and 6% sequentially to $2 billion in 3Q17. Trulicity exceeded half a billion dollars in quarterly revenue and was once again the primary growth driver for the diabetes business – by our calculations, the GLP-1 agonist accounted for 49% of portfolio growth (compared to 26% of total portfolio revenue). Basaglar accounted for a 22% share of growth (compared to 7% of total revenue), while Jardiance accounted for a 14% share of growth (compared to 6% of total revenue). Humalog was responsible for 10% of portfolio growth, capturing 35% of overall diabetes sales, while DPP-4 inhibitor Tradjenta was responsible for 7% of growth, capturing 8% of overall sales. Reflecting on 3Q17, management mentioned “outstanding diabetes growth.”

2. GLP-1 agonist Trulicity continued its strong performance in 3Q17, with revenue more than doubling YOY to $528 million (from $244 million in 3Q16). The product drove 49% of diabetes portfolio growth, by our calculations, compared to 26% of total diabetes revenue. Trulicity seems on track to top $2 billion in 2017, compared to $249 million in 2015 and $926 million in 2016. Management was optimistic about underlying momentum in the GLP-1 agonist class, suggesting that new market entries like Novo Nordisk’s once-weekly semaglutide will spur whole class growth as opposed to stealing share from Trulicity. Management pointed to ~25% YOY growth in total GLP-1 prescriptions written in the US (TRx) in 3Q17, which was driven by PCP adoption. Still, GLP-1 agonist prescriptions represent <30% the volume of the basal insulin market, leaving ample opportunity for expansion to new patients. In addressing the SUSTAIN 6 retinopathy signal, management underscored Trulicity’s strong safety/efficacy profile, and suggested that a “nominal number of patients” get proper eye screening within diabetes care. On the flip side, experts at the FDA Advisory Committee meeting for semaglutide established that clear protocols are in place for retinopathy screening/treatment, arguing that this risk could be well-managed in the real world. Whether or not these best practices are implemented by the average diabetes care provider is another story, and we definitely want more real-world data on this front. That said, it seems unlikely that the retinopathy findings will prevent semaglutide’s approval, and structures seem to be in place for Novo Nordisk to successfully promote semaglutide despite any retinopathy concerns.

3. Lilly’s share of revenue from BI-partnered SGLT-2 inhibitor Jardiance (empagliflozin) more than doubled YOY and rose 23% sequentially to $127 million in 3Q17. The product drove 14% of growth for Lilly’s overall diabetes portfolio (and accounted for 6% of total sales). We estimate global sales of Jardiance (including BI’s portion) at $385 million for 3Q17 (this also represents >doubling YOY, from a base of $144 million in 3Q16). This is only speculation since BI does not report revenue publically, but we calculate it based on BI’s 2015 diabetes update, when the company listed global net sales for the Jardiance franchise at €165 million, or ~$183 million, while Lilly collected $60 million in revenue from Jardiance in 2015, putting Lilly’s share at ~33%. Management recognized underlying class growth, but pointed to the black box warning for amputations on the Invokana label (J&J’s canagliflozin), asserting that Jardiance has “quickly become the new standard” for starting a patient on SGLT-2 therapy. That said, there was notable disappointment from management at the exclusion of Jardiance from the CVS Health national formulary for 2018 (with Invokana added back as a preferred drug). All in all, we wouldn’t count Invokana out just yet, even though J&J’s SGLT-2 business did decline 19% YOY in 3Q17.

4. Lilly’s reported sales of BI-partnered Basaglar (biosimilar insulin glargine) jumped to $146 million from a low base of $19 million in 3Q16, representing ~seven-fold YOY growth and 68% sequential growth (from $87 million in 2Q17). Basaglar was the second biggest growth driver in Lilly’s diabetes portfolio, accounting for 22% by our calculations (vs. 7% of total revenue). US sales nearly doubled sequentially to $115 million – the biosimilar has shown encouraging growth since US launch in December 2016. Ex-US sales totaled $31 million in 3Q17, up 57% YOY and 13% sequentially.

5. Humalog (insulin lispro) posted $696 million in revenue, up 9% YOY and 3% sequentially. The product drove 10% of portfolio growth in 3Q17 (capturing 35% of total revenue). Very notably, management announced that Lilly will not pursue legal action against Sanofi’s biosimilar insulin lispro, EMA-approved in July and tentatively approved by the FDA under brand name Admelog in September. Presumably, this means US launch of Admelog may come sooner than we initially expected, but we await further commentary during Sanofi’s 3Q17 earnings call on November 2.

6. Humulin (human insulin) sales fell 7% YOY and 16% sequentially to $301 million. This decline was largely driven by a challenging ex-US market, where sales dropped 23% YOY to $98 million. By comparison, US sales rose 4% YOY in 3Q17 to $203 million.

7. Lilly’s share of revenue from BI-partnered DPP-4 inhibitor Tradjenta (linagliptin) totaled $153 million, up 33% YOY and 8% sequentially, also driving 7% of overall growth in Lilly’s diabetes portfolio (vs. 8% of total revenue). Despite fluctuating sales of-late for the overall DPP-4 inhibitor market, Tradjenta has experienced strong growth in 2017 to-date. We estimate total Tradjenta franchise revenue, including BI’s share, at $425 million. This is based on our estimate of 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 by BI in a diabetes update.

8. Sales of Lilly’s glucagon rose 2% YOY to reach $47 million in 3Q17 – an all-time high, based on our records.

Pipeline Highlights

9. Management highlighted that LY900014, Lilly’s internally-developed ultra-rapid-acting insulin candidate, has advanced into phase 3 for type 1 and type 2 diabetes. The initiation of phase 3 studies (PRONTO-T1D and PRONTO-T2D) comes on schedule with 3Q17 guidance, and these trials are expected to complete in September and February 2019, respectively.

10. Management announced that a phase 2 trial of high-dose dulaglutide (3.0 and 4.5 mg) has been completed (Trulicity is currently available as 0.75 mg or 1.5 mg once-weekly doses). No specific data was shared, but we’re eager to see these results. Novo Nordisk’s high-dose GLP-1 agonist liraglutide (3.0 mg, branded Saxenda) has shown profound efficacy in obesity and in diabetes prevention – we wonder if Lilly will consider an obesity indication for its high-dose GLP-1 formulation (the company has previously had a laser focus on type 1 and type 2 diabetes rather than obesity, but we could imagine this shifting if weight loss results are very positive for 3.0 mg/4.5 mg dulaglutide).

