Lilly 2Q16 – Diabetes portfolio up 27% to $1.4 billion; Trulicity drives over half of Lilly diabetes growth with sales up nearly 5-fold to over $200 million; Jardiance remains growth driver but SGLT-2 inhibitor class growth plateauing – July 26, 2016

Lilly provided its 2Q16 update this morning in a call led by CEO Mr. John Lechleiter. Sales for the company’s diabetes portfolio totaled $1.4 billion, a 27% year-over-year (YOY) increase from 2Q15 and the largest YOY margin we’ve seen for Lilly diabetes since 2006. Continuing a trend we’ve seen across several quarters, growth was driven primarily by GLP-1 agonist Trulicity (dulaglutide), which while just 14% of overall Lilly Diabetes sales, accounted for over half of the portfolio’s growth. The product continued its promising trajectory with $201 million in sales in 2Q16, a whopping 4.5-fold year-over-year (YOY) increase from a $44 million base in 2Q15, and a 40% sequential increase from $144 million in 1Q16.

Lilly’s flagship product Humalog (insulin lispro) bounced back from its low-performing 1Q16, rising 7% YOY to $702 million. It comprised 16% of the growth for Lilly’s diabetes portfolio, and 2% of the overall volume growth of the company’s entire portfolio. Humulin (human insulin) revenues rose 5% YOY to $332 million. Basaglar (biosimilar insulin glargine, marketed as Abasaglar ex-US) revenues rose 45% sequentially to $16 million in international sales, up from $11 million.

Lilly’s portion of revenue for its BI-partnered SGLT-2 inhibitor Jardiance (empagliflozin) franchise nearly quadrupled YOY and rose 5% sequentially to $40 million. We estimate overall 2Q16 revenue for the Jardiance franchise to be ~$121 million but this is an estimate as BI does not report these numbers regularly. Despite the strong growth (from a low base, given it’s a new product), management shared that overall new patient starts for the SGLT-2 inhibitor class have flattened and Jardiance’s share of new prescriptions (NBRx) has shown strong but flattening growth. Furthermore, management cited higher utilization of co-pay cards as another factor contributing to below-expected revenue growth. To counter this in part, management also mentioned the roll out of new, “less generous” co-pay cards. Looking forward, Lilly was optimistic that the FDA will grant an expanded indication for Jardiance/Synjardy following a close vote of 12-11 in favor at last month’s FDA Advisory Committee and continued to express optimism for an inflection in sales following a label update.

Lilly reported revenue of $201 million for the Trajenta franchise. Although the BI portion is not disclosed, we estimate total worldwide sales for Lilly/BI’s DPP-4 inhibitor Tradjenta (linagliptin) and DPP-4 inhibitor/metformin fixed-dose combination (linagliptin/metformin) in 2Q16 at ~$336 million. Lilly’s share of revenue for Tradjenta rose 37% YOY in constant currencies (51% as reported) to $121 million in 2Q16. Tradjenta sales accounted for 14% of the growth of Lilly’s overall diabetes portfolio in 2Q16.

On the pipeline front, Lilly announced a delay in the timeline for the advancement of its Adocia-partnered ultra-rapid-acting insulin BioChaperone Lispro into phase 3: the trial is now expected to initiate in 2017, rather than by the end of 2016 as previously expected. Although we’re very eager for “next-generation” rapid-acting insulins, this timeline shift isn’t meaningfully different at this stage. In disappointing news, Lilly has decided not to pursue a diabetic nephropathy indication for its rheumatoid arthritis drug baricitinib – we’re very disappointed about that. On the more positive side, Lilly has officially advanced its GIP/GLP-1 dual agonist into phase 1 development – this candidate was previously introduced in the company’s May R&D update.

Read on below for an item-by-item overview of highlights from the call, followed by relevant Q&A.

Table 1: 2Q16 Financial Results for Lilly’s Major Diabetes Products

Financial Highlights

• Total sales for Lilly’s diabetes portfolio reached $1.4 billion in 2Q16, rising 27% year-over-year (YOY) as reported. This marks the highest YOY growth for the company’s diabetes portfolio since at least 2006. That said, the growth occurred against a relatively easy comparison, as the portfolio declined 4% YOY as reported in 2Q15, the only quarter in which sales fell since 2012. Sequentially, revenue for all diabetes drugs rose 13% as reported from $1.3 billion in 1Q16. As usual, by geography, US performance was stronger than ex-US performance, accounting for 62% and 38% of overall sales, respectively. US sales totaled $890 million, rising 28% YOY. US sales rose 9% sequentially against an easy comparison, as US sales fell 8% sequentially in 1Q16. Ex-US sales rose 24% YOY and 20% sequentially to $554 million in 2Q16. This growth also occurred against an easy comparison, as the company’s diabetes portfolio declined 15% YOY as reported in 2Q15 and declined 8% sequentially in 1Q16 ex-US. See the tables above for an overview of 2Q16 financial results for Lilly’s major diabetes products.
• GLP-1 agonist Trulicity (dulaglutide) was the primary driver of growth for the company’s diabetes portfolio, as has been the case for several quarters. Trulicity was responsible for a whopping 52% of the portfolio’s YOY growth by value, and management noted that Lilly has benefited from the overall strength of the GLP-1 agonist class. Humalog (insulin lispro) accounted for 16% of the portfolio’s growth, DPP-4 inhibitor Tradjenta (linagliptin) for 14%, SGLT-2 inhibitor Jardiance (empagliflozin) for 10%, Basaglar/Abasaglar (biosimilar insulin glargine) for 5%, Humulin (human insulin) for 5%, and glucagon formulations for 1%. Impressively, every diabetes product for which Lilly breaks out sales experienced growth in 2Q16 (as of 1Q16, Lilly no longer breaks out revenue for TZD Actos [pioglitazone]). That said, Humalog’s positive growth is driven in part by an easy comparison to 2Q15, when Humalog sales fell 7% YOY as reported. By volume, Lilly management emphasized that US growth was driven primarily by Trulicity and Humalog, with strong performances by Jardiance, Humulin, and Tradjenta as well. Moreover, 6% of the company’s total volume growth in 2Q16 was attributed to new products including Trulicity, Jardiance, and Basaglar.

