Lilly 2Q18 – Diabetes portfolio grows 29% YOY to $2.4B driven by Trulicity (+62% YOY), Basaglar (+133% YOY); Relatively weaker quarter for Jardiance (+43% YOY, -3% QOQ); GIP/GLP-1 dual agonist moves to phase 3, Nasal glucagon goes to FDA/EMA– July 25, 2018

Lilly reported 2Q18 financial results this morning in a call led by CEO Mr. Dave Ricks. Click here for the press release, here for the presentation slides, and here if you’d like to listen to the call. Below, you’ll find our eight highlights covering financial results and pipeline updates, plus historical graphs of sales trends and tables of product revenue and Lilly’s diabetes pipeline.

2Q18 Financial Results for Lilly’s Major Diabetes Products

Financial Highlights

1. Lilly’s Overall Diabetes Portfolio Climbs 29% YOY to $2.4 Billion, Driven By 31% Gain in US Volume; Trulicity Drives 54% of Growth and Overtakes Humalog as Top Revenue Generator

Lilly’s diabetes portfolio grew 29% YOY to $2.4 billion in 2Q18, from a base of $1.9 billion in 2Q17 when the portfolio rose 30% YOY. Diabetes revenue also rose 6% sequentially in 2Q18, from $2.3 billion in 1Q18 when it posted a lower 2% sequential gain. Trulicity drove 54% of YOY portfolio growth, Basaglar 21%, Humalog 17%, and Jardiance 8%. At $780 million for the quarter, Trulicity sold $10 million more than Humalog in 2Q18, overtaking the mealtime insulin analog for the first time ever (both now comprise 32% of Lilly Diabetes). Cited as the primary driver of the company’s overall growth (+9% YOY in both volume and revenue), Lilly’s diabetes business continued on the growth trajectory it has followed since 3Q15; indeed, since 4Q16, the portfolio has posted >20% YOY growth every quarter and climbed 32% in 2017 overall. The company’s diabetes sales were, as usual, driven primarily by US performance: US sales climbed an impressive 35% YOY to $1.6 billion (+5% sequentially), while OUS diabetes revenue bounced back from a 2% sequential loss in 1Q18, increasing 16% YOY and 7% sequentially to $773 million. Over the years, US sales have made up a larger and larger share of Lilly’s global diabetes sales, now comprising 68%.  Of particular note, management highlighted very early in the call that US diabetes volume is up 31% YOY, driving most revenue growth stateside; relative to the 35% YOY growth in US revenue, this suggests that Lilly is also seeing a higher price on at least some products. Ultimately, Lilly’s diabetes business continues to grow fast from a high base (as a reminder, the company recently surpassed Merck to because the second-largest diabetes company). For rough comparison, in 1Q18 Novo Nordisk’s diabetes/obesity portfolio grew 6% YOY to $3.7 billion, AZ’s diabetes portfolio also grew 6% YOY to $607 million (driven solely by SGLT-2 Farxiga), and Sanofi’s diabetes portfolio fell 20% YOY ($1.6 billion) due to further Lantus decline.

• We heard surprisingly little talk about insulin pricing pressure, which is likely at least partly responsible for the sequential drop (-3%) taken by Humalog in 2Q18 (and, more generally, the stagnation of the rapid-acting insulin market). The pricing pressure that results from the PBM/payer/manufacturer price- and rebate-setting axis has been repeatedly criticized by key thought leaders and has become a near-fixture on the earnings calls of insulin manufacturers. Particularly in the rapid-acting class, where payers consider products easily interchangeable, different products are pitted against one another when PBMs/payers design exclusive formularies, resulting in a high rebate paid out by the manufacturer, driving down the realized price. We suspect Lilly is paying increasingly high rebates on Humalog to secure commercial coverage, though this is speculation on our part (we would love to see how rebates have changed over the years relative to list prices, which would require some transparency around PBMs). That said, management acknowledged on Lilly’s 1Q18 call that segment mix for Humalog has been more favorable in 2018, meaning there are fewer Medicare claims and more high-paying, commercially-insured patients. Indeed, Humalog is the only rapid-acting insulin that experienced sales growth in 1Q18 (+12% YOY); Novo Nordisk’s NovoLog fell 12% YOY and Sanofi’s Apidra fell 7% YOY. In 2Q18, Humalog revenue climbed 13% YOY, and we do sincerely hope that this positive performance isn’t coming at the expense of patient access. Starting August 1, Lilly will launch a new helpline to give patients personal advice on how to secure the maximum discount on Humalog, Humulin, and Basaglar – we’ll be curious to see how many patients this helps (of the 400,000 targeted), and if it affects mean realized price on Humalog at all.

• Lilly once again updated its financial guidance for 2018, explicitly on the strength of its diabetes portfolio as well as favorable foreign exchange rates. Total revenue is now projected at $24.0-$24.5 billion, up $300 million from the $23.7-$24.2 billion projection made in 1Q18 (which also represented an increase from a December announcement of $23-$23.5 billion). Management noted that favorable foreign exchange rates accounted for 2% of their overall portfolio growth, attributing the other 7% completely to volume; the pharma business alone saw an even greater +9% change in volume.