11. Lilly’s presentation slides noted the discontinuation of a once-weekly basal insulin candidate from phase 1. While this wasn’t explicitly discussed, management did comment on increasing competitive pressure in the basal insulin market. As Head of Diabetes and Senior Vice President Mr. Enrique Conterno put it: “The basal insulin class was relatively more open when we launched Basaglar. Now, the situation is such that it’s harder to create viable economics for the payer.” So it seems that Lilly has de-prioritized bringing a new long-acting insulin to market in favor of focusing on its existing insulin portfolio (Basaglar, Humalog, Humulin), on the later-stage ultra-rapid-acting candidate, and perhaps on the basal insulin/dulaglutide fixed-ratio combination in phase 1.

12. We noticed substantial enthusiasm for the diabetes pipeline during prepared remarks and Q&A, with management emphasizing the promise of all three candidates in phase 2: (i) DACRA-042 (a dual amylin/calcitonin receptor agonist acquired from KeyBioscience), (ii) high-dose dulaglutide, and (iii) a GIP/GLP-1 dual agonist (phase 2 study expected to complete May 2018).

Financial Highlights

1. Diabetes Portfolio Grows 39% YOY for Third Consecutive Quarter; Revenue Reaches $2 Billion on Strength of Trulicity, Basaglar, Jardiance

Lilly’s overall diabetes portfolio grew 39% year-over-year (YOY) and 6% sequentially to $2 billion in 3Q17. Impressively, this marks the third consecutive quarter of YOY growth in the 30%-40% range for Lilly Diabetes. Moreover, the portfolio faced relatively tough YOY and sequential comparisons in 3Q17, given the 14% YOY rise in 3Q16 and the 13% sequential rise in 2Q17. Management called out GLP-1 agonist Trulicity, biosimilar insulin Basaglar, and SGLT-2 inhibitor Jardiance for their contributions to worldwide volume growth. Trulicity exceeded half a billion dollars in quarterly revenue and was once again the primary growth driver for the diabetes business – by our calculations, the GLP-1 agonist accounted for 49% of portfolio growth (compared to 26% of total portfolio revenue). Basaglar accounted for a 22% share of growth (compared to 7% of total revenue), while Jardiance accounted for a 14% share of growth (compared to 6% of total revenue). Humalog was responsible for 10% of portfolio growth, capturing 35% of overall diabetes sales, while DPP-4 inhibitor Tradjenta was responsible for 7% of growth, capturing 8% of overall sales. Reflecting on 3Q17, management mentioned “outstanding diabetes growth.” This 3Q17 performance is consistent with Lilly’s forecast in its 2017 financial guidance that Trulicity, Basaglar, and Jardiance would be among the company’s main growth drivers for the year 2017, and we find it extremely encouraging to see diabetes therapies featured as bright spots in the company’s larger pharmaceutical portfolio.

• We’re particularly happy about the strong quarter for Lilly Diabetes following the company’s September announcement of cost-cutting measures and streamlining procedures affecting ~3,500 positions. The job cuts are coming primarily from a voluntary early retirement program and from site closures, including an R&D office in Bridgewater, NJ and the Lilly China Research and Development Center in Shanghai. These changes should be fully-implemented by December 31, 2017. When questioned about the rationale for cost-cutting, Lilly management reiterated that the program is intended to “free up cash flow and operating spending capacity to invest in R&D.” This is logical, given Lilly’s expansive pipeline of development programs. On the other hand, we also assume the efforts are partly motivated by challenges facing US pharmaceutical manufacturers (and on this note, Merck recently announced similar cuts for its US business); while we’re reassured by very strong US sales for Lilly Diabetes in 3Q17 ($1.3 billion, up 51% YOY) in this quarter, our guess is that overall profitability isn’t growing nearly as fast as sales. In any case, we are very disappointed to hear about the workforce reductions, and we applaud the affected employees for their service to people with diabetes and other diseases.

2. Trulicity Dazzles: Sales >Double YOY to $528 Million, Drives Nearly 50% Portfolio Growth

GLP-1 agonist Trulicity (dulaglutide) continued its strong performance in 3Q17, with revenue more than doubling YOY to $528 million (from $244 million in 3Q16). This marks a 10% sequential increase from $480 million in 2Q17. The product drove 49% of diabetes portfolio growth, by our calculations, compared to 26% of total diabetes revenue. Trulicity seems on track to top $2 billion in 2017, compared to $249 million in 2015 and $926 million in 2016. As we have come to expect for GLP-1 agonists, Trulicity’s growth was largely driven by the US market, where sales totaled $413 million (more than doubling YOY and up 8% sequentially). That said, ex-US revenue was also strong at $115 million (more than doubling YOY and up 16% sequentially). Management was optimistic about underlying momentum in the GLP-1 agonist class, pointing to ~25% YOY growth in total GLP-1 prescriptions written in the US (TRx) in 3Q17. We were pleased to hear that this was driven by PCP adoption. Still, GLP-1 agonist prescriptions represent <30% the volume of the basal insulin market, leaving ample opportunity for expansion to new patients. Trulicity’s market share by volume is growing, topping ~37% TRx in 3Q17 (up from ~20% in 3Q16) according to management. By value, Trulicity captured 30% of pooled GLP-1 agonist sales in 2Q17 (up from 26% in 1Q17), behind market-leader Novo Nordisk’s Victoza (liraglutide) with 56%. See our 2Q17 pooled class analysis for much more information on the GLP-1 agonist market (an overarching trend to note is the closing gap between Victoza’s and Trulicity’s share of total volume/sales, although we see more than enough room here for multiple agents to be successful). We’ll be back with an updated 3Q17 analysis after AZ’s earnings call on November 9 (reporting revenue from exenatide products Bydureon and Byetta).