Humalog

• Revenue for Lilly’s flagship product Humalog (insulin lispro) rose 8% YOY in constant currencies and 7% as reported to $702 million. This was against an easy comparison, as Humalog sales fell 7% as reported in 2Q15. Sequentially, sales increased 16% from $606 million in 1Q16. This was also against an easy comparison as revenue fell 10% sequentially in 1Q16. Humalog performed well in both US and ex-US markets. US sales grew 5% YOY to $420 million and ex-US sales grew 11% YOY to $282 million. Lilly attributed the growth in both US and ex-US revenues to increased demand, though management acknowledged that this was slightly offset by lower realized prices domestically and unfavorable foreign exchange rates internationally. In prepared remarks, Lilly repeatedly highlighted Humalog as a strong driver of the company’s growth; the product accounted for nearly 2% of the overall volume growth of the company’s entire portfolio (including non-diabetes products), and drove 16% of the overall YOY growth for Lilly’s diabetes products. That said, Humalog’s growth numbers by value are inflated to a certain degree by the product’s weak performance in 2Q15. By volume, Lilly’s supplemental presentation slides indicated that Humalog’s total prescription (TRx) share continues to grow, climbing to over 48% following an uptick from 44-45% beginning in 4Q15. The underlying rapid-acting insulin analog market appears to have plateaued at 2%-3% YOY growth since the beginning of the year. Lilly also remarked at least once on the lower average gross margin of its insulins, including Humalog.
  
  • Specifically, in Q&A, management notably pointed out that insulin is a less profitable flagship product for Lilly compared to the small molecules that were its highest-revenue products in the past. In response to a question about changes in Lilly’s gross margin over time, management noted that the profit margins for Lilly’s current flagship insulin franchises are lower than those for its previous small molecule flagship products, such as antidepressant Cymbalta (duloxetine). We expect this is likely due to the fact that, as a biologic, the manufacturing and quality assurance process for insulin is much more complex than that for a small molecule. Dr. Lutz Heinemann pointed out the difficulties of biologic manufacturing in a recent presentation at Keystone 2016, highlighting the difference in development costs for a small molecule generic ($1-$4 million) compared to a biosimilar biologic ($40 million-$375 million).

Figure 1: Humalog Sales (2Q15-2Q16)

Humulin

• Humulin (human insulin) revenues rose 7% YOY in constant currencies and 5% as reported to $332 million. Revenue fell 7% sequentially from $356 million in 1Q16, despite the fairly easy comparison with 1Q16’s 1% sequential drop in revenue.  Humulin’s favorable YOY performance is attributable to its success in the domestic market. US revenues rose 9% YOY to $204 million whereas ex-US revenues remained flat at $128 million. The company cited increased demand as the reason for Humulin’s strong domestic performance – a departure from previous quarters, in which Humulin’s rising revenues were attributed primarily to price increases. We’re not surprised that price increases have become less of a driver of growth, given the current cost-conscious payer environment even for a “generic” insulin and rising frustration from patients and providers alike over steep insulin costs. 

Figure 2: Humulin Sales (2Q15-2Q16)

• Humulin’s TRx share in the US appears to have steadied at 54-55%, following a slight uptick from 53% in January 2016. Humulin’s growth is fairly impressive in the context of the underlying human insulin market, which has been falling 5%-6% YOY since January 2016, after hitting a low of ~7% YOY decline in November 2015. Management also shared that Humulin sales drove 0.4% of the overall global volume growth of Lilly’s diabetes and non-diabetes products this quarter. By value, Humulin accounted for 5% of the overall growth in Lilly’s diabetes portfolio. Humulin accounted for 8% of the US growth of Lilly’s diabetes portfolio.
• We’d imagine the launch of the Humulin U500 KwikPen earlier in 1Q16 played a role in the rise in US Humulin revenues and market share, though the product is still very early in its launch cycle. The Humulin U500 KwikPen, targeted toward patients with very high insulin requirements, holds 1,500 units of insulin and is more cost-effective per unit of insulin than the Humulin U100 KwikPen, which holds five times less insulin. According to the company’s past updates, demand for this U500 formulation of Humulin has been driving the franchise’s US performance since 3Q15. The increased demand for concentrated insulins from the growing population of patients with high insulin requirements likely incentivized the development of the Humulin U500 KwikPen as well as the recently approved dedicated syringe for the U500 formulation. The approval of the KwikPen and the syringe eliminate the need for complicated dose conversions for Humulin U500, improving the safety and convenience of the formulation. Given the new options that ease administration, we wouldn’t be surprised to see Humulin’s sales and market share continue to grow in coming quarters.

Trulicity

• Once-weekly GLP-1 agonist Trulicity (dulaglutide) continued its promising trajectory with $201 million in sales in 2Q16, a nearly five-fold YOY increase from $44 million in 2Q15. Trulicity sales rose an impressive 40% sequentially from $144 million in 1Q16, a particularly notable statistic against a difficult comparison: sales grew 28% sequentially in 1Q16. Trulicity boasted very strong performance in the domestic and international markets alike in 2Q16. US sales more than quadrupled YOY to $161 million and ex-US sales nearly quintupled YOY to $40 million. Trulicity sales represented 14% of the company’s total diabetes portfolio revenue and drove 52% of its YOY growth. Both US and ex-US growth contributed to this performance – Trulicity sales accounted for 63% of the growth of Lilly’s diabetes portfolio in the US and 30% of the growth outside of the US. Lilly management highlighted Trulicity as a major driver of volume growth for its overall portfolio – new products including Trulicity, Jardiance, and Basaglar accounted for over 6% of the volume growth in Lilly’s entire product portfolio, including non-diabetes products.