• There was substantial interest from investors on coming changes in US drug pricing and rebate policy, but Lilly management offered few details; we noticed much less specific discussion of pricing and access for particular products than usual. In two interesting comments, CEO Mr. Dave Ricks asserted that decreases in list price (theoretically enabled by lower rebates) would increase volume and improve costs and adherence, but also explained that there’s “a lot of room for disruption in a negative way here.” He also reinforced the company’s commitment to innovation, alluded that Lilly would support patients being able to access a post-rebate level list price, and directly opposed the importation of drugs from outside the US, instead calling for a direct fix to broken US regulatory processes. We would have loved to hear a more robust commitment to patients and affordability from Lilly, though we are impressed by some of the company’s efforts on this front. For example, the new insulin helpline has the potential to help 400,000 US patients access Lilly insulins at the lowest possible price for them, and Lilly has pledged to offer personalized support and solutions. Of note, Lilly saw an overall 24% net margin in 2Q18 at a company level – we’d love to know what diabetes was – and management noted that margins continue to expand. Innovation is absolutely needed, and currently the US is subsidizing R&D globally; unfortunately, no other countries are subsidizing US affordability. We’d like to see more resources here, not only from industry but from other large players. Moreover, it’s high time for better policies from insurance companies and a bright light put squarely on PBMs so that their value is clarified.

2. Trulicity Continues to Impress with 62% YOY Growth to $780 Million; Leads Overall Diabetes Portfolio with 54% Share of Growth; REWIND on Track for Topline Readout by Year-End

Sales of GLP-1 agonist Trulicity (dulaglutide) rose 62% YOY to $780 million in 2Q18, from a base of $480 million in 2Q17, also growing an impressive 15% sequentially (~$100 million) from a high base of $678 million in 1Q18. This represents a very encouraging rebound in Trulicity’s growth trajectory, after sequential growth slowed to +5% in 1Q18, its lowest rate since sales were first reported in 4Q14; we continue to be impressed by Trulicity’s steep growth from a high base, particularly as we understand the product is driving underlying class growth more than anything else. Within Lilly’s overall diabetes portfolio, Trulicity accounted for a 54% share of growth in 2Q18, while reflecting 32% of total revenue. Indeed, the GLP-1 has edged out mealtime insulin Humalog for the first time ever, by a margin of $10 million; it was more than clear to us through the call that Trulicity is a bright spot within Lilly’s entire pharmaceutical portfolio. By geography, Trulicity remains remarkably strong in the US, where sales grew 61% YOY and 16% sequentially to $612 million. OUS growth was similarly encouraging, with revenue climbing 69% YOY and 12% sequentially to $167 million. Lilly’s press release attributed US growth to (i) higher demand related to increased market share for Trulicity and to (ii) underlying GLP-1 agonist class growth, while OUS growth was attributed to increased volume and, in part, favorable exchange rates that more than compensated for lower realized prices OUS.

• Although according to Lilly’s slide deck (slide 16), Trulicity is now the market leader by volume in the US, with 37% TRx (share of total prescriptions), we aren’t sure that reflects current information. Lilly’s supplementary slides (slide 27) indicate that total US GLP-1 prescriptions are currently growing at an annual rate of ~26%, and Trulicity captured ~37% of those prescriptions as of June 2018. However, this 37% is identical to the US volume share we noted for Trulicity in 1Q18, and our initial suspicion is that Novo Nordisk’s new GLP-1 agonist Ozempic has shifted the rest of the GLP-1 class breakdown – we’ll know more when Novo Nordisk reports on August 8. That said, we’ve heard Ozempic has been hard for patients to get their hands on.  

Trulicity Sales (4Q14-2Q18)

• Anticipation continues to build around the REWIND CVOT, and Head of Diabetes and SVP Mr. Enrique Conterno confirmed that a topline release is expected in 4Q18. Expected completion according to ClinicalTrials.gov is July 2018, though the CVOT is still listed as active. As in the past, Mr. Conterno reinforced his confidence in the trial’s design, explaining that REWIND is differentiated by its longer duration of follow-up, which is expected to give >5 years of time on-treatment (get a rundown of all reported and ongoing CVOTs here). To be clear, Lilly designed REWIND with a longer follow-up to balance the trial’s very large (69%) primary prevention population – while some have speculated that REWIND may not show CV superiority due to the lower-risk population, this is also one of the reason’s we’re most excited about the trial. To-date, all CVOTs to demonstrate cardioprotection have enrolled few to no participants without baseline CV disease, so REWIND could vastly expand the population considered to reap CV benefit from GLP-1 agonists – and also, potentially, give Trulicity a much broader CV indication.

  • We can’t overstate the importance of REWIND to Trulicity’s competitive position in the GLP-1 agonist class. When asked whether Trulicity could continue to grow without CV superiority, Mr. Conterno touted the product’s “unmatched patient experience,” explaining how once-weekly autoinjection improves adherence and real-world efficacy. He also cited an “excellent access position.” While we’re very much on board with the patient friendliness of Trulicity, and the access could not be more important, we do think that long-term commercial success is greatly dependent on its CVOT assuming that payers value cardioprotection as much as we believe they will. We do think patients over time will be asking for it though for now, they are just beginning to understand it.

  • In our view, REWIND results that do not show CV protection would represent a major blow to Trulicity’s overall package, and its ultimate growth trajectory (although Lilly could obviously propel further growth by making it available to payers for a better deal, cardioprotection will also enable it to have high profitability). Novo Nordisk’s market-leading GLP-1 agonist Victoza (liraglutide) already boasts a CV indication based on very clearly positive LEADER data, and the company’s next-gen, once-weekly Ozempic (semaglutide) has also demonstrated CV benefit in SUSTAIN 6. What’s more is that Ozempic showed significant superiority over Trulicity in the head-to-head SUSTAIN 7 trial (at both low and high doses, dulaglutide was less efficacious in lowering A1c and stimulating weight loss), and Novo Nordisk is heavily promoting these comparative results. As patients/providers start to demand more than just glucose-lowering from their diabetes drugs, our sense is that any new product will have to show CV efficacy, let alone an agent that’s already facing tough in-class competition on other metrics. To be clear, we think Trulicity is an outstanding therapy that could help so many patients with type 2 diabetes, and we see ample room for the entire GLP-1 class to continue to climb. As Dr. Ralph DeFronzo stressed at AACE 2018, we should be careful about over-emphasizing head-to-heads among different GLP-1 agonists, because the reality is that being on any one of these agents would be better than being on none of them, and due to serious access challenges most patients are on none of them.

