Lilly(Eli) & Company (LLY) — Q2 2021 Earnings Call Transcript

Good morning. Thank you for joining us for Eli Lilly and Company's Q2 2021 earnings call. I'm Kevin Hern, Vice President of Investor Relations. And joining me on today's call are Dave Ricks, Lilly's Chairman and CEO; Anat Ashkenazi, Chief Financial Officer; Dr. Dan Skovronsky, Chief Scientific Officer and Medical Officer; Anne White, President of Lilly Oncology; Ilya Yuffa, President of Lilly Bio-Medicines; Mike Mason, President of Lilly Diabetes; and Jake Van Naarden, CEO of Loxo Oncology at Lilly. We are also joined by Lauren Zierke and Sara Smith of the Investor Relations team. During this conference call, we anticipate making projections and forward-looking statements based on our current expectations. Our actual results could differ materially due to a number of factors, including those listed on Slide 3. Additional information concerning factors that could cause actual results to differ materially is contained in our latest Forms 10-K and subsequent Forms 10-Q and 8-K filed with the Securities and Exchange Commission. The information we provide about our products and pipeline is for the benefit of the investment community. It is not intended to be promotional and is not sufficient for prescribing decisions. As we transition to our prepared remarks, a reminder that our commentary will focus on non-GAAP financial measures. Now I'll turn the call over to Dave for a summary of our second quarter results.

Thank you, Kevin. Q2 of last year was the peak of the pandemic’s negative impact on our business. And one year later, I'm proud of the innovation and resilience displayed by my Lilly colleagues to deliver against our objectives in new ways, while also mobilizing to develop treatments to help combat COVID-19. Looking at Q2 2021, we were encouraged by the increasing worldwide vaccination rates, as well as the underlying environment in most of our major markets. COVID-19 related stocking in Q1 followed by destocking in Q2 of last year complicates quarterly performance comparisons. Therefore, looking at revenue growth in the first half of 2021 better reflects the underlying trends in our business. On today's call, we will provide year-over-year comparisons for both Q2 and the first half of the year. In the first half of 2021, we delivered 11% growth in our core business. This excludes COVID-19 antibody revenue; this was buoyed by strong volume-driven growth across key brands in major geographies, including the U.S., Europe, and China. Turning specifically to Q2, revenue grew 23% compared to Q2 2020, or 20% in constant currency. This performance was driven entirely by volume growth of 22 percentage points. As previously highlighted, in Q2 2020, we saw reversal of the $250 million pandemic-related product stocking which occurred in Q1 2020. When excluding COVID-19 antibody revenue, the Q2 2020 COVID-19 related destocking, and the sale of Cialis in China, our core business grew 12% for the quarter, up from 7% in Q1 on the same basis. We were pleased to see sequential top-line growth in the core business this quarter, signaling that healthcare systems continue recovery from the pandemic and the strength of our underlying business. Key growth products continue to drive our revenue growth and represent 54% of our core business this quarter. Our non-GAAP gross margin was 79.3% in Q2, or 79.7% excluding the impact of foreign exchange on international inventories sold. Excluding the FX impact, our gross margin increased by approximately 60 basis points compared to last year. Our non-GAAP operating margin was 29.4%, representing an improvement of nearly 140 basis points. We were pleased to see operating margin expand year-over-year, and we expect continued expansion in the second half of this year. On the pipeline front, we achieved multiple milestones since our earnings call in April, including receiving Breakthrough Therapy designation for donanemab and announcing our plan to submit to the FDA under the accelerated approval pathway; announcing positive Phase 3 results for tirzepatide SURPASS-4 trial with planned global submissions of the SURPASS program for tirzepatide in type 2 diabetes by the end of 2021; obtaining approval for Jardiance in partnership with Boehringer Ingelheim for HFrEF in Europe; and announcing positive Phase 3 results from the EMPEROR-Preserved trial for Jardiance in HFpEF, the first and only successful trial for this patient population; and initiating Phase 3 trial results for pirtobrutinib in mantle-cell lymphoma, tirzepatide in HFpEF, Verzenio in HR+ HER2+ early breast cancer, and now, prostate cancer. We also continue to augment our pipeline with business development deals and announced the acquisition of Protomer Technologies. We welcomed the Protomer team to Lilly and are excited to bring this technology to our diabetes pipeline, as we believe glucose sensing may become the next generation for insulin treatment to improve the quality of life for people living with diabetes. Lastly, on financials, we returned approximately $1.3 billion to shareholders via dividends and share repurchases in the quarter and authorized the repurchase of up to $5 billion in stock in addition to the $500 million authorization remaining under our 2018 share repurchase program. Moving on to slides 5 and 6, you'll see a list of key events since our Q1 earnings call, including a May webcast, which highlighted our updated environmental, social, and governance strategy and our sustainability efforts, as well as the launch of a new ESG website to serve as a comprehensive resource to provide increased transparency regarding the company's ESG goals and progress. Further, as part of our goal to become carbon-neutral in our own operations at our manufacturing plant in Kinsale, we recently inaugurated a new solar field, which is now the largest in Ireland. We also announced donations of COVID-19 therapies at no cost to low-income and lower-middle-income countries heavily impacted by the pandemic, and are proud of the impact we’re having around the world as we work to combat COVID-19. Now, I'll turn the call over to Anat to review our Q2 results and to provide an update on our financial guidance for 2021.