Figure 3: Trulicity Sales (2Q15-2Q16)

• Lilly highlighted Trulicity’s momentum within the GLP-1 agonist class, pointing out that it now holds nearly 20% of the GLP-1 agonist market by total prescriptions (TRx), up from 17% in 1Q16. Although the entire GLP-1 agonist market growth by volume has leveled off around 30% in recent months, Trulicity’s percent share continues to demonstrate a steadily increasing trajectory. Lilly has previously characterized Trulicity as a catalyst for renewed growth in the GLP-1 agonist class, which experienced a significant rebound in growth from ~8% at the beginning of 2015 to ~33% at the start of 2016. Management shared that Trulicity now captures approximately 28% of US patients beginning treatment with a GLP-1 agonist, and it has become the most-prescribed brand in Germany for patients new to the drug class. Lilly noted that this uptake is comparable to that of Novo Nordisk’s GLP-1 agonist Victoza (liraglutide) when it was first launched. That said, we expect Trulicity will face increased pressure if Novo Nordisk receives a label update for Victoza reflecting the cardioprotective benefit demonstrated in the LEADER trial. Furthermore, we expect the upcoming potential approvals and launches of Novo Nordisk’s semaglutide (in both once-weekly injectable and once-daily oral formulations), Intarcia’s ITCA 650 (implantable exenatide mini-pump), and GLP-1 agonist/basal insulin fixed-ratio combinations from Novo Nordisk (Xultophy [insulin degludec/liraglutide]) and Sanofi (LixiLan/iGlarLixi [lixisenatide/insulin glargine]) will exert additional pressures on Trulicity by providing more individualized, patient-friendly options.
• Management also highlighted the AWARD-9 results demonstrating superior A1c and weight reductions with Trulicity and insulin glargine vs. insulin glargine and placebo. The trial (n=300), presented at this year’s ADA, demonstrated that Trulicity in combination with insulin glargine significantly reduced A1c (-1.4% vs. -0.7%) and body weight (-1.91 kg [~4.2 lbs] vs. +0.5 kg [~1.1 lbs]) compared to insulin glargine and placebo. Lilly appears to be positioning Trulicity as a strong option for patients in need of basal insulin intensification. Lilly management previously suggested that the use of Trulicity with daily basal insulin injections could be an effective alternative to the GLP-1 agonist/basal insulin fixed-ratio combinations on the horizon. As Trulicity is a once-weekly injection, the use of both Trulicity and a basal insulin represents only eight weekly injections, compared to seven weekly injections for a GLP-1 agonist/basal insulin combination and 14 weekly injections for a once-daily GLP-1 agonist such as Victoza in combination with a basal insulin. Furthermore, Lilly shared in its May R&D update that it is developing a next-generation once-weekly basal insulin specifically for co-formulation with Trulicity – a once-weekly GLP-1/basal insulin combination delivered through Trulicity’s patient-friendly, IDEO-designed delivery device would be a welcome addition  to the diabetes armamentarium. 

Jardiance/Glyxambi/Synjardy

• Lilly’s portion of revenue for its BI-partnered SGLT-2 inhibitor Jardiance (empagliflozin) franchise nearly quadrupled YOY to $40 million in 2Q16, from $11 million in 2Q15. These figures include sales for DPP-4 inhibitor/SGLT-2 inhibitor fixed-dose combination (FDC) Glyxambi (empagliflozin/linagliptin) and SGLT-2 inhibitor/metformin FDC Synjardy (empagliflozin/metformin) as well as standalone Jardiance. Sequentially, sales of the franchise rose 5% as reported, from $38 million in 1Q16. The seemingly strong YOY growth for the franchise occurred against an easy comparison, with sales falling 43% between 1Q15 and 2Q15. On the other hand, Jardiance’s 2Q16 sequential growth faced a tough comparison, as sales increased by 162% between 4Q15 and 1Q16. Despite this upswing, Jardiance accounted for only 10% of growth in Lilly’s diabetes portfolio (meanwhile, GLP-1 agonist Trulicity was responsible for 52% of growth).
• We estimate overall 2Q16 revenue for the Jardiance franchise to be ~$121 million. We recently learned from BI’s diabetes update that global set sales for the Jardiance franchise totaled €165 million (~$183 million) in 2015. Lilly reported $60 million in total revenue for Jardiance in 2015, suggesting that Lilly’s share of Jardiance revenue is about 33% of the total revenue for the franchise..For comparison, 2Q16 sales of Janssen’s Invokana (canagliflozin) franchise totaled $383 million worldwide. Sales of AZ’s Farxiga (dapagliflozin) totaled $165 million in 1Q16 – we expect 2Q16 sales figures when AZ provides its quarterly update on Thursday, July 28. Jardiance’s comparatively lower sales are likely due to its status as the third-to-market SGLT-2 inhibitor in both the US and Europe, though consideration of the full combined revenue for Jardiance from Lilly and BI certainly narrows the gap between the franchise and the others in the class. During the update, Lilly management repeatedly highlighted the (very narrow) positive FDA Advisory Committee vote (12-11) in favor of a label update for Jardiance to reflect reduced incidence of cardiovascular (CV) mortality as shown in the EMPA-REG OUTCOME trial, and expressed confidence that an expanded FDA indication would boost worldwide sales. We remain fairly optimistic that the FDA will grant some sort of expanded indication for Jardiance, but our hopes were tempered a bit by the contentious discussion and split vote at the meeting.
• By geography, Lilly’s Jardiance revenue more than doubled YOY both in the US and outside the US, reaching $26 million and $14 million, respectively. For US sales, this represents a 163% YOY increase from 2Q15 and a 13% sequential decline against a tough comparison, as Jardiance sales in the US more than tripled in 1Q16. Management cited growth of the SGLT-2 inhibitor class and increased market share for Jardiance as reasons for the strong US performance. That said, Janssen’s Invokana continues to lead the class in sales as well as share of new and total prescriptions. At $14 million, Jardiance sales outside the US were nearly seven-fold the ex-US sales of $1.2 million for the franchise in 2Q15. Ex-US sales rose 66% sequentially from $9 million in 1Q16. AZ’s Farxiga continues to lead the SGLT-2 inhibitor market ex-US, likely driven by its position as the first-to-market SGLT-2 inhibitor in Europe.