• Trulicity was the subject of many inquiries during Q&A, leading Lilly management to comment on upcoming competition from Novo Nordisk’s once-weekly candidate semaglutide. Last week, an FDA Advisory Committee voted 16-0 in favor of semaglutide approval, and a regulatory decision is expected in 4Q17. On Lilly’s call, management maintained that new market entries will spur whole class growth as opposed to stealing share from Trulicity (we emphasize again how GLP-1 agonist prescriptions represent <30% of basal insulin prescriptions, so there’s plenty of room to grow). Management also underscored Trulicity’s strong safety/efficacy profile in light of the retinopathy signal from SUSTAIN 6 (the pre-market CVOT for semaglutide). Head of Diabetes and Senior Vice President Mr. Enrique Conterno suggested that a “nominal number of patients” get proper eye screening within diabetes care. He cited that ~30% of people with diabetes have some degree of retinopathy in their lifetime, while 3%-5% experience vision-threatening retinopathy. Management seemed to imply that retinopathy could have an adverse effect on semaglutide sales, but that it certainly won’t cloud over the entire GLP-1 agonist class. We actually think retinopathy risk will prove quite manageable in the real world for patients taking semaglutide. Experts at the FDA Advisory Committee meeting established that clear protocols are in place for retinopathy screening/treatment. Whether or not these best practices are implemented by the average diabetes care provider is another story, and we definitely want more real-world data on this front. That said, it seems unlikely that the retinopathy findings will prevent semaglutide’s approval, and structures seem to be in place for Novo Nordisk to successfully promote semaglutide despite any eye-related concerns. Moreover, it’s possible that retinopathy only appeared in SUSTAIN 6 (a two-year trial) due to “early worsening phenomenon,” when rapid A1c reductions cause a transient increase in retinopathy/neuropathy. The DCCT showed this early worsening on retinopathy, but over time, intensive glycemic control greatly reduced risk for diabetic eye disease, because persistent hyperglycemia is the root cause of microvascular complications after all. In this scenario, the retinopathy signal in SUSTAIN 6 would only be another sign of semaglutide’s potency and glycemic efficacy – indeed, topline results from Novo Nordisk’s SUSTAIN 7 showed a statistically significant ~0.4% A1c treatment difference favoring semaglutide over dulaglutide when compared head-to-head. Lilly management declined to comment specifically on SUSTAIN 7 without seeing the full results.
• We’re curious as to how the new CV indication for Victoza – FDA-approved in August – will influence dynamics in the GLP-1 agonist market. Victoza is the first GLP-1 agonist and second of all diabetes drugs (after Lilly/BI’s Jardiance) with a CV indication on its product label. Time will tell whether Trulicity will receive a similar indication down the line – REWIND is expected to complete in July 2018. On Lilly’s 2Q17 earnings call, when asked about these upcoming CVOT results, management stated “we like our chances.” They cited Lilly’s experience in designing/executing diabetes CVOTs (EMPA-REG OUTCOME broke ground as the first positive one). There’s debate in the field as to whether or not cardioprotection could be a class effect for GLP-1 agonists. Some argue for a divide between human GLP-1-based molecules (liraglutide, semaglutide, dulaglutide) vs. exendin-4-based molecules (exenatide, lixisenatide), but in our view, AZ’s EXSCEL trial for Bydureon (exenatide once-weekly) did more to support a class effect than to refute it. EXSCEL results were neutral, but the hazard ratio for three-point MACE just barely missed the statistical threshold for superiority (HR=0.91, 95% CI: 0.81-1.00, p<0.001 for non-inferiority, p=0.06 for superiority), and this could very well be attributed to the pragmatic trial design (a lower-risk population at baseline, no run-in period to exclude patients with poor adherence, wide range of concomitant medications permitted, etc.). Regardless, REWIND will contribute to our knowledgebase on GLP-1/CV effects, and we are eagerly awaiting the readout, especially as CV risk becomes a more central component of diabetes management.
• Intarcia’s ITCA 650 (implantable exenatide mini-pump) could also represent competition for Trulicity down the line, but not for some time due to a September Complete Response Letter (CRL) for the product. According to Intarcia’s announcement, management hopes to meet with FDA soon to discuss next steps, but the content of the CRL suggests that no new pivotal trials or long-term activities will be necessary, so we are anticipating a swift resubmission. The real prize in ITCA 650 would lie in very low injection burden and near-perfect adherence, but again, this mini-pump is unlikely to reach the commercial market in the near-future.

Figure 1: Trulicity Sales (4Q14-3Q17)

3. Jardiance Sales >Double YOY to $127 Million; Leads SGLT-2 Class in New Therapy Starts

Lilly’s share of revenue from BI-partnered SGLT-2 inhibitor Jardiance (empagliflozin) more than doubled YOY and rose 23% sequentially to $127 million. The franchise – including standalone Jardiance, fixed-dose combination Synjardy (empagliflozin/metformin), and fixed-dose DPP-4/SGLT-2 combo Glyxambi (linagliptin/empagliflozin) – drove 14% of growth in Lilly’s diabetes portfolio (and accounted for 6% of total sales), on par with recent quarters. We estimate total global sales of Jardiance in 3Q17 at $385 million (this also represents >doubling YOY, from a base of $144 million in 3Q16), though this is speculation since BI’s portion of revenue is not reported publically. In a full-year 2015 diabetes update, BI listed global net sales for the franchise at €165 million, or ~$183 million, while Lilly collected $60 million in revenue from Jardiance in 2015, putting Lilly’s share at ~33%. Lilly/BI’s SGLT-2 business is faring well across geographies: Jardiance franchise sales >doubled YOY in the US to $84 million, driven by increased market share and underlying class growth (as management explained), and revenue nearly tripled YOY outside the US to $43 million, driven by increased volume in several countries. Management noted that Jardiance is now the market leader in US NBRx (new-to-brand prescriptions are “north of 50%,” up from 45% in 2Q17) and new therapy starts. The product captured 63% of all SGLT-2 inhibitor prescriptions written by a US endocrinologist and 48% of all SGLT-2 scripts written by a US PCP in 3Q17 – without a doubt, this is a remarkable showing on market share. Management recognized underlying class growth, but pointed to the black box warning for amputations on the Invokana label (J&J’s canagliflozin), asserting that Jardiance has “quickly become the new standard” for starting a patient on SGLT-2 therapy. The impressive NBRx numbers support this, as does commentary from diabetes thought leaders: Following the CANVAS presentation at ADA, Dr. Daniel Drucker predicted a gradual attrition away from Invokana toward Jardiance, as new therapy starts favor the latter (near-equivalent cardioprotection without the amputation signal). More recently, at CMHC 2017, Dr. Jay Skyler shared that he’s switched all his patients on Invokana (there were few to begin with) over to Jardiance. On the other hand, there was notable disappointment from Lilly management at the exclusion of Jardiance from the CVS Health national formulary for 2018 (with Invokana added back as a preferred drug). Dr. Anne Peters has suggested that canagliflozin shows better glucose-lowering and weight loss vs. empagliflozin in her patients. She also explained that the commercial success of both products will likely depend largely on reimbursement status. All in all, we wouldn’t count Invokana out just yet, even though J&J’s SGLT-2 business did decline 19% YOY in 3Q17. The favorable positioning of canagliflozin products on the CVS Health formulary could be a significant tailwind for the franchise in 2018. Moreover, we caution against over-comparison between CANVAS and EMPA-REG OUTCOME due to key differences between the CVOTs (method of data collection for amputations, baseline population, etc.). There’s a chance that future post-hoc analyses of CANVAS will attribute the amputation signal to something other than the canagliflozin molecule, although this hasn’t been shown to-date. Lilly management emphasized the enormous opportunity for SGLT-2 inhibitor class growth, as only 10 million out of 160 million oral diabetes medicine prescriptions are for SGLT-2 agents. We fully agree with this assessment, and we think this story should be less about in-class competition and more about whole class growth. We’ll be back with our pooled class analysis after AZ reports 3Q17 Farxiga (dapagliflozin) sales on November 9. Can’t wait? See our 1Q17 + 2Q17 industry roundup – SGLT-2 inhibitors rose ~19% YOY in total sales and drove 19% of overall industry growth.