• Oral semaglutide, the likely first-to-market oral GLP-1 agonist, looms large: Novo Nordisk’s phase 3 PIONEER program has released topline results for about half of its 10 studies, and PIONEER 10 will compare oral semaglutide head-to-head vs. Trulicity. At Keystone 2018, Dr. Michael Bush noted how the ease of use of oral medications could shift patient preferences away from Trulicity ­– though whether a once-weekly injection is easier or harder than a daily pill (with fasting requirements) will depend on the patient. We see expanded choice within the GLP-1 class as a fantastic opportunity for more personalized diabetes treatment, where patients can express preferences for oral vs. injectable or daily vs. weekly administration, because all those options exist. Novo Nordisk intends to position oral semaglutide slightly earlier in the course of type 2 diabetes development, so that it competes with SGLT-2s and DPP-4s more so than with injectable GLP-1s (after all, Novo Nordisk has its own injectable GLP-1 agonists to keep selling). Given all this, we expect that oral semaglutide will expand the GLP-1 market instead of “stealing share,” but we can’t deny that pressure is mounting for Trulicity. As we embark on an empowering new era of diabetes care defined by outcomes beyond A1c (CV and renal outcomes, hypoglycemia, weight loss), patients can and should demand more from their therapies; the coming months will tell us more about what Trulicity has to offer to this end.

• While management did not mention Trulicity’s recent label update indicating safety/efficacy in patients with moderate to severe CKD (based on AWARD-7), the change was highlighted in the press release and on slide 7. In AWARD-7, Trulicity offered similar A1c reductions to basal insulin glargine (Sanofi’s Lantus) when both were given alongside mealtime insulin lispro (Lilly’s Humalog), but Trulicity gave significant benefit on weight loss and hypoglycemia risk. What’s more, Trulicity actually appeared to decelerate eGFR decline in this patient population with CKD (REWIND will give more robust data to this point). From a baseline 38 ml/min/1.73m², eGFR dropped a mean 1.9 ml/min/1.73m² after 26 weeks of insulin glargine therapy, but only fell 0.1-0.4 ml/min/1.73m² with dulaglutide over the same time frame (p<0.05 for both comparisons). Of late, we’ve noted renewed interest in the renal protective effects that seem to come with GLP-1s – now we’re just waiting for a dedicated outcomes trial, as SGLT-2 Invokana just proved itself in the CREDENCE renal outcomes trial. 

3. Jardiance Posts $147 million in “Flat-ish” Quarter (+43% YOY, -3% Sequentially); Management Vaguely Cites Changes in Rebate Estimates and Discount Normalization; Robust Clinical Trial Program Holds Promise for Growth

Lilly’s share of SGLT-2 inhibitor Jardiance (empagliflozin) sales increased 43% YOY to $147 million, rising from $103 million in 2Q17 and representing significant slowing vs. +104% YOY growth in 1Q18 and +122% growth in 2017 overall. Jardiance is part of Lilly’s diabetes alliance with BI (a private company that does not report revenue publically), and we estimate entire franchise revenue at $445 million, growing from $312 million in 2Q17. These estimates assume Lilly collects ~33% of total Jardiance sales based on reported figures from 2015: For that year, Lilly recorded $60 million in empagliflozin sales while BI listed global net sales at ~$183 million. Sequentially, Jardiance revenue actually fell 3% from $151 million in 1Q18. Moreover, the 43% YOY increase for 2Q18 is by far the lowest we’ve seen for the franchise since we could perform this calculation – the next lowest is 88% YOY growth in 4Q17. By our calculations, Jardiance drove only 8% of growth in Lilly’s diabetes portfolio in 2Q18, comprising 6% of total diabetes revenue. Indeed, over the last year, Jardiance has driven a relatively lower share of growth for Lilly’s diabetes portfolio, falling from 14% in 2Q17 and 3Q17, to 13% in 4Q17, 12% in 1Q18, and now only 8% (though some of this is due to strong performances from Lilly’s other diabetes products). When asked about this “flat-ish” quarter for Jardiance during Q&A, Mr. Conterno suggested that the decrease was due to a “double whammy” of changes in rebate estimates and discount normalization over 1Q18 and 2Q18. He explained that US sales of Jardiance rose 28% YOY as reported while TRx climbed 58% YOY in 2Q18 (i.e. volume growing at a faster pace than revenue); lower realized price (due to pricing pressure, higher Medicaid utilization, patient discounts, etc.) is cited in the press release as contributing to this disconnect. Ultimately, we found this explanation a bit vague, although J&J has also cited pricing pressure as a headwind for the Invokana franchise, meaning SGLT-2 inhibitors are not immune to this major commercial challenge in the US. We’re hoping for more color on what’s going on in the SGLT-2 market as AZ reports (and Merck, although Steglatro is so new and poorly reimbursed right now that it’s unlikely to make a dent). We also found Lilly’s supplementary slide for Jardiance (slide 31) incredibly illustrative: Since June 2016, YOY total prescription market growth has plummeted from ~30% to ~6% in June 2018 (in the US); though Jardiance leads the market with ~40% market share by volume (slide 16), underlying class growth is quite limited.