Figure 4: Lilly’s Share of Jardiance Franchise Sales (1Q14-2Q16)

• In Q&A, management noted that growth of Jardiance within the SGLT-2 inhibitor class was lower than expected and attributed this to flattening new patient prescriptions for the SGLT-2 inhibitor class as a whole. Though Jardiance’s share of new therapy starts among endocrinologists exceeds 35% according to the most recent data, management commented that this number has started to plateau, fluctuating just above 30% since 4Q15. Share of new therapy starts among primary care physicians has also flattened at just above 20% since 4Q15. While growth in total prescription volume for the SGLT-2 inhibitor class remains strong at 30%, management shared that the number of new patients initiating SGLT-2 inhibitor therapy has essentially flat-lined. Management suggested that recent FDA and EMA warnings for the SGLT-2 inhibitor class may be contributing to the below-expected growth, namely the DKA warning that is affecting all products in the SGLT-2 inhibitor class. However, the company also emphasized that the strengthened bone fracture and kidney function warnings for some SGLT-2 inhibitors do not apply to Jardiance and that an updated label for Jardiance incorporating results from the EMPA-REG OUTCOME trial would likely serve as an inflection point to spark an increase in the drug’s total prescription (TRx) share and drive substantial sales growth overall. In particular, management suggested that the incorporation of a cardioprotective benefit into Jardiance’s indication could serve as a catalyst for overall SGLT-2 inhibitor class growth as well. We imagine the unexpected and unprecedented nature of the EMPA-REG OUTCOME results and the lack of consensus on the mechanism of cardioprotective benefit have hindered growth for the Jardiance franchise to some extent. 
• Disappointingly, it sounds like Lilly management is rolling out a new, “less generous” co-pay card for Jardiance (or making changes to the program), in part to increase revenue for the product. During Q&A, management suggested that greater utilization of the current, fairly generous co-pay cards has led revenue to be lower than would be expected based on the number of prescriptions for Jardiance. As we understand it,Lilly has gone out of its way up to this point to make its co-pay cards very generous, and we hope that these changes do not significantly impact access to the drug for patients who are underinsured, especially given the exciting CV mortality and kidney benefits seen in EMPA-REG OUTCOME. We’re also curious how revised co-pay cards might influence Jardiance’s share of new therapy starts. We wonder if a less generous co-pay scheme will deter patients from starting Jardiance, and we are ultimately skeptical that less generous co-pay cards will drive expansion of the company’s SGLT-2 inhibitor franchise. 
  • Jardiance has an advantage over Invokana in terms of coverage on commercial plans, but still lags behind for Medicare Part D reimbursement. Lilly management noted during the call that Jardiance is covered by 85% of commercial payers and 70% of Medicare Part D plans. During J&J’s 2Q16 update last week, the company mentioned that Invokana is covered by 70% of commercial plans and 90% of Medicare Part D plans. Notably, Invokana and J&J’s SGLT-2 inhibitor/metformin FDC Invokamet (canagliflozin/metformin) were excluded from the 2016 CVS formulary – a notable win for Lilly’s Jardiance franchise. Jardiance’s status as the third-to-market SGLT-2 inhibitor may explain its smaller presence within Medicare Part D.
• Management expressed high hopes that the FDA Advisory Committee vote (12-11) in favor of a label update for Jardiance/Synjardy will translate into an expanded indication by late 3Q16 or early 4Q16. This was clearly a key event for Lilly in 2Q16, and much of the expected growth for the company’s SGLT-2 inhibitor franchise hinges on the FDA’s decision later this year. During Q&A, management stated in a question from the very smart and very highly regarded Goldman Sachs analyst Jami Rubin that it would not be satisfied with a label that merely incorporates data from EMPA-REG OUTCOME without an expanded indication, since Lilly firmly believes that the findings from the CVOT provide substantial evidence for reduced incidence of CV death in patients with type 2 diabetes and established CV disease. We certainly feel that an expanded indication would be incredibly valuable for the type 2 diabetes community, allowing more providers and patients to consider, discuss, and prescribe Jardiance and benefit from the significantly decreased risk of CV mortality. To this end, we appreciate that management is ambitious with regard to the pending label update. We do note, however, that 12-11 in favor was a close vote rather than a substantial majority, and that the FDA could be more conservative in expanding indications without substantial majority support from the Advisory Committee voting panel. Still, we feel even the data on the label would be a major improvement. We’ll eagerly await the outcome of the FDA’s decision in 3Q16 or 4Q16.
• Lilly also highlighted data demonstrating renal benefits of Jardiance, presented at ADA and published in the NEJM. Renal outcomes data from EMPA-REG OUTCOME showed a significant 39% risk reduction for diabetic nephropathy with Jardiance vs. placebo. Given the enormous unmet need for new treatments for chronic kidney disease, we hope these results will prompt Lilly/BI to conduct dedicated trials of Jardiance in this population that could support an expanded indication.
• Lilly once again called attention to two clinical trials of Jardiance that will be conducted in patients with heart failure – we see this as a tremendously exciting area. When pressed during Q&A, management did not share details on the design of these trials, but reiterated that participants would be people with and without diabetes and clarified that one trial would be conducted in patients with reduced ejection fraction while the other would be conducted in patients with preserved ejection fraction. In addition, management implied that, through these clinical trials, the company will seek further indications for heart failure for Jardiance in patients with and without diabetes. The recent FDA Advisory Committee meeting discussion suggested that the EMPA-REG OUTCOME trial would not support a specific heart failure indication on its own and we’re glad that Lilly is investing in these trials to further characterize, and potentially better publicize, the heart failure benefit.
• As in 1Q16, the company did not break out sales of Synjardy (empagliflozin/metformin) or Glyxambi (empagliflozin/linagliptin). Neither product received distinct attention during the webcast or press release. At this time, Lilly/BI are only seeking a label update for Jardiance and Synjardy, but not Glyxambi, most likely because CV outcomes data for the DPP-4 inhibitor linagliptin component of Glyxambi is not yet available. The CAROLINA CVOT for Tradjenta (linagliptin) is expected to complete in January 2018. Management further noted that the company is still waiting on regulatory action for Glyxambi in Europe.