Thanks, Dave. Slides 7 and 8 summarize financial performance in the second quarter and year-to-date. I'll focus my comments on non-GAAP performance. Revenue increased 23% this quarter compared to Q2 2020, or 12% excluding the items Dave mentioned earlier, representing strong momentum for our core business. Given the COVID-19 related stocking and destocking seen between Q1 and Q2 of 2020, our first half performance of 11% revenue growth or 8% in constant currency, excluding COVID-19 antibody revenue is a more accurate reflection of underlying performance, and the sequential quarter-over-quarter revenue growth better represents the strength in our core business. Sequential revenue growth from Q1 to Q2 for core business, increase in vaccination rates in many major markets, and the majority of our sales reps now being back in the field suggest the recovery from the pandemic was in line with our expectation for the quarter. We're particularly pleased with the strong volume growth across key brands like Trulicity, Taltz, Verzenio, and Jardiance. Verzenio in the U.S. grew nearly 6 percentage points in share total prescription exiting June compared to the prior year. While Trulicity, Taltz, and Jardiance increased their leading market share in the same period, class growth accelerated. These products, along with other key growth products, represented 54% of revenue in core business this quarter. Gross margin as a percent of revenue declined 30 basis points to 79.3% in Q2. The decrease in gross margin percent was driven primarily by unfavorable effects of foreign exchange rates on international inventories sold. Excluding this FX impact, gross margin as a percent of revenue grew 60 basis points this quarter. Total operating expenses grew 18% this quarter compared to the same quarter last year. Marketing, selling, and administrative expenses increased 16% as the base period in Q2 2020 included a meaningful reduction in direct-to-consumer marketing and customer-facing expenses as healthcare systems closed. R&D expenses increased 20% driven by investment in exciting late-stage pipeline opportunities, including pirtobrutinib, tirzepatide, donanemab, and lebrikizumab. In Q2, we also invested approximately $85 million in COVID-19 therapy R&D bringing our total COVID-19 R&D investment to approximately $300 million year-to-date. Net of COVID-19 R&D investment, operating expense growth was 18% compared to Q2 of 2020, and 10% for the first half of the year. Operating income increased 29% compared to Q2 of 2020, and operating income as a percent of revenue was 29.4% for the quarter, an increase of 140 basis points compared to the prior year. This increase was driven by revenue growth outpacing expense growth, and we expect continued margin expansion in the second half of 2021. Other income and expense was income of $5 million this quarter compared to expense of $57 million in Q2 2020, driven by income from European patent settlements for Alimta. Our effective tax rate was 14.4%, an increase of 350 basis points compared with the same quarter last year. The effective tax rate for both periods was reduced by net discrete tax benefits, with the lower net discrete tax benefit reflected in the second quarter of 2021. At the bottom line, net income and earnings per share increased 29% in Q2, and 22% year-to-date or 30% and 24%, respectively, in constant currency. On Slide 9, we quantify the effect of price, rate, and volume on revenue growth across the world. We are encouraged by the growth seen across most of our major geographies. This quarter, U.S. revenue grew 18% compared to the second quarter of 2020. Adjusted for COVID-19 antibody revenue and the Q2 2020 COVID-19 related destocking, the core business grew 8% in the U.S., up from 5% in Q1 on that same basis. These results were driven entirely by volume led by Trulicity, Taltz, Verzenio, and Jardiance. Pricing was a 1% drag on U.S. revenue growth this quarter, with increased rebates to maintain excellent access and higher growth in lower net price segments, largely offset by lower utilization into the 340B segment, changes for estimates to rebates and discounts, and to a lesser extent, modest list price increases. The year-to-date price decline of 3% in the U.S. is in line with our net price expectations for the full year. Specific to Taltz in the U.S., performance for the quarter was in line with the expectation we described on the Q1 earnings call. We were pleased to see a return to net sales growth this quarter, as volume gains more than offset price decline. Taltz Q2 performance benefited from a favorable change to prior estimates for rebates and discounts, and COVID-19 related inventory destocking last year. Excluding these items, Taltz still returned to double-digit growth in the second quarter. We believe the net price decline for Taltz in the first half of 2021 represents the underlying full price trend, and that continued volume growth will drive net sales acceleration in the second half of the year. While midterm price trends are currently stable, given increasing variability in payer mix, we continue to expect quarterly variability in reported U.S. net price changes across our business. Moving to Europe, revenue grew 27% in constant currency. Excluding COVID-19 antibody revenue and the negative impact of Q2 2020 COVID-19 related customer buying patterns, revenue grew 14% in constant currency, driven entirely by volume growth, primarily for Trulicity, Taltz, Alimta, and Olumiant. We continue to be pleased with the momentum of our business in Europe and expect continued growth in the second half of this year, excluding the expected impact from the loss of exclusivity for Alimta. In Japan, revenue grew 2% in constant currency, driven primarily by the launches of Olumiant and Emgality. Revenue growth in Japan continues to be negatively impacted by the decreased demand for several products that have lost market exclusivity. But our key growth products grew 21% in Q2 in Japan and represented approximately 50% of the business there. Recent surges of COVID-19 cases continue to negatively impact recovery in Japan. So we currently expect improved revenue growth in the second half of the year based on the uptake of newer products. In China, revenue grew 106% in constant currency, primarily driven by the divestiture of Cialis and the launches of Tyvyt and Trulicity. Excluding the impact from the Cialis transaction, our revenue in China grew 35% in constant currency. We are excited about the continued momentum in China, as sales of new medicines have accelerated significantly in the past three quarters. Revenue in the rest of the world increased 5% in constant currency, driven primarily by our key growth products. At the bottom of the slide is the price rate volume effect on revenue for our June year-to-date results, which shows double-digit growth across all major geographies, except Japan.

