Lilly(Eli) & Company (LLY) — Q4 2020 Earnings Call Transcript

Good morning. Thank you for joining us for Eli Lilly and Company's Q4 2020 Earnings Call. I'm Kevin Hern, Vice President, Investor Relations. Joining me on today's call are Dave Ricks, Lilly's Chairman and CEO; Josh Smiley, Chief Financial Officer; Dr. Dan Skovronsky, Chief Scientific Officer; Anne White, President of Lilly Oncology; Ilya Yuffa, President of Lilly Biomedicines; and Mike Mason, President of Lilly Diabetes. We're also joined by Sara Smith and Lauren Durfy 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. The additional information concerning factors that could cause actual results to differ materially is contained in our latest Form 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, which exclude the financial contribution from Elanco during 2019, and present earnings per share as though the full disposition via the exchange offer was complete on January 1, 2019. Now, I'll turn the call over to Dave for a summary of our 2020 results.

Thanks, Kevin. On our guidance call in December 2019, we provided a framework for how we are thinking about the 2020 to 2025 period, noting our expectations to continue to deliver top-tier revenue growth and operating margin growing into the mid-to high 30s while continuing to increase R&D productivity. As we met with investors throughout 2020, it was evident that while there was good insight into management's expectations for the next few years, investors were increasingly focused on our ability to grow in the second-half of this coming decade. In the past two months, we have begun to deliver answers to that question, with positive data for LOXO-305, tirzepatide and donanemab, each with a chance to significantly improve patient outcomes in areas of high unmet medical need. We believe these are three of the most important and exciting pipeline assets in our industry and provide meaningful support for Lilly's growth potential beyond 2025. Together with Verzenio data in early breast cancer last summer, we have significantly reinforced our growth prospects for the midterm and upgraded them for the long term. While these readouts drive incredible momentum for our future, I am also pleased with the way that we delivered in a complex and challenging 2020. Our 2020 revenue enabled us to exceed our midterm revenue goal of 7% CAGR from the years 2015 to 2020. Turning to the quarter. Revenue grew an impressive 22% versus Q4 2019, or 20% in constant currency. This strong performance was driven entirely by volume growth of 24 percentage points despite continued pricing headwinds and demand pressure from the effects of COVID-19. Excluding bamlanivimab, revenue grew over 7% for the quarter. Key growth products continue to drive this volume and our revenue, now representing 55% of our base business. We continue to advance our productivity agenda in Q4 as the combination of strong revenue growth and modest operating expense growth drove significant margin expansion. Our non-GAAP operating margin was approximately 33% with and without COVID-19 therapies, an improvement of roughly 650 basis points versus Q4 2019. With continued margin expansion for the quarter and full-year, excluding the impacts of COVID-19 on our business in 2020, we would have achieved our midterm operating margin as a percent of revenue goal of 31%. We're proud to have delivered nearly 1,000 basis points of operating margin expansion since 2016. In addition to the strong business performance, we achieved multiple pipeline milestones since our Q3 earnings call. These include the positive results I noted for LOXO-305, tirzepatide and donanemab; the FDA granting emergency use authorizations for bamlanivimab and baricitinib to help patients with COVID-19; the submission of empagliflozin for heart failure in people with reduced injection fraction in the U.S., Europe and Japan, this is done in collaboration with Boehringer Ingelheim; and the submission of Verzenio in early breast cancer in the U.S. During Q4, we put our growing operating cash flow to work, announcing a 5% increase in the dividend for the third consecutive year, as well as continuing to pursue external innovation to augment future growth prospects with the acquisition of Prevail Therapeutics. This acquisition adds a promising new modality for Lilly by creating a gene therapy program that will be anchored by Prevail's portfolio of clinical-stage and late-preclinical-stage gene therapies across Alzheimer's, Parkinson's, dementia, ALS, and other neurodegenerative disorders.