Tradjenta

• Lilly’s combined portion of revenue for BI-partnered DPP-4 inhibitor Tradjenta (linagliptin) and DPP-4 inhibitor/metformin FDC Jentadueto (linagliptin/metformin) rose a very substantial 37% YOY in constant currencies (51% as reported) to $121 million. Sequentially, revenue rose 28% against an easy comparison, as sales declined 7% sequentially in 1Q16. These figures only account for Lilly’s share of revenues – see below for our best estimate of total Tradjenta revenue for 2Q16. Tradjenta was responsible for 14% of growth for Lilly’s overall diabetes portfolio in 2Q16, and according to management, made meaningful contributions to US volume growth. US sales reached $49 million, marking 43% YOY growth and a 27% sequential increase. Ex-US sales reached $72 million, marking 57% YOY growth and a 29% sequential increase.
• We estimate overall 2Q16 revenue for the Tradjenta franchise to be ~$335-$340 million but this has not been announced so this is speculation. Based on BI’s recent diabetes update, total worldwide revenue for the Tradjenta franchise in 2015 totaled €909 million (~$1.0 billion), suggesting that Lilly’s share of revenue for the franchise ($357 million for full-year 2015) accounts for 36% of the total revenue. This estimated total worldwide revenue places Tradjenta as the second best-selling DPP-4 inhibitor globally by value. For comparison, Novartis’ DPP-4 inhibitor Galvus franchise, which reported 2Q16 results last week, posted total sales of $306 million. The other DPP-4 inhibitor companies have yet to report 2Q16 results, but 1Q16 sales totaled $1.4 billion for Merck’s Januvia (sitagliptin), $211 million for AZ’s Onglyza (saxagliptin), $93 million for Takeda’s Nesina (alogliptin), and $40 million for Daiichi Sankyo’s Tenelia (teneligliptin). 

Figure 5: Lilly’s Portion of Tradjenta Sales (2Q11-2Q16)

• Tradjenta, like Galvus, saw substantial YOY growth in 2Q16 despite flattening and decline in previous quarters due to tough market competition from emerging drug classes. We are curious how prescriptions and sales of DPP-4 inhibitors will continue to be affected by emerging drug classes. Recent strong commentary from Dr. Jay Skyler (University of Miami, FL) suggested that DPP-4 inhibitors should not be the preferred choice for patients due to their lower A1c efficacy compared to GLP-1 agonists, and that SGLT-2 inhibitors may also be preferable to DPP-4 inhibitors due to demonstrated renal-protective effects. On the other hand, at Keystone 2016, we heard ADA Chief Medical Officer Dr. Robert Ratner advocate for use of DPP-4 inhibitors in the elderly based on their benign safety profile and extremely easy dose administration and very clean side effects. Ultimately, we think familiarity and tolerability will mean that DPP-4 inhibitors remain a central part of diabetes care for the near future, though their long-term prospects are not quite as certain – we imagine once they go generic in years and years from now, they will be extremely popular and even more so than today. AZ, Merck, and Takeda, which also have DPP-4 inhibitors in their portfolios, will report 2Q16 results later this week. We’ll be back with a class-wide look at revenues for this drug class.
• Lilly management highlighted the phase 3 MARLINA trial results (presented at ADA) showing that Tradjenta reduces A1c in adult type 2 diabetes patients at risk for kidney impairment. The trial found that 36% of Tradjenta-treated patients achieved a target A1c <7% vs. 9% of placebo-treated patients and that the drug had a renal safety profile comparable to that seen in other trials. These findings lend additional support to the continued use of DPP-4 inhibitors in for a wide range of patients with diabetes based on their benign safety and tolerability profile.
• Lilly also achieved FDA approval for its BI-partnered once-daily Jentadueto (linagliptin/metformin) extended release combination in 2Q16. Jentadueto XR is the third-to-market DPP-4 inhibitor/extended release metformin combination, following AZ’s Kombiglyze XR (saxagliptin/metformin XR) and Merck’s Janumet XR (sitagliptin/metformin XR).

Basaglar/Abasaglar

• Basaglar (biosimilar insulin glargine, marketed as Abasaglar ex-US) revenues rose 45% sequentially to $16 million in international sales, up from $11 million in 1Q16. These sales are fairly sluggish for a product early in its launch cycle – Basaglar has been on the market for less than a year. That said, Basaglar sales accounted for 5% of the growth in Lilly’s diabetes portfolio, similar to last quarter. Accordingly, Lilly cited Basaglar as one of its principal “engines of growth.”
• Nine months following its 3Q15 launch, Basaglar’s share of the basal insulin market varies widely by geography: the product holds 25% market share in Slovakia, 13% in Japan, 6% in the Czech Republic, and 2% in Germany. This represents an increase from the 17%, 11%, and 4% market share Basaglar held in Slovakia, Japan, and the Czech Republic, respectively, in 1Q16; Germany’s market share has remained flat at 2%. Management continued to emphasize that Basaglar’s performance in individual markets is heavily dependent on the product’s copay advantage relative to Sanofi’s Lantus (insulin glargine). In prepared remarks, the company noted that Basaglar holds about 3%-5% of the basal insulin market in countries like Spain and Germany where it does not have a copay advantage over Lantus. Indeed, Dr. Lutz Heinemann, who is based in Germany, offered a fairly critical view of the safety and quality assurance of biosimilar insulins in a recent presentation at Keystone 2016. That said, we imagine the fact that Basaglar is manufactured by Lilly, which is known for its experience in the insulin field, should help reassure patients and providers on its safety. This may be less true for insulin glargine biosimilars from other companies like Biocon with less experience in this specific market, though Biocon has substantial experience with biosimilars in the developing world and a partner (Mylan) with significant experience marketing small molecule generics.
• Management confirmed that Basaglar is still slated to launch in the US on December 15, 2016, following the terms outlined in Lilly’s patent lawsuit settlement with Sanofi. We’re eagerly looking forward to this launch and are curious what Basaglar’s discounting will look like relative to Lantus in the US. 