Thanks, Anat. 2021 has clearly been a productive year for R&D at Lilly, with continued strong progress in our pipeline and more potential catalysts on the way. Before I get into the broader portfolio update, I'll spend a few minutes highlighting several updates for our late-stage pipeline. I'll start with donanemab. In Q2, the first amyloid lung agent for the treatment of Alzheimer's disease was approved under the FDA's accelerated approval pathway based on plaque lowering, which we believe reflects a shift in policy and sets a new path for Alzheimer's drug approval in the U.S. Lilly has long been an advocate for using biomarkers for amyloid plaque and neurofibrillary tangles to identify patients for treatment and to monitor their response to therapy. We were pleased to see the FDA's conclusion that improvements in brain pathology are appropriate surrogates for clinical efficacy of Alzheimer's drugs. Based on data we've seen to date, we believe donanemab clears plaque faster and deeper than previously seen with other therapies and achieved complete plaque clearance in a majority of patients in TRAILBLAZER-ALZ after only a limited duration of dosing. On the basis of the clinical evidence for donanemab, we were pleased to have received Breakthrough Therapy designation from the FDA. At the Alzheimer's Association International Conference last week, we shared additional important analyses from donanemab TRAILBLAZER-ALZ. Briefly, I'll highlight several findings. First, we shared detailed exploratory statistical analyses comparing a variety of methods beyond MMRM and DPM, as summarized on Slide 14. We are pleased to see that these new analyses showed consistency of effects on primary and secondary outcomes across all statistical methods. Notably, all of the new analyses conducted showed good separation of treatment from placebo with statistical significance achieved for most endpoints at nearly all relevant time points measured. The robustness of the treatment efficacy across analytical methods increases our confidence in the potential clinical benefit of donanemab. While all statistical methods evaluated showed similar results, we note that the Bayesian disease progression model, DPM, closely reflected the raw observed data with the smallest standard error of any method. These results reinforced our hypothesis that DPM is a preferred analytical method for Alzheimer's trials. Additionally, we shared new data showing a relationship of amyloid plaque reduction and slowing of cognitive decline, as shown on Slide 15. To our knowledge, this is the first time such results have been available. When we initially reported the results of TRAILBLAZER-ALZ, we commented that at a group level, patients treated with donanemab showed both statistically better plaque reduction and slowing of cognitive decline at 18 months. But patient-level correlations between the degree of plaque reduction and the magnitude of slowing or cognitive decline were not significant. Now using a more sophisticated PK iADRS exploratory analysis that uses all the available time course data, we showed a highly significant relationship between the degree of amyloid plaque reduction and slowing of cognitive decline with P less than 0.001. The Conrado model, shown here, was published in 2014 and is the result of efforts from the Coalition Against Major Diseases, CAMD, which collected placebo data from 15 randomized trials, including almost 4,500 participants. We introduced a treatment arm and incorporated percent amyloid plaque removal into this model to generate these results. We believe this is important support for the use of amyloid plaque reduction as a surrogate for clinical efficacy. Notably, these data suggest that full clearance of amyloid plaque is required for the highest efficacy as model results predict that patients achieving a 100% clearance of amyloid plaque could have more than a 40% slowing of disease progression. Moving to Slide 16, we show an exploratory analysis looking at the effect of donanemab's plaque clearance on the development of tau pathology. Tau pathology is an exciting biomarker since measures of Alzheimer's disease tau, unlike measures of amyloid plaque, have been correlated with clinical measures of cognitive and functional decline as noted here. Importantly, we have previously shown that donanemab-treated patients had slower accumulation of regional brain tau pathology than placebo-treated patients. This is an important finding because the amount of brain tau pathology is an excellent predictor of subsequent cognitive decline, a finding we observed with solanezumab in EXPEDITION3 and reproduced once again in TRAILBLAZER-ALZ. Now we've extended these results to show that the donanemab-treated patients who achieved complete clearance of amyloid plaque by 6 months had the most marked slowing of tau spread with nearly complete abrogation of progression in the frontal lobe. This reinforces our hypothesis that both deep and rapid amyloid plaque clearance is required for optimal drug efficacy. With this new data we presented last week, we have now linked degree of amyloid plaque reduction with degree of clinical benefit as well as degree of amyloid plaque reduction with degree of benefit on brain tau pathology, which is itself linked to clinical benefit. As displayed on Slide 17, we have just recently obtained data with our plasma tau biomarker, phospho-tau217. These new data demonstrate that amyloid plaque clearance with donanemab also resulted in reversal of the typical increases of phosphorylated tau seen in the blood with decreases from baseline of more than 24% and a change from the untreated arm with a P-value of less than 0.0001. This highly significant effect was seen as early as 3 months following initiation of treatment and could reflect a combination of less tau spread in the brain as well as less neuronal damage, which could account for tau leakage into the periphery. You can see on the right side of the slide that the effect on plasma tau is also correlated to a degree of plaque reduction, with nearly every patient on treatment who achieved substantial plaque clearance showing flat or declining plasma phospho-tau. We are delighted to see the potential utility of P-tau217 not just for diagnosing disease but also for monitoring treatment efficacy. We believe this could be another important contribution to the Alzheimer's field. Finally, on Slide 18, the significant relationship between plasma P-tau217 reduction and the slowing of cognitive decline is shown. This additional biomarker for efficacy links the donanemab mechanism of amyloid plaque clearance with positive effects on both clinical outcomes and tau pathology. These data suggest that patients who achieved a 30% decrease in P-tau217 from baseline showed more than a 40% slowing of disease progression. The three main findings I just discussed. One, the consistency of clinical benefit across statistical methods; two, the correlation of plaque lowering the clinical benefit, the patients who achieved the greatest plaque clearance; and three, the correlation between achieving complete plaque clearance and beneficial effects on tau pathology seen in the brain and measured in the periphery, which themselves are predictors for clinical benefit, strongly support the efficacy of donanemab and give us confidence that the remarkable levels of amyloid plaque clearance achieved by donanemab could translate into a meaningful breakthrough for patients. Moving to Slide 19, accordingly, we've announced that we plan to submit to the FDA under the accelerated approval pathway before the end of this year based on data from completed studies, supplemented by additional safety data from the ongoing TRAILBLAZER-ALZ 2 study. We remain focused on enrolling TRAILBLAZER-2 with the aim to replicate the positive results of TRAILBLAZER-1. Replication is important to overcome skepticism in the field. We hope that TRAILBLAZER-ALZ 2 will generate important confirmatory data for patients, physicians, and payers and help us understand how to make sure the right patient gets the right duration of therapy at the right stage of disease. We are pleased to announce today that we have closed screening for TRAILBLAZER-ALZ 2, with an adequate number of subjects now in the trial's screening process to fully enroll the study. Given that conducting and processing and imaging studies used during screening take several weeks to complete, we expect that the final subject will complete screening procedures and receive their first dose of donanemab or placebo by the end of the third quarter, and the study will complete 18 months later. Given this progress in enrollment, we are confident that we will achieve the number and duration of drug exposures needed to appropriately characterize the safety profile of donanemab, allowing for regulatory submission by the end of this year. Discussions with the FDA are consistent with our prior statement supporting a submission before the end of 2021. I also want to provide a few comments on how we believe the national coverage determination opened for monoclonal antibody therapies targeting amyloid by the Centers for Medicare & Medicaid Services may impact Lilly and donanemab. We believe this NCD is a clear opportunity to focus treatment on the patients most likely to benefit from amyloid plaque-reducing therapies. This would align with our goals, which have long been to use advanced diagnostic tools to identify the right patients that can benefit the most from amyloid-reducing therapies. We're particularly encouraged that our progress with the plasma P-tau217 assay could open up broader access to diagnostic tools. Still, despite the advances in diagnostics and the promise of donanemab, we acknowledge the current skepticism in the national discussion. And we hope that each drug will be evaluated by payers and prescribers based on its own data. This could be particularly important given the data I've shared today, which suggests that the degree of donanemab's amyloid plaque clearance relates to clinical benefit. In summary, we look forward to submitting donanemab to the FDA later this year with the potential to bring a robust amyloid plaque-clearing agent with limited treatment duration to market for early symptomatic Alzheimer's patients in 2022 with potential replicated clinical efficacy results expected in 2023.