• The waning in Jardiance’s growth was entirely due to sales in the US, where 28% YOY growth and a 10% sequential drop gave $86 million in revenue as reported by Lilly. Performance OUS remained strong (+70% YOY and +10% QOQ), with revenue totaling $62 million for Lilly.

Lilly’s Reported Jardiance Sales (3Q14-2Q18)

• With phase 3 trials in type 1 diabetes, heart failure, and chronic kidney disease completed, underway, or soon-to-be initiated, we see a high ceiling for Jardiance and overall SGLT-2 inhibitor class growth. Topline results from the EASE program in type 1 diabetes were just presented at ADA and were highlighted in Lilly’s slide deck; 2.5 mg, 10 mg, and 25 mg doses of empagliflozin all gave significant A1c and weight loss benefits, and the lowest dose did so without an increase in DKA (with efficacy that apparently matched the 10 mg dose). We’ll have to wait for EASD 2018 in October to see the full results, and to answer the question on everyone’s mind – is this too good to be true? Additionally, the EMPEROR HF-Preserved and EMPEROR HF-Reduced trials in heart failure with preserved and reduced ejection fraction are well into recruitment of people both with and without diabetes; both are slated to complete June 2020. Given robust CVOT and real-world evidence in factor of a class-wide heart failure benefit for SGLT-2s, these readouts are strongly expected to be positive. Finally, the EMPA-KIDNEY outcomes trial will enroll ~5,000 patients with CKD, with or without diabetes. On the heels of J&J’s CREDENCE renal outcomes trial for Invokana, which was just stopped a year early for meeting its primary endpoint, we have incredibly high hopes for the success of EMPA-KIDNEY (and Dapa-CKD for AZ’s Farxiga). All of these trials stand to vastly expand Jardiance’s patient population while also filling unmet needs in type 1, heart failure, and CKD.

• The SGLT-2 market remains relatively young and hectic – for both good and bad reasons. Between shifting formularies, evidence for positive benefits (renal and CV protection) and negative side-effects (amputations, DKA in type 1), and still very low uptake (one recent study found that only ~7% of second-line type 2 diabetes prescriptions in the US go to an SGLT-2), it has been difficult to predict how SGLT-2s will fair commercially from quarter to quarter. After seeing Lilly’s 1Q18 financial results, we suspected that Jardiance might have very strong growth to look forward to for the rest of this year, after getting through what was thought to be the brunt of its exclusion from the CVS Health national formulary. We do think other SGLT-2s are probably being made more available to payers at a lower price – as well, of course, some of the revenues we do not see due to Boehringer! For example, J&J has attributed falling Invokana sales to increased discounts, higher rebates, and competitive pressure, indicating some level of pricing pressure in the market. That said, there is plenty to look forward to for Jardiance in 2018 – namely, readouts from EASE-2 and EASE-3 (Jardiance in type 1) and the official start of EMPA-KIDNEY, plus the potential bolstering of class effects from the DECLARE CVOT for Farxiga. We see oceans of room for each SGLT-2 to garner far more commercial success, given both the class’s current low uptake and the variety of cardiometabolic benefits these agents offer.

4. Basaglar Revenue >Doubles YOY to $202 Million; Impressive 22% Sequential Rise; Accounts for 21% of Diabetes Portfolio Growth

After underwhelming with an 8% sequential rise in 1Q18, Basaglar (biosimilar insulin glargine) more than doubled its revenue YOY (+133%) and posted 22% sequential growth to hit $202 million in 2Q18. Notably, this represents only Lilly’s share of Basaglar sales. Our rough estimate for total global Basaglar revenue in 2Q18 is $404 million; BI’s portion is not publicly disclosed, so we speculate that the companies split revenue 50/50. This does not meaningfully affect YOY or QOQ growth. While Basaglar was not discussed beyond name on the call, it did drive 21% of growth in Lilly’s diabetes portfolio and has now grown to comprise 8% of total diabetes sales, up from 19% and ~7% in 1Q18. Within the US, Basaglar now holds ~16% of total basal insulin prescriptions (slide 30) – this is very positive growth, and we expect more details on basal insulin market shares when Novo Nordisk reports on August 8. We’re encouraged to see Basaglar’s volume and revenue gains go hand-in-hand, especially in the context of continued pricing pressure in the US. Basaglar is preferred over Lantus (Sanofi’s insulin glargine) on the CVS Health and UnitedHealthcare formularies, as well as Medicare Part D. On the subject of pricing, management did not mention Lilly’s new Diabetes Solution Center, which will launch August 1 with an aim to connect US patients to their most affordable insulin. By geography, Basaglar grew 163% YOY and 23% sequentially to $157 million in the US, while OUS revenue grew 67% YOY and 16% sequentially to $45 million.

• Two other biosimilar insulin glargines are on the horizon: Merck’s Lusduna Nexvue and Mylan/Biocon’s Semglee. While both products should eventually help push down costs for patients, they are each currently facing patent infringement lawsuits from Sanofi. Lusduna Nexvue has been tentatively approved by FDA, pending lawsuit resolution. Semglee recently received a CRL from FDA despite approval in Europe. Notably, Basaglar faced a similar lawsuit from Sanofi back in the day, and we expect both Lusduna Nexvue and Semglee to reach the market within the next few years. Having multiple biosimilars available could be a big win for patients, not only in terms of expanded choice, but because typically two or more generics are needed on the market to appreciably drive down patient costs.