Other Products

• Global sales of Lilly’s glucagon formulation of $31 million rose 17% YOY as reported and in constant currencies and 1% sequentially. As has been the case historically, glucagon revenue was driven primarily by US sales, which accounted for $29 million, up 18% YOY and 2% sequentially from $30 million in 1Q16. Ex-US sales, which have remained relatively low and flat since 1Q12, accounted for the remaining $2 million. Glucagon was responsible for 1% of growth of Lilly’s overall diabetes portfolio.

Figure 6: Glucagon Sales (1Q12-2Q16)

• As in 1Q16, Lilly did not break out its share of revenue from Actos (pioglitazone). Actos sales have been declining steadily for at least five years, due to the loss of patent exclusivity and concerns over heart failure, bone fractures, and bladder cancer.

Pipeline Highlights

Ultra-Rapid BioChaperone Lispro

• Lilly announced that it now plans to initiate a phase 3 trial for its Adocia-partnered ultra-rapid-acting insulin BioChaperone Lispro in 2017, a slight delay from the previous late 2016 expectation. We’ve been eagerly looking forward to a “next-generation” rapid-acting insulin for some time and believe that the change shouldn’t signal major stress. Innovation in the rapid-acting insulin analog field has not progressed quite as rapidly as in the basal insulin field (which recently saw the launch of Novo Nordisk’s Tresiba [insulin degludec] and Sanofi’s Toujeo [U300 insulin glargine], both of which are doing extremely well based on our “real world” contacts) though several interesting new rapid-acting insulins are in the mid- to late-stage pipeline. Novo Nordisk’s faster-acting insulin aspart is closest to market – the company submitted the product to regulatory authorities last December and expects decisions in 4Q16. Full phase 3 data from the Onset 1, Onset 2, and Onset 3 trials for faster-acting aspart were presented at ADA 2016. Both faster-acting aspart and BioChaperone Lispro appear to offer meaningful but incremental benefits over existing rapid-acting insulins and aren’t expected to be paradigm shifting. Furthermore, we expect that the rapid-acting insulin market will continue to face significant pressure with the growing popularity of GLP-1 agonists and GLP-1 agonist/basal insulin fixed-ratio combinations for postprandial glucose management. That said, we’re particularly intrigued by Dr. Tim Heise’s (Profil GmbH, Neuss, Germany) suggestion at ADA 2016 that BioChaperone Lispro may support more flexible dose timing in addition to improved post-prandial control, which he views as an especially attractive feature for elderly patients or children. Indeed, we expect many patients would appreciate greater flexibility in mealtime insulin dosing, as it is more consistent with how patients dose mealtime insulin in the “real world.” On the basal insulin front, the dosing flexibility offered by Tresiba has garnered significant attention and praise from clinicians.
• BioChaperone Lispro recently reported positive phase 1b meal study results at ADA 2016, demonstrating greater improvements in postprandial glucose (PPG) with Lilly/Adocia’s ultra-rapid BioChaperone Lispro compared to Humalog. Compared to Humalog, BioChaperone Lispro demonstrated improvements in mean one-hour PPG (42 mg/dl difference), two-hour PPG (27 mg/dl difference), and significant reductions in PPG excursions as measured by area under the curve (51% reduction at 30 min post-meal [p=0.0004], 58% reduction at one hour [p<0.0001], 61% reduction at two hours [p<0.0001], 75% reduction at three hours [p=0.0002], and 42% reduction at eight hours [p=0.04]). One- and two-hour post-meal blood glucose concentration and peak post-meal blood glucose reductions all significantly favored BioChaperone Lispro over Humalog as well.
• Results from an ongoing phase 1b trial in patients using BioChaperone Lispro in an insulin pump are expected in 4Q16. The trial (n=80) includes two parts. Part A involves a 14-day crossover trial (n=36) evaluating postprandial control and a variety of PK/PD parameters vs. Humalog in pumps in patients with type 1 diabetes. Part B (n=44) will evaluate PK/PD, pump compatibility, and safety of BioChaperone Combo vs. Humalog administered via two different pumps (Roche’s Accu-Chek Spirit and Medtronic Paradigm Veo) and administered via syringe in a four-period crossover study. The trial is expected to complete in October 2016. For more on BioChaperone Lispro, see our report on Adocia’s 2Q16 update.

GLP-1/Glucagon and GIP/GLP-1 Dual Agonists

• Lilly management highlighted the advancement of a once-weekly GIP/GLP-1 dual agonist into phase 1 development. We assume this was the previously undisclosed phase 1 biologic for diabetes from Lilly’s past quarterly updates. The GIP/GLP-1 dual agonist was first publicly disclosed at Lilly’s R&D update meeting in May. At the same meeting, management also shared that the company is developing a preclinical next-generation once-weekly GLP-1/glucagon dual agonist (an oxyntomodulin analog). We’re glad to see continued investment on Lilly’s part in early-stage diabetes candidates, though the decision to invest further in incretin dual agonists came as a slight surprise to us, as Lilly recently declined to advance its phase 2 Transition Therapeutics-partnered GLP-1/glucagon dual agonist LY2944876/TT401 into phase 3 trials. Lilly management previously emphasized that TT401 did not meet the company’s “very high” internal bar for efficacy and potency for advancement into phase 3, but noted that the combination of weight loss and A1c reduction in the phase 2 trial does provide strong support for the dual agonist hypothesis as a whole – the company presumably has reason to believe that this candidate will be more potent. Lilly even suggested that the company could eventually co-formulate its preclinical once-weekly long-acting glucagon with GLP-1 agonist Trulicity or create a tri-agonist with the long-acting glucagon and the phase 1 GIP/GLP-1 dual agonist (management noted that it’s unlikely Lilly will develop the long-acting glucagon as a standalone product). The GLP-1/glucagon dual agonist field has received significant industry investment of late – see our competitive landscape for more.