Thank you, Dan. I appreciate that. Before we proceed to the Q&A, I want to summarize the progress we've made this quarter. We've experienced strong performance in our core business during the first half of the year and gained momentum in Q2. This growth is primarily due to increased volumes across our key brands in most major regions. We're happy to report sequential top-line growth this quarter alongside year-over-year margin expansion. We've made significant strides in developing new medicines, and Q2 marked another positive period for our pipeline as we announced plans to submit tirzepatide for type 2 diabetes and donanemab for Alzheimer's disease later this year, as well as an approval for Jardiance in HFrEF, with positive results in HFpEF as Dan mentioned. We returned nearly $800 million to shareholders through dividends and completed $500 million in share repurchases, underscoring our confidence in the continued strength of our business. As we look ahead to the rest of the year, we remain optimistic about our long-term outlook. Before we move into the Q&A, I want to also mention that we will hold a live investor meeting this December to showcase our R&D pipeline and progress. During this meeting, we will also provide our initial guidance for 2022. Due to limited physical space, this event will include a webcast. We hope to host it in person, but we are closely monitoring the pandemic situation and will pivot to a virtual format if necessary. Our Investor Relations team will reach out in the coming weeks with invitations and further details about this meeting. Now, I will hand it over to Kevin to lead our Q&A session.

Thanks, Dave. We'd like to take questions from as many callers as possible, so we ask that you limit your questions to two per caller. Lois, can you please provide the instructions for the Q&A session, and then we're ready for the first caller?

Maybe, Dan, I was just wondering if you could elaborate at all on your comments regarding your discussions with the FDA on donanemab. It sounds like they're consistent with your expectations. But any more color you can provide if they've actually signed off fully on your plans to file the BLA? And then how much additional safety data would they want to see from the ongoing TRAILBLAZER-2 study?