Thanks, Dave, and good morning. Slide 7 summarizes our non-GAAP financial performance in Q4 and 2020. As Dave mentioned, revenue increased 22% this quarter compared to Q4 2019 and increased 7%, excluding bamlanivimab sales. Gross margin as a percent of revenue declined 130 basis points to 78.6%. Excluding the impact of bamlanivimab revenue and the related manufacturing costs, gross margin as a percent of revenue was 79.9%, in line with Q4 2019 performance. Moving down the P&L. Operating expenses grew 3% compared to the same quarter last year. Marketing, selling, and administrative expenses were down 8% as reduced activity due to COVID-19 and productivity measures offset investments in key growth products. R&D expenses increased 16%, driven by investment in COVID-19 therapies. Net of COVID-19 expenses, baseline R&D was relatively flat, and total operating expenses decreased over 5% compared to Q4 2019. Operating income increased 53% compared to Q4 2019 as revenue growth far outpaced expense growth, resulting in operating income as a percent of revenue of 33% for the quarter. Excluding the impact of COVID-19 therapies, operating income grew 34% in the quarter and the operating margin for our base business was 32.7% for Q4. Other income and expense was income of $477 million this quarter compared to income of $206 million in Q4 2019, driven by investment gains on public equities. As we noted in our Q3 earnings call, beginning in 2021, we will exclude the gains or losses due to equity investments from our non-GAAP measures. We have posted a supplemental investor workbook for Q4 on that basis to enable you to have an apples-to-apples comparison as we move into 2021 and compare to 2020 non-GAAP performance. Our tax rate was 14.4%, an increase of 180 basis points compared with the same quarter last year, driven primarily by net discrete tax items in both quarters. At the bottom line, net income increased 58%, while earnings per share increased 59%. Net of COVID-19 therapies, net income and earnings per share increased 43%. Moving to Slide 8. You can see these same non-GAAP measures for the full-year. In spite of the ongoing demand impact from the pandemic, we grew the top line at 10% or 6%, excluding bamlanivimab. Excluding COVID-19 therapies, our operating margin expanded by 300 basis points contributing to 30% EPS growth while continuing to invest behind our newer products and pipeline.

Thanks, Josh. We had an exciting start to 2021 as we read out positive results for gentimab in the Phase 2 Trailblazer Al study. Lilly has spent more than 30 years dedicated to finding solutions for Alzheimer's disease, and we are proud of our progress in advancing the science and providing hope for patients and their families suffering from this devastating disease. On Slide 14, you can see our key takeaways from this exciting trial. We are encouraged by the strong efficacy results, where in a relatively small study, we overcame the scale of the study with precision on patient enrollment and a very potent and effective plaque-clearing drug, becoming the first-ever disease modification study to hit statistical significance on its primary endpoint, with a slowing of decline by 32% relative to placebo as measured by the integrated Alzheimer's Disease Rating Scale. ADAS is a clinical composite tool, combining two well-accepted measures in Alzheimer's disease: ADAS-Cog 13 for cognition; and ADCS IADL, Instrumental Activities of Data Living, for function. While the study was not powered for assessing multiple endpoints, we're very encouraged by the consistent improvements observed on all pre-specified secondary endpoints for cognition and function compared to placebo though ganitumab did not reach statistical significance on every secondary endpoint. The consistency across time points and across statistical methods was very encouraging, particularly the disease progression model, which is becoming more accepted by the scientific community. In addition, we saw rapid and deep amyloid plaque clearance for ganitumab-treated patients who, on average, showed an 84 centiloid reduction of amyloid plaque at 76 weeks compared to a baseline of 108 centiloids. Since below 25 centiloids is a negative amyloid scan, this means that the average ganitumab-treated patient had a negative scan by the end of the study. Finally, the safety profile was consistent with observations from Phase 1. Amyloid-Related Imaging Abnormalities or ARIA were observed, which is consistent with plaque-clearing antibodies. In the ganitumab treatment group, ARIA E occurred in 27% of treated participants with an overall incidence of 6% of patients experiencing symptomatic ARIA E.

Well, thanks, Dan. Before we go to Q&A, let me briefly sum up the progress we made in 2020. 2020 was a remarkable year as Lilly worked to fulfill its purpose in new and important ways. In addition to many contributions in the fight against the global pandemic, our business grew 10% in 2020 driven by strong volume growth from our key growth products launched since 2014. These products now account for more than half of our revenue for the first time. We continued our productivity journey, delivering nearly 300 basis points of operating margin expansion for our base business. We made significant progress on our innovation-based strategy, with LOXO-305, tirzepatide and Verzenio early breast cancer readouts delivering potential category-changing data. While January's donanemab top line success was a first in Alzheimer's. With EUAs for bamlanivimab and Olumiant to combat COVID-19 and bolt-on acquisitions of Dermira and Prevail bookending the year, the past 12 months have been an exceptional example of Lilly's success in leveraging internal and external innovation. We returned nearly $3.2 billion to shareholders via the dividend and share repurchase, and we will have another meaningful dividend increase, which we announced in December, reflecting significant confidence in the ongoing strength of our business. All of this was accomplished against the headwind of a pandemic that is still raging. While the New Year does not free us from that near-term challenge, our long-term outlook has never been stronger.