Other Pipeline Highlights

• Disappointingly, Lilly discontinued development of its rheumatoid arthritis drug baricitinib (JAK1/JAK2 inhibitor) for a diabetic nephropathy indication. Lilly also dropped a phase 1 anti-VEGF monoclonal antibody for this indication from its pipeline in 1Q16. Lilly’s decision to move away from this field is particularly disappointing given the high unmet need. That said, the clinical development process for diabetic nephropathy is often extremely long and costly, given the lack of appropriate surrogate endpoints for this indication. We would love to see regulatory authorities continue to work with industry to develop appropriate endpoints for this field – in the meantime, we’re glad to see that the diabetic nephropathy competitive landscape remains fairly crowded and we’re particularly optimistic about the potential of SGLT-2 inhibitors, as well as Bayer’s phase 3 mineralocorticoid receptor antagonist finerenone and AbbVie’s phase 3 endothelin-receptor antagonist atrasentan, for this indication.
• Lilly publicly entered the NASH competitive landscape with a discussion of its NASH strategy at its May R&D update. Management confirmed that Lilly is investigating its DGAT-2 inhibitor for NASH; the candidate is also in phase 1 for dyslipidemia. Ionis Pharmaceuticals added a DGAT-2 inhibitor for NASH to its own phase 1 pipeline in 3Q15. Interestingly, the company chose to list the new candidate within its diabetes pipeline, rather than its cardiovascular pipeline. Ionis has previously suggested that it may investigate its DGAT-2 inhibitor for type 2 diabetes in addition to NASH and we continue to hope that Lilly is considering this option as well. Lilly also shared during its R&D update that it intends to study the once-weekly GLP-1/glucagon dual agonist for NASH as well, citing promising preclinical data in diet-induced obese mice.
• There were no additional updates on Lilly’s other diabetes pipeline candidates. Lilly announced several phase 1 and preclinical candidates at its R&D update day in May that have yet to be added to its pipeline webpage. These include a phase 1 soluble glucagon, a phase 1 ghrelin O-acyltransferase (GOAT) inhibitor, a preclinical next-generation basal insulin, and a preclinical once-weekly insulin. Lilly’s pipeline webpage lists an undisclosed biologic for hypoglycemia and an undisclosed chemical entity for diabetes (licensed from a third party) in its phase 1 pipeline, which we assume could refer to the soluble glucagon and the GOAT inhibitor, respectively. In its R&D update, management particularly highlighted the potential use of the phase 1 soluble glucagon in bi-hormonal closed loop systems. The company also shared that the next-generation basal insulin would enter phase 1 trials by the end of 2016 and was specifically designed to eventually support once-weekly combination use with Trulicity. The company’s intranasal glucagon acquired from Locemia remains in phase 3 and PCKS9 inhibitor LY3015014 remains in phase 2. 

Questions and Answers

Jardiance

Q: Could you comment on the disparity between cardiologists and endocrinologists voting in the Jardiance AdComm and what that could potentially mean for the label change and treatment guidelines?

A: In the Jardiance AdComm, we had a positive 12 to 11 vote that we had substantial evidence of reduction in the incidence of cardiovascular death in patients with type 2 diabetes and established cardiovascular disease with Jardiance treatment. If I recall, the five practicing cardiologists voted in favor. Excluding the pediatric endocrinologists, there were five endocrinologists on the panel. Of those, two voted in favor and three voted against. I think it's difficult to try to assign a particular view to either cardiologists or endocrinologists based on the vote of 10 people. What I would say is we were very pleased with the overall discussion. We thought it was robust and we believe that the right discussion happened. For that reason, we feel optimistic that we have a really good chance of getting an indication come late Q3, early Q4.

Q: Naturally Jardiance will be the only one with cardiovascular outcome data in the label in 2016 and 2017. Do you think that will lead to significant formulary coverage shifts in favor of your product in 2017? Or could that be viewed as a class effect by payers, and you could find incumbent products like Invokana actually keeping a good portion of their formulary positioning?

A: There is no question that a new label with an indication for a reduction of incidence of cardiovascular death will be very significant from a payer perspective. Now, in this particular case, we already have excellent coverage and access. We are at 85%-plus when it comes to commercial access and above 70% in Part D. So our access is very good and it can only get stronger once we get the label.

Q: Can you give some commentary on the design of your heart failure trials and how you think those trials might be able to capture data that could help with the additional build-out of your label on Jardiance?

A: We are not in a position right now to disclose the specific design for our heart failure trials. We will be conducting two trials, one for reduced ejection fraction and one for preserved ejection fraction. We expect the first of those trials to start this year. Also, note that we are studying heart failure in people with and without diabetes. So clearly, we will be seeking an indication when it comes to heart failure with those trials.

Q: You mentioned that the year-over-year growth for the SGLT-2 inhibitor class is a little less than what you had expected. Can you comment on why you think that is? Is it some of these safety concerns we're seeing across the class or anything else that's keeping growth a little bit below what you were expecting?

A: Over the last year or so, we’d had a number of strengthened warnings or new warnings when it comes to products in the SGLT-2 inhibitor class, starting with the DKA warning, which was added across the class. More recently, we’ve had two strengthened warnings that have covered other products but have not covered Jardiance. One warning for bone fractures was specific to canagliflozin and a strengthened warning on acute kidney injury covered both dapagliflozin and canagliflozin.