Lilly’s Reported Basaglar Sales (3Q15-2Q18)

5. Humalog Sales Rise 14% YOY to $770 Million; Management Cites Lower Medicaid Utilization, Higher Realized Price as Main Contributors

Humalog (insulin lispro) revenue of $770 million grew 14% YOY from $678 million in 2Q17, falling 3% sequentially from $792 million in 1Q18. Humalog continues to make up a sizable chunk of Lilly Diabetes (32%), and we’re impressed that such an established product drove 17% of growth in the company’s diabetes portfolio in 2Q18. Looking more closely at our model (see the graph below), however, we also note that 2Q17 was a low-point in revenue for Humalog. This means that the “excellent” 14% YOY growth extolled by management in 2Q18 occurred against a relatively easy comparison and is likely reflective of a short-term fluctuation in product revenue vs. a long-term growth trend. In what Mr. Conterno deemed an “excellent” quarter for Humalog, lower utilization in the Medicaid channel was again identified as a key factor in the product’s success. This was clarified in the press release, which detailed higher realized prices due to changes in rebates, discounts, and segment mix, plus a small contribution by increased volume. This is concerning, as it suggests that Medicaid beneficiaries may have limited access to Humalog, and that patient assistance funding may be on the decline for those commercially insured – this would be very negative for patients, obviously. It also speaks to the power of insulin pricing pressure, which remains abundant in the rapid-acting insulin landscape. It is our understanding that PBMs and payers view all products in this class (Humalog, Novo Nordisk’s NovoLog, Sanofi’s Apidra, MannKind’s Afrezza) as largely interchangeable, pitting manufacturers against one another in a competition to offer higher rebates in exchange for exclusive formulary positioning. To this end, we can only speculate that Lilly pays high rebates for commercial coverage of Humalog, driving down realized profit. We’re curious to know what prescription volume looks like in the Medicaid channel, and how the average patient discount on Humalog specifically may have changed from 2017 (when Lilly’s mealtime insulin sales were decidedly more sluggish). Altogether, we’d love to see more resources being put towards access and affordability of insulin, and in our opinion, it’s high time for better policies around pricing and reimbursement, and for a bright light to be shined squarely on PBMs so that their value is clarified. We’d also like to understand profitability of the diabetes business and to see what the investments are in marketing and R&D. Currently, the US is subsidizing R&D for the world in diabetes and presumably this will not be sustainable. At the same time, many patients do not understand the tradeoffs between greater investment in affordability programs vs R&D vs marketing (Kelly and Adam included) and more information on this would be very helpful.  

• While no specific details on Humalog volume were released, Mr. Conterno clarified that 9 points of the 14% YOY growth in Humalog revenue was due to changes in estimates for rebates and discounts, in line with the higher realized pricing detailed above. He left the other 5% unspecified, though presumably at least some of this was due to volume growth in non-Medicaid channels – that is, higher TRx in commercial and Medicare channels. 

Humalog Sales (1Q11-2Q18)

• Mr. Conterno conjectured that Sanofi’s new biosimilar Admelog (biosimilar insulin lispro) will likely have its biggest impact in Humalog’s share of the managed Medicaid population. He acknowledged that it is difficult for Lilly to forecast exactly how Admelog will affect future revenue, but it’s clear from our view that Humalog sales have not yet felt the full brunt of its new biosimilar competitor, which Sanofi has notably positioned as a revenue driver for 2019 rather than 2018. Of course, Lilly management has maintained that Admelog won’t be a significant commercial threat (see CEO Mr. Dave Rick’s comments from JPM that Humalog and NovoLog are “already on the floor, with some of the lowest pricing in the market.” ) because realized price on rapid-acting insulin is already at the floor (or said another way, rebates are already at the ceiling). Perhaps this will be true in commercial channels, while Admelog will effectively “steal share” among Medicaid beneficiaries – Sanofi has stated that securing strong Medicaid coverage is its first priority for Admelog (which launched earlier this year in the US). For now, we’re in wait-and-see mode on the landscape of biosimilar mealtime insulins.

• US sales of Humalog surged 19% YOY but fell 8% sequentially to $465 million. Meanwhile, OUS sales grew 6% both YOY and QOQ to $305 million. This once again highlights the persistent problem of US pricing pressure around insulin.
  

6. Tradjenta Sales Flat at $142 Million (No Change YOY, 1% Sequential Growth); No Discussion of CARMELINA Topline Results

Lilly’s reported revenue from BI-partnered DPP-4 inhibitor Tradjenta (linagliptin) was flat YOY and sequentially at $142 million in 2Q18. Tradjenta comprised 6% of total diabetes revenue (contributing 0% of growth). We estimate total Tradjenta franchise revenue, including BI’s portion (not reported publicly), at $394 million worldwide ($158 million in the US and $236 million OUS in 2Q18). These calculations assume that Lilly collects ~36% of Tradjenta franchise sales in all geographies. Per the norm for Tradjenta, a majority of sales came from OUS markets ($85 million, +4% YOY, -2% QOQ) vs. the US ($57 million, -6% YOY, +5% QOQ); these mixed results are the norm for the DPP-4 inhibitor market, which overall has tightly but consistently fluctuated over the past few years. Tradjenta has actually been a rare if modest success story as far as DPP-4s are concerned, climbing a gentle growth curve since its launch. During Q&A, Mr. Conterno indicated that Tradjenta volume is steady (i.e. little growth or loss in total prescriptions QOQ), and that realized price is holding steady. Interestingly, Merck recently commented that volume for Januvia (sitagliptin) is increasing even while revenue falls, which signals pricing pressure (high rebating and generous patient discounts). Unfortunately, manufacturers rarely comment on DPP-4s, so it’s hard to tell exactly what’s going on in terms of volume vs. revenue. Despite DPP-4s falling out of fashion with some providers, we expect the class will continue to pool to nearly $10 billion annually for some time to come. While GLP-1s and SGLT-2s offer better glucose-lowering, weight loss, and often even CV and renal protection, DPP-4 still have, by far, the best safety, tolerability, and usability profile known in diabetes, and we continue to see a vital role for these agents earlier in the course of disease/in recently-diagnosed patients. We can’t help but imagine what role they might play in prediabetes (though, of course, this would be off-label use). Finally, while DPP-4s remain out of reach for so many patients, their use will only increase once they go generic.