• As far as regulatory milestones, Lilly’s slide deck mentioned that Jardiance was approved in China in 3Q17. Importantly, the product label in China includes data on reduced risk for CV death in patients with type 2 diabetes and established CV disease. China is now home to the world’s largest type 2 diabetes epidemic: ~10% of the country’s nearly 1.4 billion people have diabetes, though less than half are diagnosed. Clearly, this is a massive market for diabetes pharmaceuticals. For certain, much more work needs to be done on education, screening/diagnosis, and patient access to advanced drugs (in China and elsewhere), but it’s encouraging to see the most effective diabetes therapies becoming available across the globe. Lilly also submitted Glyxambi (empagliflozin/linagliptin fixed-dose) to the Japanese PMDA in 3Q17.
• Management shared no updates on ongoing clinical trials of empagliflozin. The EMPEROR HF studies appear to be on track, according to ClinicalTrials.gov, and there’s still no word on timing for the planned CKD outcomes study of Jardiance. EMPEROR HF-Reduced (in patients with reduced ejection fraction) and EMPEROR HF-Preserved (in patients with preserved ejection fraction) both began in March 2017 to investigate Jardiance in heart failure. Both studies are still recruiting, and both are expected to complete in June 2020. ClinicalTrials.gov does not yet list any Lilly/BI-sponsored study of empagliflozin in CKD, though plans were announced back in June. Expanded indications for Jardiance in heart failure or kidney disease would have major influence on Lilly/BI’s SGLT-2 business. Interest in these expanded indications extends to all SGLT-2 manufacturers: AZ has also initiated trials of dapagliflozin in heart failure (Dapa-HF, expected to complete December 2019) and in CKD (Dapa-CKD, expected to complete November 2020), while J&J’s CREDENCE trial of canagliflozin in people with diabetic kidney disease is expected to complete in June 2019 (J&J management recently highlighted this readout as a potential bright spot for the Invokana franchise). Lastly, we noticed a possible delay in the EASE program for empagliflozin in type 1 diabetes. EASE-2 and EASE-3 were expected to complete in October and September, 2017, respectively, though both are still ongoing according to their ClinicalTrials.gov pages. Notably, AZ recently reported positive data on dapagliflozin in type 1 diabetes at EASD (the DEPICT 1 trial). See our pipeline summary (table 2) below for more details.

Figure 2: Jardiance Sales (3Q14-3Q17)

4. Basaglar Climbs Skyward, Growing ~7x YOY and 68% Sequentially to $146 Million

Lilly’s reported sales of BI-partnered Basaglar (biosimilar insulin glargine) jumped to $146 million from a low base of $19 million in 3Q16, representing ~seven-fold YOY growth and 68% sequential growth (from $87 million in 2Q17). Basaglar was the second biggest growth driver in Lilly’s diabetes portfolio, accounting for 22% by our calculations (vs. 7% of total revenue). US sales nearly doubled sequentially to $115 million – the biosimilar has shown encouraging growth since US launch in December 2016. Ex-US sales totaled $31 million in 3Q17, up 57% YOY and 13% sequentially (perhaps less striking, but still impressive, and not unexpected considering Abasaglar has been on the market longer in Europe vs. the US). Management underscored that total basal insulin prescription share (TRx) for Basaglar is now above Novo Nordisk’s Tresiba and Sanofi’s Toujeo, and the product is contributing to Lilly’s worldwide pharmaceutical volume growth. We attribute Basaglar’s commercial success at least in part to its exclusive positioning over Sanofi’s Lantus on the CVS Health and UnitedHealthcare formularies, plus equal footing to Lantus on the Express Scripts formulary (this was implemented in 2017 and will carry over into 2018). Stay tuned for our pooled class analysis of the basal insulin market after Sanofi reports on November 2 (at that point, we’ll have revenue numbers for 3Q17 from the big three insulin giants, also including Novo Nordisk).

• Of note, the FDA granted tentative approval to Merck’s Lusduna Nexvue (biosimilar insulin glargine) this past July, pending resolution of a patent infringement lawsuit from Sanofi. Merck has remained quiet on the candidate since this tentative approval, perhaps because it could be up to 30 months before Lusduna Nexvue is launched and impacting revenue. Lilly/BI faced a similar lawsuit from Sanofi over Basaglar, which was eventually settled with an agreement to delay US launch and a royalty deal in exchange for license of the disputed insulin glargine patents – still, this took nearly 30 months in total, following Sanofi’s filing of the lawsuit. We’re enthusiastic about Lusduna Nexvue’s potential to drive down basal insulin prices, to expand patient access, and to grow this class as a whole. Past experience in the generic marketplace suggests that two or more generics must be available in order to drive down prices appreciably. Biosimilars are an important addition to the basal insulin commercial landscape, offering a lower-cost option to patients, and we believe Lusduna Nexvue could really help rather than hurt the Basaglar business.

Figure 3: Basaglar Sales (3Q15-3Q17)

5. Humalog Sales Grow 9% YOY to $696 Million; Management Confirms – No Lawsuit Against Sanofi’s Biosimilar Insulin Lispro

Humalog (insulin lispro) posted $696 million in revenue, up 9% YOY and 3% sequentially. By geography, US sales grew 10% YOY to $415 million, while ex-US sales grew 7% YOY to $281 million. Although significantly from the all-time highs of over $800 million per quarter, this financial performance for 3Q17 is still relatively impressive, especially considering (i) ongoing pricing pressures in the US insulin market; (ii) increasing competition for rapid-acting insulins from SGLT-2 inhibitors and GLP-1 agonists (advanced therapy classes that address postprandial excursions with lower hypoglycemia risk); and (iii) relative move away from MDI for some patients. Humalog’s volume and sales growth occurs against the backdrop of slowing growth for the overall rapid-acting insulin market: In contrast to consistent 6%-8% YOY growth in total US rapid-acting insulin prescriptions (TRx) in 2014, 2017 has been marked by only 1%-2% YOY volume growth for the class (slide 35). Consistent with this, pooled class sales for rapid-acting insulins were down 7% YOY in 1Q16, up a modest 1% YOY in 2Q16, down 6% YOY in 3Q16, down 2% YOY in 4Q16, up 14% YOY in 1Q17 (a rare double-digit rise), and up again 1% YOY in 2Q17. According to Lilly’s presentation materials (slide 35), Humalog held ~48% of US TRx as of September 2017 and has consistently held a solid 47%-49% share since January 2016, up from a long plateau at 44% previously. Moreover, Humalog holds a sizeable value share of the market, capturing 43% of pooled sales ($1.6 billion total) in 2Q17; Novo Nordisk’s NovoRapid (insulin aspart) held 50%, while Sanofi’s Apidra (insulin glulisine) held the remaining 7% of whole class revenue. We’ll be back with a pooled class analysis on rapid-acting insulin for 3Q17 after Novo Nordisk’s and Sanofi’s earnings calls on November 1 and 2, respectively.