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. Tony, please provide the instructions for the Q&A session and then we're ready for the first caller.

Okay, thanks. Good morning, everyone. Thanks for the question. Just have a couple. Dan, on donanemab, I know you're planning on having regulatory discussions, but beyond expanding TRAILBLAZER-ALZ 2, is it reasonable to start a third study just to expand the safety database in the treatment experience? And then, Josh, you mentioned you're still seeing commercial impact from COVID. What would you say are the franchises that were mostly affected in 2020? And maybe just review your assumptions for normalization of some of those in 2021? Thank you.

Thanks, Geoff. Your question is whether we'd consider starting a third donanemab study to improve the safety database. No, we haven't considered that at present. Of course, as we said, our next step is to discuss the dataset we have with regulators. I think if we determine that we need more patients, the place to do that is the TRAILBLAZER-2 study. With respect to additional studies, I think there could be opportunities to explore other populations, and we're working through those possibilities right now, but we don't see that as necessary for this current population.

Thanks, Geoff. I think when we look across our therapeutic areas, it's relatively consistent at this point that we're still not quite back to new prescriptions in key areas, like immunology, pain, and diabetes, although the diabetes numbers are looking stronger as we get here into January, but I think we're still seeing some suppressed demand. Physicians have, I think, in most markets, figured out how to see patients safely, and we're seeing nothing like what we saw back in April and May in the U.S. So, I think we just have to be cautious as we get into the first quarter and realize that the more complex treatments have some higher degree of variability against them, and that includes starting new patients in areas like migraine. I think maintenance has been good throughout the pandemic. So, we do expect it, as we get through the first half of this year, we'll see a return to fully normal levels. But I think it's fair to assume that in the first quarter, we'll still be, in many of the therapeutic areas, a little bit below pre-COVID baselines in terms of new prescription starts.

Probably worth mentioning, Geoff, as well as dermatology, because I think that one also, you see some suppression in new patient starts, not affecting our share anything but just the overall volume in the category.

Thank you. A couple on donanemab. Investors are naturally wondering what the odds are that you could file for approval based on this first Phase 2 trial. To me, it seems highly unlikely given the size of the trial and the different subgroups, but just wondering, if you can share your latest thoughts on that? And then, Dan, your view of the need to continue to knock down what is sometimes described in the industry as toxic oligomers, which is what you might achieve by giving monoclonal chronically well after patients have already seen plaque normalization?

Tim, so your first question is on the possibility of approval from a single study in Alzheimer's disease. Look, we - in general, we don't disclose our back and forth with FDA or other regulators. In this case, of course, we said that next steps are discussions with regulators. So clearly, that hasn't happened yet. Still, we understand the regulatory threshold traditionally has been adequate and well controlled trial that means more than one trial. In this case, we have a single adequate and well controlled trial in Alzheimer's disease. That, as I said, has not been the standard in this area. Of course, in other disease areas, notably oncology, drugs can be approved from a single trial, usually, that's an accelerated approval. Usually, it's a group that's well defined by pathologic characteristics and biomarkers. Usually, there's dramatic pathologic response as well as clinical outcomes. Of course, that's also true in the case of this donanemab trial, but again, oncology is quite different than Alzheimer's disease. Your second question was on the question of toxic oligomers. It's long been unclear what is the toxic species of amyloid beta: is it monomers? is it oligomers? Is it plaques? This antibody was designed to be exquisitely specific for amyloid plaques. We don't think it binds oligomers. And thereby, the efficacy – therefore, the efficacy we see here seems to imply that it's the plaques that are the toxic species rather than the like oligomers. However, we can't rule out that these two things are in equilibrium with each other, and perhaps by clearing plaques, you remove oligomers as well.