Lilly’s Reported Tradjenta Sales (2Q11-2Q18)

• No mention was made of recently-released topline CARMELINA results, which showed Tradjenta’s CV safety vs. placebo (the readout was listed on slide 6). Although not surprising, CARMELINA found that Tradjenta is non-inferior to placebo on three-point MACE, a result in line with those for Merck’s Januvia, Takeda’s Nesina, and AZ’s Onglyza. Full results are scheduled to be formally presented at EASD 2018. Tradjenta has often been touted by Lilly/BI as the only DPP-4 inhibitor available that does not require dose-adjustment with declining kidney function; in this light, we remain particularly curious about upcoming renal data from CARMELINA that could indicate long-term renal safety (and maybe protection?). Lilly’s presentation (slide 20) also notes that topline disclosures for CAROLINA, comparing linagliptin vs. SU glimepiride, could occur in 2018. CAROLINA is another very intriguing trial, as evidence for linagliptin’s long-term superiority over SUs could further shift opinion away from these generics. Sulfonylureas are still prescribed today for their low cost, despite targeting no core pathophysiology of diabetes and despite causing hypoglycemia, weight gain, beta cell burnout, and possible CV harm. Seeing as generic DPP-4s are due to enter the market within the next ~5-10 years, patients would widely benefit from evidence of their superiority over sulfonylureas.

Pipeline Highlights

7. GIP/GLP-1 Dual Agonist to Enter Phase 3 by Early 2019; Phase 2 Data Presentation Slated for EASD 2018; Management Emphasizes New Standard for Efficacy

With phase 2 data available in-house, Lilly management announced the advancement of their GIP/GLP-1 dual agonist into phase 3, with trial initiation anticipated in late 2018 or early 2019. Phase 2 data will be presented in October at EASD 2018 in Berlin, and an investor call will follow the presentation – indicating to us that Lilly is very enthusiastic about and committed to this candidate. Moreover, with this decision, Lilly’s candidate has become the leader (for now) of the much-hyped dual-agonist landscape: Our GIP/GLP-1 competitive landscape includes a handful of other phase 1 and preclinical candidates (including GIP/GLP-1/glucagon tri-agonists from Novo Nordisk and Sanofi), while the larger GLP-1/glucagon competitive landscape features four promising phase 2 candidates. To our knowledge, Lilly’s is the very first of these dual agonists to reach phase 3 for diabetes. What’s more, Head of Diabetes and SVP Mr. Enrique Conterno emphasized the company’s “high bar for any type of next-generation incretin,” and Lilly has previously explained how critical it is for this candidate to demonstrate superiority vs. existing GLP-1 agonists, so we have high expectations for the phase 2 data. We’re very excited to see Lilly moving quickly on this candidate – if everything goes as planned, we’d estimate it could be approved by 2021 – and, pending the phase 2 readout, we’re also curious about the therapy’s potential in obesity. For reference, the candidate, LY3298176, completed a phase 2 trial (n=300) in type 2 diabetes in March 2018, in which it was compared to both placebo and dulaglutide (Trulicity); LY3298176 is an injectable therapy. Lilly CSO and SVP Dr. Daniel Skovronsky explained the company’s commitment to building on its experience with incretin therapies, also pointing to the company’s oxyntomodulin analog (GLP-1/glucagon dual agonist) and describing the efficacy of the GIP/GLP-1 as highly-differentiated on weight loss and glucose control.

• So far, diabetes combination therapies – to date, combinations of already-existing drugs – have seen limited commercial uptake; will a medication that is inherently a combination be stymied by the same factors? A GIP/GLP-1 or GLP-1/glucagon formulation is inherently a combination, but we’re optimistic that patients, providers, insurers, and manufacturers might treat them as monotherapy, because they are being developed as such. Hopefully, this will translate both to better reimbursement and greater clinical utilization. To be sure, this is a complex issue with important nuances between therapies. SGLT-2/DPP-4 inhibitor combinations (Lilly/BI’s Glyxambi, AZ’s Qtern, and Merck/Pfizer’s Steglujan) are one major class, and our sense is that the major barrier to uptake of these has been reimbursement. Sales representatives have actually shared that they typically recommend, for example, that a patient uses separate copay cards to buy an SGLT-2 and DPP-4 inhibitor because it’s significantly cheaper than buying the more expensive combination pill: Convenience comes at a premium. But we’ve been even more disappointed at the low commercial success for fixed-ratio basal insulin/GLP-1 agonist combinations (Novo Nordisk’s Xultophy and Sanofi’s Soliqua). While these have done slightly better in 2018, on the heels of what we perceive as increased commitment from the manufacturers, they’re still being devastatingly underutilized. Thought leaders extoll the benefits of these agents – glucose-lowering on par with basal-bolus therapy, without hypoglycemia risk or weight gain – but clinical enthusiasm has far outpaced commercial performance. The issue partially stems from very narrow indications and the inherent limitations these fixed-ratio combinations place on which basal and GLP-1 a person must use, as well as on dose titration (i.e. you can’t combine Lantus and Victoza, and the ratio of insulin to GLP-1 cannot be adjusted). In addition to barriers of cost and reimbursement, however, our understanding is that many providers aren’t used to initiating combination therapy, especially early on in the course of disease. This “treat-to-fail” paradigm is starting to be overturned in guidelines, and we think dual agonists could actually play a role in helping to break down these traditional barriers to combination therapy, faster.