• Very notably, management announced that Lilly will not pursue legal action against Sanofi’s biosimilar insulin lispro, EMA-approved in July and tentatively approved by the FDA under brand name Admelog in September. Elaborating during Q&A, management explained that the only remaining patents for Humalog pertain specifically to the KwikPen device, which is “fundamentally different” from the device Sanofi plans to use for its biosimilar insulin lispro, thus leaving no basis for litigation. We learned in Sanofi’s EASD exhibit hall booth that Insulin lispro Sanofi (the EU brand name) is available in SoloStar pens (very familiar to patients), and we’d venture a guess that Admelog will also be dispensed in SoloStar pens for US patients and that this will be a popular route for Sanofi. Sanofi previously specified that although Admelog has met all regulatory requirements for commercial approval (including positive phase 3 data in type 1 and type 2 diabetes), the product will not hit US pharmacy shelves until all patent infringement disputes are settled. Presumably, Lilly’s lack of legal action against Sanofi indicates that US launch of Admelog may come sooner than we initially expected, but we await further commentary during Sanofi’s 3Q17 earnings call on November 2. Notably, Sanofi seems committed to making Insulin lispro Sanofi/Admelog a commercial success – the product occupied a substantial amount of square footage in the EASD exhibit, where it was more or less debuted with strong messaging about equivalent efficacy at a lower cost (“everything you’d expect from a mealtime insulin, without the brand name”). As the first-to-market biosimilar prandial insulin, Admelog stands to make a big splash, and we’re excited about the prospect of a more affordable alternative from the patient perspective. To be sure, there are some limitations to biosimilar insulins, namely reluctance from HCPs to prescribe them (although Sanofi’s extensive history in insulin manufacturing could be reassuring) and the notion that they aren’t as cost-saving as we might hope. We’ll have to wait-and-see how Admelog is priced relative to Humalog in the US.
  
  • Lilly management further emphasized that Humalog – unlike Sanofi’s biosimilar insulin lispro – boasts a full range of formulations, including concentrated Humalog U200 and the newly-launched Junior KwikPen. The disposable Junior KwikPen comes pre-filled with U100 insulin lispro and allows for half-unit dosing for children and others with low insulin requirements. The product hit US pharmacies on October 4, and is lighter/smaller than other half-unit insulin pens (including the NovoPen Junior for Novo Nordisk’s NovoRapid). We applaud Lilly for the introduction of this more patient-friendly option (an especially important factor for young children and adolescents with diabetes, for whom discretion translates to much higher quality of life).
• Though not mentioned on the call, earlier this week Lilly announced plans to pursue a $72 million upgrade to one of its Indianapolis manufacturing facilities. The upgrade will supply a new insulin vial filling line, one that comes with modern design and state-of-the-art technology. This investment at the ground floor will help Lilly meet growing demands for Humalog and Humulin, and will prepare the company for potential new insulins in the portfolio (namely, phase 3 ultra-rapid-acting insulin lispro – more on this in the pipeline updates section below). Given Lilly’s diversified diabetes portfolio, we view this investment as a strong vote of confidence in the durability of the company’s insulin business, despite current challenges in the market. No timeline was specified for the project. The announcement further notes that Lilly expects to continue these investments in US diabetes product manufacturing operations, which have totaled >$1.2 billion since 2012. Notably, Novo Nordisk also has a major manufacturing project in the works – a new $1.8 billion facility for GLP-1 agonist and insulin production in Clayton, NC is expected to be fully operational by 2020. Overall, we view both these moves as a reflection of growing demand for insulin and diabetes medications in general as diabetes prevalence continues to rise. Furthermore, the US location of both of these manufacturing projects is likely no coincidence, given that this is the largest market for Lilly’s and Novo Nordisk’s products.

Figure 4: Humalog Sales (1Q11-3Q17)

6. Humulin Sales Decline 7% YOY and 16% Sequentially to $301 Million

Humulin sales fell 7% YOY and 16% sequentially to $301 million. This decline was largely driven by a challenging ex-US market, where sales dropped 23% YOY to $98 million. By comparison, US sales rose 4% YOY in 3Q17 to $203 million. Humulin’s TRx share in the US has been falling in recent months, dropping to an all-time low of 51% in September 2017. This plunge in volume share comes after an all-time high point of >54% in 4Q16. We suspect these volume losses are to Novo Nordisk’s human insulin, which led the class by value in 1H17. We also note the backdrop of decline for this particular therapy class, with pooled sales down 10% YOY in 1Q17 and flat in 2Q17 at $776 million. The overall market for human insulin as measured by total prescriptions has also been falling 4%-6% YOY since January 2016, and is currently at a low of nearly 7% YOY decline as of June 30, 2017 (slide 39). We imagine this combination of diminishing TRx in a shrinking market will create further negative impacts on Humulin revenue in future quarters. That said, with the complexity of the insulin pricing environment, it is difficult to predict these dynamics with certainty. To this end, Humulin sales surprisingly rose 8% YOY in 2Q17, despite the same challenging landscape.  

Figure 5: Humulin Sales (1Q11-3Q17)