Yes. Sure, Terence. Thanks for that question on donanemab and FDA and safety. In June, when we got the Breakthrough Therapy designation and we announced our expectations to file the BLA by the end of the year, that was based on our current understanding of the situation. Since then, things have progressed, and I would say I'm even more confident now than I was then, that we should have an adequate package to support a complete submission by the end of this year. That includes, of course, our confidence that we have enough safety data to support a full evaluation of the benefit/risk of this drug. I think given the limited duration of dosing, that helps as well as given the near completion now of enrollment in TRAILBLAZER-2. So it's our intent then to use combined safety data from the completed Phase 1 and Phase 2 studies as well as an early look at safety data from that ongoing Phase 3 study to support the package. Now of course, with any ongoing study, there's always risk. We don't know what that safety data is going to show. If it's consistent with safety data we've collected prior to the study, then I think we should also be confident that would support a positive benefit/risk assessment and put us on track to launch next year, as we said.

Two, the first one, I'll stay with donanemab. You have suggested in your comments there that there will be a good chance to use some of the biomarkers that you are developing in the early commercial use of donanemab. Can you talk a little bit about what markets do you expect to have provided by when? And how do you see the entire patient passage through the use of donanemab going forward? And how does it differ from other amyloid β? And second, Basaglar seems to have a bit of a price drop this quarter. Can you discuss a little bit what you're seeing here? What are you expecting with the approval of the first interchangeable biosimilars? Any impact there? And as we go forward, how should we think about that franchise?

Thank you, Ronny, for your question about biomarkers and their commercial use. This area has seen significant progress and investment from Lilly, and we remain focused on it. The advancements we are witnessing in Alzheimer's disease are largely due to our ability to select patients for treatment and monitor their responses using biomarkers. We believe these tools should be accessible to patients undergoing clinical treatment for Alzheimer's in the future. Currently, both PET agents, tau PET imaging with Tauvid and amyloid PET imaging with Amyvid, are FDA approved, though availability is limited, especially for Tauvid. However, this could change quickly with the launch of donanemab. The phospho-tau217 assay is expected to be the most accessible option for patients. As we continue to enhance this assay, we are increasingly impressed by its effectiveness in identifying patients and tracking their progress. We are committed to making it available soon. There's often less regulatory burden for in vitro diagnostics compared to in vivo diagnostics, which presents good potential. Regarding patient flow once these tools are approved, it is likely to coincide with the launch of donanemab. Screening might begin with basic cognitive exams by physicians to determine eligibility for early Alzheimer's, followed by a blood test like phospho-tau217. If that result is positive, it may indicate a basis for treatment if supported by data or guide patients to PET scans for further evaluation.

Thank you for your question regarding Basaglar. The performance we observed in the second quarter of 2021 is mainly due to pricing and volume pressures in the Medicaid segment. To clarify how the Medicaid segment operates, there are two types of states: some have a single unified preferred drug list for managed Medicaid and fee-for-service, while others maintain separate preferred drug lists for each. We have noted that when states transition from two lists to one unified list, the financial benefits tend to favor established brands like Lantus. Consequently, in states with managed Medicaid, this leads to a shift back to Lantus. This has impacted our performance in Q2. Additionally, in the managed Medicaid sector, we’ve encountered pricing pressures from Semglee, necessitating increased rebates to maintain our volume. Moving on to Semglee, it's important to note that Semglee has achieved interchangeability only with Lantus, not with Basaglar, so we don’t expect any immediate effects on Basaglar. As for Semglee's performance, it has captured approximately 2% of the market share in total prescriptions and about 1% of new treatment initiations, primarily from Medicare Part A and the Medicaid segment. Semglee is currently priced at $99 per vial and around $150 for a five-pack of pens. We fully support measures that improve affordability for diabetes patients, which is why anyone can purchase their monthly prescription of Lilly insulin for $35 or less through our Insulin Value Program. This program has successfully reduced the average out-of-pocket costs for a monthly prescription of Lilly insulin, which includes multiple vials or pen packs, to $28.05 despite rising health insurance deductibles. It’s encouraging that individuals living with diabetes have access to various options to lower their out-of-pocket expenses. Thank you for your inquiry.

A couple of questions. Just your general thoughts on subcutaneous dosing with antibodies to plaque, does that offer meaningful differentiation? At a high level, the benefits would seem quite obvious to being able to dose a drug at home. But some argue that it falls outside of a Medicare Part B framework, so maybe doctors would be more inclined to stick with an infusion. And I believe you originally did not pursue subcutaneous because you're worried you wouldn't get enough drug across the blood-brain barrier. Roche has shown us that they can achieve that. So can we expect Lilly might also pursue a subcutaneous? And would this require a formal Phase 3 study looking at plaque reduction as a primary endpoint? And then last quick question, why wouldn't something like P-tau217 become a separate meaningful revenue stream in its own right for Lilly?