8. Nasal Glucagon Filed with FDA and EMA in First Submission from Next-Gen Glucagon Competitive Landscape; Decision Expected in ~2Q19

Lilly has submitted nasal glucagon to both FDA and EMA, per the company’s pipeline update (slide 19). This marks the first regulatory submission to come out of the next-generation glucagon competitive landscape. Assuming a standard 10-12 month review period, we estimate that an FDA decision on Lilly’s nasal glucagon will come in 2Q19, or even early 3Q19 depending on when exactly Lilly submitted the NDA. Also coming down the pipeline, Xeris is slated to submit its liquid-stable glucagon autoinjector (MyGluca) in 3Q18, a small delay that accompanied the company’s IPO, and Zealand is on track to submit its liquid-stable rescue pen in 2H19. Another handful of earlier-stage candidates stand to further expand glucagon offerings. All in all, we’re very excited to see this submission reach the desks of both agencies: Current glucagon reconstitution kits are cumbersome at best, which makes them dangerous at worst (in that they can take too long to use), and we’re thrilled that at least one of these game-changers could be on the market by the end of next year. Moreover, we fully expect both nasal and liquid-stable glucagon to be available by 2020 and we expect this market to grow significantly.

• Much remains to be determined in terms of market potential, but we certainly see potential for multiple new glucagon products to be commercially successful. Lilly’s Glucagon grossed $32 million in 2Q18 (quarterly sales have ranged from $26-$46 million over the past two years); unfortunately, Novo Nordisk does not break out GlucaGen sales, but we understand that they’re in the same ballpark. Xeris has officially estimated US market potential at $2 billion annually based on current glucagon kit prices and a patient population of 3.5 million, but we consider this number quite ambitious. On the other hand, Zealand estimates a >$700 million market by 2025, a more reasonable estimate that still represents more than doubling from the current base of $300 million/year (a number Zealand has supplied). To be sure, market penetration could be greatly improved, and new, patient-friendly products should help get glucagon into far more patient hands; our sense is that many patients and families do not widely embrace or consider glucagon as useful or necessary, but a more convenient product could really shift that thinking. As such, innovative glucagon products certainly stand to expand the current market. We would love to see Lilly and other manufacturers leverage these products in campaigns to both improve glucagon prescribing practice and promote uptake within the diabetes community.

Lilly Diabetes Pipeline Summary

The table below reflects the latest updates, as far as we are aware, on Lilly’s diabetes pipeline products. Items highlighted in yellow indicate notable changes to the pipeline in 2Q18.

Select Questions and Answers

On Policy and Pricing

Q: What do you view as the best-case scenario that could come out of the blueprint proposed by the HHS Secretary and White House administration?

Mr. Dave Ricks (CEO, Lilly): On the policy side, of course, the blueprint was rolled out in May, and recently all actors in the industry seem to have responded to the request for information. So you're asking me to guess at where this will land, and that's a really difficult thing to do at this point because there are dozens of ideas in there.

I think we can comment on actions that have been initiated, though, because I think that indicates where the administration might be going first. And of course, many of you may have noted, last week the administration sent to OMB a proposed rule change for the Anti-Kickback safe harbor which relates to rebates. We don't know the substance of that rule change, but this is heavily commented on through the RFI via questions. And I think you're right to assume that there would be some changes to the Anti-Kickback Statute as it relates to rebate treatment.

We don't know what those are, and I think we're preparing for all scenarios there. But this is consistent with the broad theme of looking at ways to reduce the gross-to-net spread in the industry and to allow patients to have a price point in a high deductible or without insurance that's closer to net pricing. Overall, we support that. I don't think that's a bad thing for innovators, and we'll have to watch how that rule develops. We don't have any specifics as of today.

You also saw the administration move on putting a task force together related to addressing off-patent brands who take super inflationary price increases. In the absence of IP this is really a regulatory failure, from our perspective. The administration's talking about importation. We think that's the wrong road to go down but rather we need to fix the regulatory system to begin with. Nonetheless, clearly, that is a hot-button issue, and we agree it should be solved. The method we disagree with, but we'll have more to say about that in the future.

I think I said in the Q4 call in January, we could expect a busy year in regulatory reform, and I think in fact that's what we're seeing. For innovators, at the end of the day, as long as we can embrace a pro-innovation, market-driven set of changes that embraces choice, I think we're going to be fine. Getting from A to B, we'll have to see how these changes unfold, and as of today, it's difficult to say more than I did already. But of course, we watch it very carefully, and at various investor updates will be happy to provide additional commentary as additional actions come forward.

Q: On price increases, it seems like the tone across the industry is a bit more restrained than the past on willingness to take increases. Is this a 2018 dynamic or do you see this trend continuing out into 2019 and beyond?

Mr. Ricks: Net pricing is subject not just to list price changes. And as you know, the yield of those is quite different across different categories, but also subject to prior period adjustments on assumptions made in gross-to-net, on current channel mix and many, many other things. So, you'll see, I think, in the U.S. we're reflecting a 1% positive, which reflects prior increases as well as all those other dimensions. As Josh said, we don't have list price changes in our outlook, and it's still very robust, and that's kind of where we're focused, just driving volume of new products here in the U.S. and abroad.

Q: Lilly's gross-to-net of 48% is the widest in the industry, largely because of your diabetes franchise. If there is a change to rebates, what would you expect to happen to volumes? Lilly has about $10 billion in rebates. How can this not be a net positive even if you take into consideration lower volumes and potentially lower list price? How are you thinking about the different scenarios that can play out here?