7. Tradjenta Rises Steadily to $153 Million, Up 33% YOY and 8% Sequentially

Lilly’s share of revenue from BI-partnered DPP-4 inhibitor Tradjenta (linagliptin) totaled $153 million in 3Q17, up 33% YOY and 8% sequentially. We estimate total Tradjenta franchise revenue, including BI’s share, at $425 million. This is based on our estimate of 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 by BI in a diabetes update. Also by our calculations, Tradjenta drove 7% of growth in Lilly’s overall diabetes portfolio and, comparable to past quarters, comprised 8% of total revenue for Lilly Diabetes in 3Q17. By geography, Tradjenta sales grew 43% YOY in the US to $68 million, and rose 26% YOY ex-US to $85 million. Despite fluctuating sales of-late for the overall DPP-4 inhibitor market, Tradjenta has experienced strong growth in 2017 to-date. Competition from GLP-1 agonists and SGLT-2 inhibitors is only intensifying – one drug from each class (Victoza, Jardiance) now boasts a CV indication, while the DPP-4 inhibitor class continues to face concern over heart failure. Notably, the FDA recently added heart failure warnings to all DPP-4 product labels – including Tradjenta, Merck’s Januvia (sitagliptin), Takeda’s Nesina (alogliptin), and AZ’s Onglyza (saxagliptin) – even though only the last of these showed a statistically significant risk increase for heart failure hospitalizations in a CVOT. Lilly/BI’s CVOTs for linagliptin are ongoing, and could further elucidate this signal, hopefully putting the issue to bed and establishing that heart failure is not a class effect. The CAROLINA trial (linagliptin vs. glimepiride, an SU) is expected to complete in March 2019, while the CARMELINA study (linagliptin vs. placebo) is expected to complete in December 2017. Overall, we have little doubt that DPP-4 inhibitor agents will continue to be a mainstay of type 2 diabetes management. Prescribers seem to rely on these trusted, well-studied, highly-tolerable, and familiar agents. DPP-4 inhibitors remain the preferred option for the elderly and for patients with renal impairment – interestingly, Dr. Per-Henrik Groop noted at EASD 2017 that Tradjenta is the only DPP-4 product that doesn’t have to be adjusted for dose even with falling eGFR. We’ll be back with a pooled class analysis of DPP-4 inhibitors after Novartis, Merck, Takeda, and AZ report their 3Q17 earnings. In the first half of 2017, we calculated a slight overall decrease in the DPP-4 class, but an uptick in Tradjenta market share to 16% in 2Q17, behind Januvia’s 61%.

Figure 6: Tradjenta Sales (2Q11-3Q17)

8. Glucagon Grows 2% YOY, 44% Sequentially to $47 Million

Sales of Lilly’s glucagon rose 2% YOY to $47 million in 3Q17. Sequentially, sales grew 44% to hit what is, based on our records, an all-time high. That said, we also point out the seasonality of glucagon sales, which have historically peaked in the third quarter (we hypothesize this is due to families buying new glucagon kits as children go back to school). As usual, Lilly’s glucagon revenue was driven by US sales totaling $45 million (up 2% YOY and 45% sequentially against a relatively easy comparison of 2% sequential decline in 2Q17), while ex-US sales accounted for the remaining ~$1.5 million (up 1% YOY and 20% sequentially) and have hovered between $1-$3 million in quarterly revenue since 1Q12. From our perspective, Lilly is well-positioned for innovation and growth in glucagon, having acquired phase 3 nasal glucagon from Locemia (with FDA submission expected by 1H18, per Dr. Jessica Castle’s DTM 2016 presentation – we’d certainly love to hear confirmation of this from company reps). Lilly also has a stable soluble glucagon candidate in phase 1, though no updates were shared on the 3Q17 call. We saw promising real-world data on Lilly’s nasal glucagon candidate at ADA 2017, showing that 96% of patients experiencing severe hypoglycemia recovered within 30 minutes following treatment and that most of caregivers were highly-satisfied with the device. See our glucagon competitive landscape for an overview of this therapeutic area, including products in development from other manufacturers (namely Xeris, Zealand, and Adocia).

Pipeline Highlights

9. PRONTO! Phase 3 Trials Underway for Ultra-Rapid-Acting Insulin

Management highlighted that LY900014, Lilly’s internally-developed ultra-rapid-acting insulin candidate, has advanced into phase 3 for type 1 and type 2 diabetes. The initiation of phase 3 studies comes on schedule with 3Q17 guidance. The type 1 trial, PRONTO-T1D, has an estimated enrollment of 1,119 according to ClinicalTrials.gov, and will randomize participants to LY900014 or to Humalog (insulin lispro) in combination with basal insulin glargine (Sanofi’s Lantus) or basal insulin degludec (Novo Nordisk’s Tresiba). The 26-week study is expected to complete in September 2019. The type 2 trial, PRONTO-T2D, will enroll an estimated 670 adults, again randomizing them to LY900014 or to Humalog for 26 weeks, on top of basal insulin treatment with Lantus or Tresiba. This type 2 study is expected to complete slightly sooner, in February 2019 (and as an aside, we love the name “PRONTO” signifying speed, which is so key in improving mealtime insulin options). We’re pleased to see this forward progress, especially considering that Lilly terminated its partnership with Adocia over ultra-rapid-acting BioChaperone Lispro (now phase 3-ready) to focus more on this internal candidate (as well is its non-insulin pipeline products). Moreover, we saw promising phase 2 data in both type 1 and type 2 diabetes at ADA this past June: LY900014 was associated with faster onset/faster offset vs. Humalog, including reduced time to half-maximal drug concentration and lower total glucose excursions over a five-hour meal test. Phase 3 will reveal how this faster-acting agent affects A1c (the primary endpoint in both studies), postprandial glucose control, and hypoglycemia risk. In our view, ultra-rapid-acting products on the competitive landscape are poised to be the next major wave in insulin therapy – Novo Nordisk’s Fiasp (faster-acting insulin aspart) is now approved for the US and Europe, and ultra-rapid candidates from Sanofi and Lilly are in phase 3. We’re eager for better mealtime insulin products to be in patient hands, as we continue to believe that current options are simply not good enough. Uncertainty around meals leads to increased frequency of hypoglycemia, greater fear of hypoglycemia, and worse glycemic control overall (indeed, thought leaders have repeatedly cited that 79% of type 1 patients and 58% of type 2 patients lower their insulin doses following a severe hypoglycemia episode, impeding optimal diabetes management). Faster prandial insulins (quicker onset as well as offset) could fill this gap in our diabetes treatment arsenal, though we can’t ignore the challenges that remain up ahead, including intense pricing pressure surrounding insulins in the US.

10. Management Upbeat about Phase 2 Results for High-Dose Dulaglutide (3.0 mg/4.5 mg)

Management announced that a phase 2 trial of high-dose dulaglutide (3.0 and 4.5 mg) has been completed (Trulicity is currently available as 0.75 mg or 1.5 mg once-weekly doses). No specific data was shared, but we’re eager to see these results. According to ClinicalTrials.gov, the placebo-controlled study was completed in August 2017 (after a December 2016 start date). The primary endpoint was change in A1c from baseline after 18 weeks, while secondary endpoints explored proportion of patients in each study arm (3.0 mg dulaglutide, 4.5 mg dulaglutide, a current formulation of Trulicity, or placebo) achieving target A1c <7%, weight loss over 18 weeks, fasting plasma glucose, symptomatic hypoglycemia, and treatment discontinuation due to adverse events. Notably, Novo Nordisk’s high-dose GLP-1 agonist liraglutide (3.0 mg, branded Saxenda) has shown profound efficacy in obesity and in diabetes prevention – we wonder if Lilly will consider an obesity indication for its high-dose GLP-1 formulation (the company has previously had a laser focus on type 1 and type 2 diabetes rather than obesity, but we could imagine this shifting if weight loss results are very positive for 3.0 mg/4.5 mg dulaglutide). Saxenda has been a great success for Novo Nordisk, and the company also has a dedicated obesity clinical program for its more potent, once-weekly GLP-1 agonist candidate semaglutide (with phase 3 trials slated to start in 1H18). Lilly management expressed a decidedly positive outlook on high-dose dulaglutide, which leads us to believe that these formulations demonstrated strong safety/efficacy in the recently-completed phase 2 trial, though we can only speculate until we get our eyes on the data.