Okay. Thanks. No, it's a good question and line of questions here on subcutaneous dosing for anti-amyloid therapies. Probably two factors that we have taken into account in addition to the ones you mentioned. First and most important is efficacy for patients. And I think all of the data that we have so far suggest and support the notion that deep and rapid clearance is key here. And so if you're going to go to subcutaneous dosing, it's important to make sure you do get enough drug in so that you can quickly get patients to clear. That's not going to always be possible with every drug. I think the second consideration with subcutaneous dosing is the duration of therapy. So if it's a limited duration of therapy, the difference between IV and subcutaneous, if it's once a month for six months, that's not a big difference between IV and subcutaneous. Whereas if it's for the rest of your life, maybe that is a bigger difference. Finally, with respect to our plans for subcutaneous, I do think it's an important option to offer patients, notwithstanding the previous comments, even for a limited duration therapy, some patients may prefer it, assuming you can get the same kind of efficacy. I think with donanemab, that's unlikely to be possible, and we're not pursuing it given the doses we need and the formulation we have. However, we have a second-generation antibody here that we call N3pG4, which I think is quite likely to be viable in a subcutaneous presentation. And that is our focus of development around N3pG4. My expectation around that is that it should be able to show comparable amyloid plaque lowering with subcutaneous dosing as donanemab does with IV dosing. If so, given the similarities between the drugs, we would seek an accelerated approval pathway for that drug in the future as sort of a subcutaneous version of donanemab, although it is a new enemy. Your second question was around the phospho-tau assay and whether that's a significant revenue stream. It's certainly conceivable, and we haven't sort of thought through all of our commercial plans around that. But really for Lilly, and it may be significant for some companies, for Lilly, though, our focus is on removing barriers for treatment to patients. And so as we think about how we position diagnostics and therapeutics in the marketplace, our focus will be on really making sure that most patients possible can get treated appropriately.

Thank you, Dan and Tim, for your insights on the access and payment environment. At Lilly, our primary goal is to simplify the process for patients to access therapies while we address value afterward. There are certainly short-term advantages to treatments like donanemab in Part B, and these patients will be closely monitored by their physicians at the outset. This class of medications does come with significant side effects that necessitate imaging analysis, making it a highly managed disease. Over time, similar to other treatment classes, as primary care physicians become more comfortable with using therapeutic antibodies for Alzheimer's, we envision a more convenient option being available at local pharmacies for self-injection or caregiver-administered injections. Our plans are aligned towards pursuing both channel options. Although the intensive nature of the treatment and the need for specialists currently lean towards infusions, we are dedicated to both methods and ultimately focused on enhancing patient convenience.

I have a question about donanemab and another about Verzenio. Regarding donanemab, I'm curious about the potential role of Aβ antibodies before we have definitive cognition data. Do you believe that cognition data will significantly broaden the market opportunity for these products? Or do you expect that the usage will be widespread, even if the additional cognitive readouts for donanemab are less conclusive than what we saw in Phase 2? Essentially, do you think the market focus will shift entirely to plaque regression or reduction, or is cognition still crucial for commercial opportunities? As for Verzenio, when do you expect to have the additional overall survival data for the remaining 50% of the population? Also, how challenging is it to identify higher-risk patients as we consider the initial commercial opportunity in adjuvant therapy?

Yes. Your question can be divided into two parts. First, regarding the importance of having additional cognitive benefit data for amyloid plaque drugs in the near term. I want to clarify that we already have strong clinical efficacy data for donanemab. The positive Phase II study, which demonstrated cognitive benefits, is unique and noteworthy, as published in the New England Journal. This information should be advantageous even before we get confirmatory data. Some physicians may prefer not to use a drug without cognitive data, which is understandable. However, for those who connect surrogate efficacy data with the Phase II results of donanemab, if they believe that reducing amyloid plaque is beneficial, they will likely favor the drug that has the most significant effect on amyloid plaque reduction. That's another exciting aspect of donanemab. Now, concerning the confirmatory studies, if multiple sponsors and drugs show clearly negative results, it would certainly be a setback and would prompt a reevaluation of our approach. However, I believe this scenario is very unlikely. More realistically, we might see a mixed outcome, where some drugs perform better than others, with some trials achieving significance while others do not. I have confidence in our trial, but we need to await the data. In that situation, those results could reinforce our belief that amyloid is a crucial surrogate and that lowering amyloid is beneficial.

Yes. Chris, thanks for the question. And as Dan shared, we are incredibly pleased with what we're seeing out of Verzenio and the monarchE data and as he shared, key endpoints, such as IDFS, have continued to strengthen with further follow-up. And now we have two years of median follow-up. So very pleased with that, and as he shared, we remain very confident there will be an OS trend favoring Verzenio in the broader population. So we would work with the FDA to expand our label to include these patients in the future. So obviously, this is event driven, and so the timing of this will be determined by the event rate. So our next planned analysis is in the second half of '22, and this analysis will help us really further inform the timing of that final analysis. So as you commented, the overall survival data in the broader population is still immature. We still have less than 50% of the events needed to do that final OS analysis. But with what we're seeing, and again, strong performance in both the high and low Ki-67, we remain confident to see this trend in OS favoring Verzenio replicate. As far as Ki-67, good news here is that this is really a familiar concept to physicians. It is already accepted as a prognostic factor in breast cancer. And it's really easily performed through an IHC assay, so very simple assay. And these are broadly available in the pathology labs. And the assay and the methodology that we used on monarchE is straightforward and proven to be accurate and really highly reproducible. So our belief is that oncologists will move to quickly adopt this in practice. And really, this clarity in patients with the highest risk, I think, will help to accelerate uptake in this setting. So we look forward to launching in this setting.