Mr. Ricks: The huge opening caveat here is we don't know. We don't know what this rule is. We don't know where the system will go. We paused on the list price change in the U.S. waiting to see what's happening here. But just to get the numbers right, we reported last year a 51% gross-to-net spread.

You're right in that the diabetes segment in particular drives that, primarily because you have a very large Medicaid population and you have very large gross-to-net spreads in Medicaid approaching the federal cap at 100%. We've been pretty open about all that. It depends on how this rule would be enacted, whether it would change the Medicaid system, whether it would impact commercial, etc, to really guess what would be on the other side of it.

As I said in my first response, though, we make innovative products, hopefully products people want and need. I think the current system has resulted in a structure of the industry where the hospital systems, insurance carriers, distributors and manufacturers seem to be fine with the current system, but the problem is we're shifting too much cost via list-pricing directly to consumers. I have to believe that if consumer pricing came down, it would improve volume, improve medication adherence, improve outcomes for patients. So we think, in general, this direction of travel is a good one.

Now, we have to see all the specifics because, clearly, there's a lot of room for disruption in a negative way here, and it would be very difficult for Lilly to predict what would happen to the gross-to-net line you're talking about or volume specifically without seeing the totality of the regulatory reforms.

On Therapies and Pipeline

Q: Given the strong profile for the current injectable GLP-1s and the emerging oral GLP-1, can you put a little more meat on the bones in terms of that bar you talked about [for the GIP/GLP dual agonist]? Is that the primary endpoint or is it a multitude of endpoints that shape your thinking there?

Dr. Daniel Skovronsky (CSO, Lilly): We previously said that we see incretin biology as an important platform for Lilly to continue to build on, that there are a lot more opportunities here for patients. And so, over the years, we've invested here and developed and tested now a number of different molecules that could offer differentiated efficacy in this platform.

So, for example, with the oxyntomodulin molecules, we're now testing a second one of these in clinical trials, and we have the GIP/GLP and other molecules that are in earlier development. In each of these cases, we're really looking for what we consider to be breakthrough efficacy. In the case of GIP/GLP here, we were encouraged that dual agonism of both of these receptors, which we think is very important in our preclinical studies, showed highly differentiated weight loss and glucose control, so we're encouraged by that. We took it into phase 1 clinical trials. We commented that we're encouraged by what we saw there and, therefore, moved it into phase 2 clinical trials.

Mr. Enrique Conterno (Head of Diabetes and SVP, Lilly): I think you've framed it well. But we do have a high bar for any type of next-generation incretin, and this product clearly met that bar. But we won't be able to comment on the specifics of the efficacy; we do plan to have a full presentation at EASD, and then an investor call following that.

Q: Could you help us better understand why [Jardiance and Tradjenta] were flat-ish sequentially, despite the prescription growth that we saw?

Mr. Conterno: Let’s start with the easier one. Tradjenta net sales in the U.S. reflect the trend that we see when it comes to total prescriptions. Basically, it's a steady number of prescriptions quarter-on-quarter and that's basically what we see when it comes to net sales.

When it comes to Jardiance, just maybe to provide a bit more color, we show a 58% TRx growth versus the previous year quarter-on-quarter. On a sequential basis, that growth is 13%. So there’s very solid growth when it comes to Jardiance. But in the net sales, there's a disconnect there because the net sales for the quarter versus the previous year only increased 28%. We did have an adjustment due to changes in estimates for rebates and discounts; normalizing for that, the growth versus the previous year would have been 46%.

Now, when we look on a sequential basis, in the case of Jardiance, some of those changes in the estimates for rebates and discounts really belonged in Q1, so you basically have a double whammy effect when you look at the sequential quarter-on-quarter comparison.

Q: On Humalog, you've had some benefit in terms of the changes in estimates, but just on the rebates. Can you talk about where you are in terms of comps? And then kind of moving forward into 2019, how comfortable are you at this stage that the biosimilar competitor won't really represent a meaningful source of competitive pressure to the business?

Mr. Conterno: Humalog had an excellent quarter. We did have a benefit in Q2 due to changes in the estimates for rebates and discounts. That benefit was, relative to the previous year, about 9 points, so an important benefit. It’s difficult for us to be able to project forward, but we are seeing a bit of a benefit also when it comes to mix when we look at the different segments, which is a positive for Humalog and for the insulin franchise.

Now, as it relates to Admelog, it's very difficult for us to forecast and predict how the competitor will basically play in the market. Clearly at this point in time, we basically see them as gaining share in particular in the area of Managed Medicaid. But it's difficult for us to predict what type of access and uptick they'll be able to have in 2019.

Q: What gives you confidence that you will hit superiority versus just non-inferiority in the REWIND trial for Trulicity. And even if it's not superior, can Trulicity still continue its robust growth trajectory?

Mr. Conterno: Trulicity is having fantastic growth. Clearly the GLP-1 market is growing very fast at 26%, but Trulicity's also having continued share growth over the last few months. We have an excellent access position and, quite frankly, an unmatched patient experience when we look at the real-world efficacy that Trulicity delivers. When we also look at patient adherence, very simply, delivered. We’re very excited about the core performance of the product. Clearly, REWIND, as we've said before, is an important trial for us and we believe that we've designed this trial in the appropriate way. We expect that we're going to have a top line sometime by the end of the year, but one of the differences for REWIND versus other trials is the time the patients are going to be on the product. This is not a short trial. It's fairly long. And we expect that patients will be on Trulicity for over five years on average.

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