11. Once-Weekly Basal Insulin Discontinued from Phase 1

Lilly’s presentation slides noted the discontinuation of a once-weekly basal insulin candidate from phase 1. While this wasn’t explicitly discussed, management did comment on increasing competitive pressure in the basal insulin market during Q&A. As Head of Diabetes and Senior Vice President Mr. Enrique Conterno put it: “The basal insulin class was relatively more open when we launched Basaglar. Now, the situation is such that it’s harder to create viable economics for the payer.” So it seems that Lilly has de-prioritized bringing a new long-acting insulin to market in favor of focusing on its existing insulin portfolio (Basaglar, Humalog, Humulin), on the later-stage ultra-rapid-acting candidate, and perhaps on the basal insulin/dulaglutide fixed-ratio combination in phase 1. There was no hint as to the phase 1 safety/efficacy data on the now-discontinued candidate, but in all likelihood, this was a decision based mainly on market forces rather than a lacking clinical profile. For context, Novo Nordisk management has attributed the oral insulin discontinuation (from phase 2) to commercial challenges for new insulin products, claiming that the agent’s safety/efficacy profile was “generally encouraging” – we imagine Lilly may be taking a similar approach to its insulin pipeline, maintaining a very high bar for early-stage insulin agents to advance. Moreover, we see tremendous clinical advantages to basal insulin/GLP-1 fixed-ratio combinations (superior A1c reductions, more weight loss, and less hypoglycemia vs. monotherapies, plus a milder side-effect profile), and we’d love for Lilly to double down its investment in a potential basal insulin/Trulicity combo (currently in phase 1, as you’ll see in the table below).

• As background, this candidate (LY3192767) was first-discussed by Lilly management during the company’s pharmaceutical R&D update in May 2016. It was then advanced into phase 1 clinical trials in 1Q17, at which point a Lilly-sponsored study appeared on ClinicalTrials.gov comparing this long-acting basal vs. Sanofi’s Lantus (insulin glargine) in healthy participants (n=~85). This trial is still ongoing per ClinicalTrials.gov, with an expected completion date of May 2018. We suspect interim or topline data seemed underwhelming (or at least, not inspiring enough to overcome the commercial obstacles facing new insulin therapies), leading Lilly to pull back investments from this particular candidate in favor of other diabetes agents in the pipeline.

12. Enthusiasm Abounds for Diabetes Pipeline (Much More Discussion than is Typical for a Lilly Earnings Call)

We noticed substantial enthusiasm for the diabetes pipeline during prepared remarks and Q&A, with management emphasizing the promise of all three candidates in phase 2: (i) DACRA-042 (a dual amylin/calcitonin receptor agonist acquired from KeyBioscience), (ii) high-dose dulaglutide, and (iii) a GIP/GLP-1 dual agonist (phase 2 study expected to complete May 2018). The company will evaluate these phase 2 readouts in parallel, to decide which should be advanced next into phase 3 (following in the footsteps of ultra-rapid-acting insulin LY900014). Each of these would be an exciting stride for diabetes therapy in its own right. Dual amylin/calcitonin receptor agonists boast potential to treat multiple physiological defects in type 2 diabetes. Levels of the hormone calcitonin are often elevated in people with diabetes, so a drug that stimulates greater uptake at the receptor could yield important therapeutic benefits. In DACRA-042, these benefits would be combined with the robust weight loss properties of amylin analogs (both Novo Nordisk and Zealand/BI are investigating amylin analogs toward obesity indications). We’ve already conveyed our enthusiasm for high-dose GLP-1 agonists and the implications on weight management (see highlight no. 10 above). Lilly’s GIP/GLP-1 dual agonist entered phase 2 clinical trials in 2Q17. To be frank, we’ve been hearing about the GIP/GLP-1 dual agonist class for a painfully long time, so we’d love to see more movement here. Above all, we gathered from this 3Q17 update that Lilly remains very committed to developing next-gen diabetes therapies to complement its strong existing portfolio – this was music to our ears, especially in light of the recent cost-cutting measures (layoffs, global workforce reduction, and an early retirement program for US employees affected 3,500 positions this Fall).

Table 2: Lilly Diabetes Pipeline Summary

The table below reflects the latest updates, as far as we are aware, on Lilly’s diabetes pipeline products. Rows highlighted in yellow indicate notable changes to the pipeline in 3Q17.

Questions and Answers

On Trulicity

Q: What are your thoughts on the curve of Trulicity as we go into 2018, considering new brand competition from Novo Nordisk's once-weekly semaglutide? Equally important, oral semaglutide, with several phase 3 readouts coming in 2018.

Mr. Enrique Conterno (Head of Diabetes, Lilly): Trulicity is having a terrific year. We see significant benefits when it comes to the GLP-1 agonist class, and the opportunity for growth is very significant. When we look at injectables, GLP-1 agonists as a class still hold less than 30% of the prescription volume of basal insulin. Trulicity has an enviable position in this market by the benefit that it offers when it comes its real-world efficacy and once-weekly dosing. We also have a single-dose pen that is ready-to-use and does not require managing or seeing the needle, which is important for a patient transitioning from oral medication. We clearly have competition, but it is going to help fuel the overall healthy growth of the class. When it comes to semaglutide, we see that innovation in that space is important, but we need to wait for all data to come out before we can make more comments.

Q: To what degree can the Trulicity growth be attributed to switches from Victoza vs. patients coming from orals straight to Trulicity?

Mr. Conterno: When it comes to the source of Trulicity growth, it is important to highlight that our strategy from the beginning was not to focus on current GLP-1 agonist users but instead those who are new-to-brand. These strategies have been highly successful. We would like for that to continue when it comes to patients on oral medicines who could benefit from better control with Trulicity.

Q: Does or would diabetic retinopathy cloud the GLP-1 agonist class, assuming it is a part of the warning on the semaglutide label?

Mr. Conterno: The data for Trulicity is very clear: there is no retinopathy signal. We need to see how the potential warning or labeling of retinopathy for semaglutide will read out, but we see significant advantages of GLP-1 agonists in general. We're counting on significant growth of this class as we think about Trulicity.

Q: What proportion of patients initiating GLP-1 agonist therapy in the US now have some evidence of diabetic retinopathy? On average, in the US, how frequently do people with type 2 diabetes have their retinas visualized?