Hi, thanks so much for taking my question. Surprisingly, I also want to talk about Alzheimer's today. Dan, I have three subparts. First, are you expecting to use interim data from your ongoing Phase III as part of the regulatory filing or during the review? Secondly, once the plaque is cleared, and I think 60% of patients have clearance by 12 months, what rate of onset of new amyloid plaque do you expect subsequently? And I'm just trying to understand your expectation on finite duration of dosing versus Biogen's opinion on continued dosing. And then finally, I'm also trying to reconcile the slide you showed on the nonlinear model, the Conrado model, suggesting a relationship between plaque decrease and a slowing in clinical progression. Are you saying there's a relationship, or are you saying there's a causality? Because you might recall the New England Journal paper on your Phase II mentioned there was no association between plaque and clinical benefit at patient level one, Biogen data suggested similar. Thank you very much.

Okay. Three great questions, Umer. Thanks. So the first question, you asked if we'd use interim data, I commented that we'll take a safety cut of data in the right way to support that submission. We don't plan to support that submission or do we see the need to support that submission with any looks at efficacy data. We have adequate efficacy data supporting the plaque lowering, which would be the basis of submission and approval under accelerated approval. Your second question is once plaque clears, how long does it take to come back? We have some data on that that was also presented at AAIC. I didn't highlight it this morning. But what we found is that off therapy, there is very slow, negligible, really, regrowth of plaque. I think if you sort of extrapolate it out, it might take 14 or 15 years or something like that to regrow amyloid plaque. The average age of patients in this trial is 75. And remember, we haven't fully halted progression of disease. So that doesn't feel like a near-term thinking on redosing will be necessary to keep them clear. But we'll have the ability to follow patients for many years and confirm that. Finally, I think you've correctly summarized the situation, which is that in our initial analysis, we didn't see a correlation, and now we are reporting that we do see a correlation. Why is that? And of course, correlation can't prove causation, so it is just a correlation. So why do we see it now? I think what we learned was quite interesting. And that's that the amount of plaque you remove depends a lot on how much plaque you have to start with. So if you only have 50 centiloids of plaque, there's only so much you can remove. If you have 100 centiloids of plaque, you can remove a lot more. So that turns out to be a pretty important compound in these kinds of correlations. The people who have the more severe disease, perhaps longer duration, lower cognitive performance, older age, they might have more plaque at baseline; you can remove more but they still might be the worst progressors than people who have lower plaque and you remove less. So I think our thinking initially, and maybe the field's thinking, was a little bit backwards on this to look for a straight correlation between change and change without adjusting for all of those important baseline covariates.

A couple of questions please. Just going back to interim analysis for TRAILBLAZER-ALZ to not so much as used to support accelerated but to accelerate the readout for the full standard regulatory review, you're using a Bayesian disease progression model. Given that you're getting such a rapid clearance of Alzheimer's and that's linked to cognition in those patients who have that, do we have to assume that the follow-up is going to go out to the full 72 weeks? Or is there a possibility of early unblinding driven by efficacy in these patients? And then second, perhaps you could comment on the manufacturing capacity for donanemab as well as the regulatory outlook for your P-tau assay, assuming that you attain regulatory approval on the accelerators. Thank you.

Yes. So Andrew, you've asked a follow-up question here, an important one, on the potential even in the face of an accelerated review for accelerated approval, rather, for plaque lowering, whether we'd still be keen to get that kind of data a bit earlier by pulling forward an interim on TRAILBLAZER-2. We haven't ruled that out. We also don't have plans at this moment in time. I think we just need to see where we are and where the field is. But really, the maintaining a pristine Phase III trial would probably be a pretty high priority, particularly if accelerated approval gives the path for patients to have access to the medicine, then it becomes less urgent to get that data faster. So that's our current thinking. We've been working hard to make sure we have manufacturing capacity. I feel good about where we are to support launch and growth of donanemab and hopefully someday enter PG4 even to follow that. With respect to the commercialization of a P-tau diagnostic, there are different paths forward for an in vitro diagnostic, including a lab-developed test or LDT, which can be done in a centralized location, for example, under CLIA. And that's a pretty fast path, and that's one of the options that we consider.

Hi, thanks for taking my question. So first question I had for you is, how do you think about a potential outcome for the national coverage determination of monoclonal antibodies to treat Alzheimer's disease? And then second question is, how would you think about pricing donanemab if it is approved? Thank you.

Thanks for the questions on donanemab. The first one regarding the national coverage determination is indeed significant. It was noted that the initial FDA approval was quite broad, but subsequently, the FDA and the sponsor narrowed the focus on the patient population. There is potential for further refinement by the experts at CMS, possibly requiring evidence of Alzheimer's pathology, such as amyloid plaques or tau pathology. This aligns with our objectives and views. It may take about nine months to see how this will develop, and we will be involved in discussions to share our data and insights appropriately. Regarding pricing, it's still early to provide any comments, and we have some time to consider this.

Dan, you're popular today, so I just have a couple more for you. On donanemab, is there a hurdle that the FDA has provided in terms of the number of patients for safety, either for the filing or during the review? And then as the data from TRAILBLAZER matures, what is your estimate on what the duration of therapy benefit could ultimately be? And then real quick on tirzepatide, just wanted to know as you guys complete the filing. At this point, what's the gating factor as you look at the different geographies and you prepare?