At the same time, we're currently unable to promote our CV data until we get our label updated. That said, overall we continue to feel that the benefit-risk profile of the class is very strong, but that is especially true for Jardiance in particular, given that our data basically speaks for itself when it comes to some of these safety assessments and when it comes to the CV benefit.

When we look at the class in general, the class is growing over 30% year-to-date, which one can say seems like very good growth. But when we look a little more closely at the new patient starts, we basically see that completely flattening. For that reason, we are a bit concerned and we are hopeful and optimistic that once we get both the new indication and new treatment guidelines, that we will see a significant inflection in sales. Jardiance needs to be the catalyst for the overall growth of the class.

Q: Could you outline your marketing plans for Jardiance assuming you get approval with the cardiovascular indication added? Are you building out your cardiovascular sales force or bolstering your primary care sales force for the product?

A: Clearly we view this as a very significant opportunity, so we're going to be fully resourced. That basically means making all of the proper investments when it comes to ensuring that we have the right reach for primary care, cardiologists, and endocrinologists. We have a very robust plan together with our partner Boehringer-Ingelheim.

Q: Would you be satisfied if the CV data were included in the label but without the CV indication per se? Can you describe the guidelines process and the importance of updated guidelines and specifically what you need to see in order to see that obviously important inflection point with Jardiance?

A: We do feel optimistic on the CV indication. You asked me if I would be satisfied with just inclusion of the data on the label. I will be honest – no, I would not be satisfied. We will make it work, but my view is that we basically have the appropriate data to be able to obtain a new indication when it comes to the reduction in the incidence of CV deaths in that specific population. Guidelines are also very important, but, in the US in particular, we view the FDA action date coming before new treatment guidelines. We could have new treatment guidelines in the US sometime early next year.

Q: With Jardiance, the second quarter sales looked a little light versus what was implied from the scripts. If you could give any more color around the impact of discounting, rebating and inventory buildup, that would be helpful.

A: If you look at the scripts and sequential script growth, it would imply higher revenues. There are two things that are weighing down on Jardiance. One is that we've had some gross-to-net adjustments during the period that were from prior periods. The second is that we also saw much higher utilization of our copay cards. We have recently changed the design of our copay card so that it's less generous than it used to be. But that takes a little bit of time to wash out, basically. So we should expect an improving picture when it comes to that. 

Trulicity

Q: On the Trulicity REWIND trial, can you remind us if it is just a futility look or if there is potential for early stoppage? If it’s the latter, how should we be thinking about this given the outcome of Novo's LEADER study and the fact that I believe you're approaching a similar duration of treatment?

A: We do have an interim analysis later this year. If we were to observe a similar hazard ratio versus what LEADER showed, we would not stop the trial, given that we're going to have a significantly less number of events than Novo had at the conclusion of their trial. We are confident in the profile of the product and now we just really need to wait for the interim. If we stop the trial we will know the results; if we continue with the trial, clearly we will have to wait for the results.

Big Picture

Q: How would you handicap the likelihood of Part D rebates between now and 2020?

A: The extreme would be the repeal of the noninterference clause. And then there's a lot of scenarios in between, for example, moving the LSI patients or the duals to more of a Medicaid-type system. I think the likelihood is low, quite honestly. I think that Part D is a rare example of a government program that comes in exceeding expectations at a cost of hundreds of billions of dollars less than had been forecast at the beginning. The Congressional Budget Office has several times stated that if price controls are placed into Part D, the only way the government would save money is to restrict access to drugs on the formulary. Today most seniors have access to the whole formulary through the private plans, and that's why senior satisfaction rates are so high. That doesn't sound like something a politician would want to mess with.

Q: Can you elaborate on some of your comments about increased pricing pressures in the US through 2020 that's reflected in the 5% minimum target? How are you thinking about pricing dynamics for the next few years? Do you see the industry with less pricing power? Do you see more volume driving growth? For 2017, are there any major shifts in coverage that we should be thinking about as we look toward next year? Specifically on diabetes, is there anything we should be keeping in mind in terms of either coverage or pricing as you've been talking to payers about the 2017 season?

A: I'd just say as a preamble, typically we comment on access much later in the year, closer to the new plan year and oftentimes need to wait to say anything until the payers themselves have announced these.

When we talk about increased pricing pressure, if you look at the trends, we're seeing a decline or a decay in the net price benefit that we're receiving on kind of an annual basis. When we look to the future, we are also seeing an environment where we're seeing increased rebating or discounting going on. We've experienced that in our diabetes business, and we know that we're looking to go into some highly competitive spaces in terms of some of our new product launches. So our focus has been on driving volume-driven revenue growth, and that really gets to the clinical differentiated profile of our new products that we are in the midst of launching and that's where we've centered our attention.

Q: On the gross margin comments, can you give us a sense of the push and pulls over the course of the decade where some of the products that have higher gross margin that'll be bringing it up and what'll be bringing it down?

A: Clearly, we endured the impact of the loss of many of our biggest products and at the same time they were also small molecule. So at the gross margin line they were very highly profitable. These are products like Cymbalta, Zyprexa and Evista. Then as those products expired, those revenues were replaced with insulins becoming our biggest brands in franchise in Lilly at the moment, which obviously is coming with a lower gross margin or lower profitability. And that's where you've seen the decline in our gross margin rate over this YZ period.

Going forward, it's really going be dependent on the nature of the molecules and which ones are launching at what time. So if you look at products like Jardiance, we booked our portion of the income so it actually has a boost to our gross margin as a percent relative to our base. Likewise, when you look at the opposite, you've got baricitinib, which we own, but we have a pretty high royalty rate that we also pay in the high 20s. So the mix effect is really what's going to drive or have a significant impact on what that gross margin profile looks like over time. However in aggregate, when we look at all the pushes and pulls, we still feel very confident that in total our gross margin will improve between now and the end of the decade.

-- by Abigail Dove, Helen Gao, Payal Marathe, Emily Regier, and Kelly Close