Mr. Conterno: When we look at the patients with diabetes, I do not have the specific number for GLP-1 agonist therapy, but generally about 30% of patients with diabetes have some degree of retinopathy. If you were to look at more serious, vision-threatening retinopathy, the number is between 3% and 5%. For the second question, unfortunately, not enough. There is a nominal number of patients with diabetes that get the screening.

On Jardiance

Q: Are you happy with the overall category growth trends for Jardiance? What do you think it is going to take to further expand usage of the SGLT-2 inhibitor class overall?

Mr. Conterno: Jardiance has quickly become the new standard when it comes to initiating patients on SGLT-2 inhibitor therapy. For new brand prescriptions, our current share is north of 50%. Overall volume and share trends are very strong. As we think about the long-term opportunity for this product, we need to focus much more on Jardiance than on the SGLT-2 inhibitor class. There are 160 million prescriptions written for oral medicines in the United States, and SGLT-2 inhibitors s have only 10 millions of those prescriptions. We are looking at a product that has an indication to reduce the risk of CV death for people with diabetes and established CV disease, which we believe is about 30% of all people with diabetes. The opportunity is enormous.

On Insulin

Q: Do you have a patent that you could potentially leverage into a delay or settlement with Sanofi over their biosimilar insulin lispro?

Mr. Mike Harrington (General Counsel, Eli Lilly): The deadline under the Hatch-Waxman Act for Lilly to challenge Sanofi’s biosimilar insulin lispro has expired, and we have not filed suit to challenge their follow-on biosimilar. The only remaining patent we have that is relevant to this discussion protects our KwikPen delivery device. We have thoroughly analyzed the Sanofi device and our own device: Lilly's KwikPen is an innovative device that offers a number of advantages for patients, and for purposes of analyzing the patents, it's a fundamentally different device than the Sanofi pen that they intend to use with their follow-on insulin lispro. Therefore we didn't have any factual or legal basis to support litigation.

Q: Can you confirm there is no settlement at present between you and Sanofi on the biosimilar insulin lispro launch? What do you think about the biosimilar lispro launch? In theory, how should or shouldn't it be different than your launch of Basaglar?

Mr. Harrington: I can confirm there is no settlement agreement in place between Lilly and Sanofi related to their follow-on Lispro.

Mr. Conterno: There are a few things to consider when we compare the situation with basal insulin vs. meal-time insulin. Firstly, the basal insulin class was significantly more open at the time of Basaglar’s entry. We have a number of different basal insulins in the formularies relative to the meal-time insulin space, where most formularies are under exclusive status. Therefore it  was easier to create viable economics for the payer in the basal insulin space relative to where we are in the meal-time insulin space. Second, in the case of Humalog, we have a full range of formulations such as Humalog mixtures. We have a concentrated Humalog U200, and we recently launched the Humalog KwikPen Junior. We believe that these offerings give us some strength when it comes to a follow-on insulin lispro coming to the market.

Q: Basaglar has a very strong uptick relative to the script. Is there anything notable in stocking or inventory change?

Mr. Conterno: We are very pleased with the performance of Basaglar and the continued adoption of this product as a key basal insulin. When we look at the quarter, there's nothing unusual on the stocking side, but there was $12 million worth of benefit related to changes in rebates and of the estimates of rebates and discounts for prior periods.

On Lilly’s Diabetes Pipeline

Q: There are eight products in your phase 2 pipeline. Can you talk about some of the products and what kind of data we should be expecting?

Mr. Jan Lundberg (President, Lilly Research Laboratories): We have two agents for diabetes. The first one is a GIP/GLP-1 dual agonist, which has two different mechanisms to lower body weight and potentially support better glucose control. We are particularly interested in this agent in some preclinical models. It had been more terrific than any other GLP-1 agonist study, including semaglutide. The DACRA is another variant in creatine-like molecules with amylin and the calcitonin stimulating receptor agent. The reason behind this is to have not only body weight lowering, but also potential insulin sensitization activity. These two agents are somewhat in parallel for readouts. We additionally have readouts for high-dose dulaglutide for obesity coming at about the same time.

Big Picture

Q: You made a cost-cutting announcement in early September about reducing 3,500 positions. What is the reason for this?

Mr. Dave Ricks (CEO, Lilly): These actions will help us become more competitive with our operating margins, and free up cash flow and operating spending capacity to invest in R&D. It is a stepping stone on a way to a more productive, higher operating margin company.

Q: As we get further into contracting season, do you have any preliminary comments about 2018 pricing or access to the portfolio?

Mr. Conterno: When it comes to 2018, and as we think about pricing and access, the different PBMs have already announced their formularies. We do see a continuation of the trends when it comes to pricing pressures in diabetes with one notable exception that was surprising – the exclusion of Jardiance from the CVS's health formulary. We are very disappointed in this decision, and we don't believe this is in the best interest of patients given the safety and CV profile of Jardiance and the fact that a competitor, canagliflozin, has a black box warning related to amputation. We will make our case with the physicians and with the patients.

Q: On the patient front, a lot of discounts and rebates aren't passed on to customers as their deductibles and co-pays go up. Is there a solution to potentially ensure that a greater amount of the discounted rebate gets back to the customers’ pockets so they can afford the new innovative medicines that you have going forward?

Mr. Ricks: For many years in the US we've seen structural problems with the PBMs having about 20% out-of-pocket co-pay by patients vs. about 4% or 5% for services like hospital services. That hasn't changed. What has changed is the number of patients who are exposed to very high deductibles, so there's a cash flow issue and overall out-of-pocket increases, particularly for branded drugs. We think an immediate step that should be taken in all sectors is to pass rebates through to patients. This is something commercial payers are now offering in the 2018 cycle. We think it is a great step to ease out of pocket versions. We advocated aggressively with CMS that this should be the policy for Medicare Part D as it’s one solution to help seniors pay for the medication. In the long-term we need to ask a broader question which we can get at through value pricing mechanism. Are medicines a better way to deliver healthcare than other parts of the healthcare system? If so, then why do we ask patients to pay more for it? It is a national debate we need to continue to engage in. We're making great strides scientifically, but the system isn't well-equipped to help patients have affordable access to these events.

Q: In the past, you've suggested that there could be a greater push to partner with technology companies to enhance your overall diabetes franchise and solutions-based approach. What can you say about that strategy at this point?

Mr. Conterno: We are excited about the opportunity brought by the convergence of pharmaceuticals and technology to be able to create differentiated solutions. For an area like diabetes, we think that it could be revolutionary. We are highly interested in it. I'm not prepared to share more than that, but can say that we're actually working on it.

-- by Ann Carracher, Abigail Dove, Payal Marathe, Elizaveta Maslak, and Kelly Close