Yes. Regarding the safety requirements for donanemab or any drug, it’s important to have sufficient exposure and duration across a diverse population to thoroughly evaluate the benefit-risk of the drug. This number varies based on the specific drug, the target population, and the length of therapy, along with details about the safety and efficacy data collected. It would certainly make it easier for sponsors and the FDA if there was a clear benchmark, but it does need to be individualized for each drug. Based on our current analyses and discussions, I believe we will meet these requirements comfortably by the end of this year. Your second question was about the duration of benefit as the TRAILBLAZER data develop. I mentioned that the duration of plaque lowering appears sustainable. However, Geoff, you are inquiring about the duration of cognitive benefit. We are observing an average slowing of cognitive decline among patients, although they are still experiencing decline. There may be questions about whether the lines of data are converging or diverging. Initially, there could have been a perception that the lines were parallel at later time points, but our additional statistical analyses, together with the raw data, indicate that the lines are indeed diverging over time. Therefore, I anticipate that the benefit of the slowed decline will persist going forward, but it is too early to provide definitive data on this.

Yes. Thanks for the question. Our Phase III SURPASS program for type 2 diabetes is done and completed. So the only gating factor here is how quickly we can summarize the data and submit to the regulators, which we plan to do by the end of the year to major global regulators.

Maybe I guess I'll try to take another stab at the pricing question. I appreciate it's early, but it is sort of the elephant in the room. And just maybe if Dave and team could comment just on the appropriateness of the pricing benchmarks in the space already today in Alzheimer's, as you think about it. And then obviously, 3Q has tripped in the past around Trulicity dynamics. So just hoping, if you could just offer a little bit more clarity there on as you think about pricing headwinds into 3Q specifically? Thank you.

Yes. I appreciate the question, and we totally get the curiosity. There's, as you understand, probably a lot of limitations of what we would say at this stage. One of the reasons for the limitation is really the ultimate label we have and the value we can demonstrate to customers is a key input at Lilly for pricing. Fortunately, we have the only study in the space that hit its pre-specified endpoint for disease reduction or disease progression reduction, and those are key. As we demonstrated at AAIC, we continue to cut that data. I think there was an earlier question about how we might differentiate in the NCD process, but that's one of them, as we have this completed study with exquisite biomarker profiles of the product and can continue to elucidate what donanemab does in the brains of Alzheimer's patients in ways that perhaps others could not. And those are inputs as well. Finally, Lilly has been a leader in value-based concepts and really partnerships to make sure that the appropriate patients can easily access at low out-of-pocket cost or medicines. And we're applying that thinking to this problem as well in the U.S. as well as outside. Our goal isn't just to get an approval but to make sure that all of the people millions in the U.S. who could qualify for it could access it on day one. So those are all inputs into that process and without throwing out a number here, which wouldn't be appropriate until we get an approval. That's how we think about it. Hopefully, that gives you some color behind the scenes.

Yes. Let me add a general comment on pricing movement throughout the year. We have had numerous discussions about this, and there tends to be some volatility. As patient flow through the healthcare system continues, we see a more pronounced impact from the coverage gap in the second and third quarters. This dynamic occurs consistently every year, as Mike mentioned. We have incorporated these assumptions into our full-year guidance regarding pricing dynamics. As we gain more insights, we will share further information. For now, our full-year estimate for U.S. pricing dynamics aligns with what we previously discussed about overall erosion. We observed a 3% decline for the first half of the year, which is what you should expect for the entire year.

Great. So really wanted to kind of focus in on abemaciclib in prostate cancer and the update that was provided, I think in an abstract published at AACR, some of the details were provided with regard to the sort of threshold for moving forward as it relates to the hazard ratio. And it cites a hazard ratio of 0.64, 80% power. So with a P-value of 0.1 to, I think, continue advancing into the next stage of CYCLONE 2. So just wanted to clarify if that information is consistent and a driving force for moving forward, that would seem like a robust piece of information to have as we head into that. And then as incremental to that, just wanted to get a sense of the magnitude of the opportunity that Lilly believes this would represent for Verzenio going forward. And if there are, let's say, RB, so the retinoblastoma-related requirements for enrollment or any other biomarker requirements that could limit the size of the patient population. And then just as a follow-up. In terms of the NCD determination, just wanted to clarify if the pricing of the initially priced product would have any impact on Lilly's ability to independently price its own product, and if that's part of the reason why Lilly has argued for the products being treated independently as part of the NCD rather than as a class.

Yes. You're asking, Seamus, very smartly about the criteria to expand the study from our Phase II study to a Phase III study. I don't think we want to get into the very precise details on what that was. But you're correct that it was a very robust threshold. So we're excited to see that happen. We take Phase III starts very seriously at Lilly. We don't want Phase III failures. So when we have studies like this one in any therapeutic area where we move from Phase II to Phase III without ever seeing the data, we set aggressive bars that data really have to match to move forward to Phase III. So you can expect that's what we did here. Anne for more detail.

Well, as Dan said, we were incredibly pleased with this outcome and the recommendation by the DMC, and we set a very aggressive threshold on this adaptive design, and we're impressed that it met that threshold. And I think it just continues to be another example of how Verzenio differentiates from the competition. So the Phase III is open. It's already enrolling patients. We anticipate the results of the analysis in 2024.