Hi, thanks so much for taking my questions. Dan, in the prior studies, the IADRS DAS endpoint, the composite they use in donanemab, it didn't correlate very well with CDR sum of the boxes. But I did find it interesting that the most recent EXPEDITION3 trial for solanezumab did have high concordance between this new composite versus CDR sum of the boxes. I guess what I'm wondering is how should we be thinking about whether IADRS versus CDR sum of the boxes correlate closely or not? Or is it more a function of the more recent trials where CDR does, in fact, correlate very closely with IADRS? I'm thinking about that heading into your donanemab trial. The other one I had is, you have this tau antibody Phase 3 coming up this year. Maybe if you could remind us how's this tau-mab similar or different than some of the other ones because the progress on this target has not been quite good to date. And I saw your trial was pushed out a little bit as well, but it'd be very helpful to have any color? Thank you.

Great, thanks. Umer, for your first question on IADRS and its correlation with CDR. Look, when we think about an endpoint for any clinical trial, there's really two things that make an endpoint a good endpoint. One is the statistical validation behind it. So in other words, is it reliable across different patients across different time points across different trials? We put together a lot of data that support that. Second, is it meaningful for patients? And in this case, of course, we believe that that's inherently true. This is a composite of two things that are widely used, both thought to be important and meaningful: ADAS-Cog and activities of daily living, obviously, activities of daily living inherently meaningful for patients. Now, why do we pick IADRS versus CDR to be the primary outcome of the study? That should be obvious; it's because we believe that IADRS would be more sensitive for measuring decline and therefore more sensitive for measuring a drug effect. That's based on all of that statistical validation data that we did. If ADAS and CDR sum of the boxes were perfectly well correlated, then ADAS couldn't be better, couldn't be more powerful. And yet, I'm telling you that our assumption going into this trial was that it would be. So of course, different outcomes will have some correlation, but they won't be perfectly correlated. Based on what we saw in this trial, I think we haven't changed our thinking on outcomes, and we still think ADAS is a very valid and important outcome for Alzheimer's trials. Of course, that's a discussion to be had with regulators in the scientific community. With the zagotenemab, this is our anti-tau antibody. Just as donanemab was a different type of anti-amyloid beta based on its specificity for plaque, zagotenemab is a different kind of anti-tau antibody. It's highly specific for aggregated tau. Now, we think that's particularly important in the case of tau because there's a lot of soluble monomeric tau, and tau antibodies, like any other antibody, not much of it gets in the brain. So if you have a lot of monomer and a little bit of antibody, it could soak up all of your antibody and not have left to go after what we think is the more important species, aggregated tau. So, we'll have to wait and see. Of course, this is a field that is younger than anti-amyloid therapies, but we've taken a lot of things we learned from anti-amyloid and applied them to anti-tau, and we're quite looking forward to getting that data later this year.

Two questions. Investor expectations are quite high for donanemab in terms of sales potential. Lilly knows the full data for TRAILBLAZER and also the largely failed A-beta antibody landscape better than any other company. Based on what you've said, including that donanemab will be a driver in 25 to 30, it sounds as though you are fully comfortable with these multibillion-dollar expectations. Is that the conclusion you want to leave us with?

Yes, Steve, I mean, we don't comment on analyst models or forecasts, and we never would. So I can't directly answer your question. I guess what I can say is, we've invested in Alzheimer's for 30-plus years and spent a lot of money, as you point out, mostly failing, because there's a huge unmet medical need, and we believe that investment is justified based on the size of the market. But we're not able to say today, donanemab is the answer, has a path to market, etc. We're looking forward to the ADPD presentation coming up. The field will survey that data and make your own conclusions. And we need to talk to the FDA in a formal way about the path forward, and then we'll get to sales forecast later, but it's just not possible to answer a question like the one you asked.

Thank you, Terence, for your question. We're very enthusiastic about the data from both CLL and CL. We are currently in discussions with the FDA regarding the possibility of accelerated approval. It's important to note that single-arm accelerated approvals for hematologic malignancies can be complex, which will necessitate further conversations with regulators. Therefore, we cannot yet commit to specifics about submissions, timing, or indications, but we will continue these discussions. We are truly excited about the data, as Dan noted. This molecule initially targeted C41 mutations, but we have seen remarkable performance in broader populations. We will maintain these conversations and keep you updated.

Great. Just one on donanemab and then just one other one. What are your thoughts on the high-tau population? I guess just based on what you saw from TRAILBLAZER, is there a strong rationale the drug could also work in some of these patients? And maybe just give us a sense of what percent of patients in TRAILBLAZER 2 we should expect to come from that group?

Thank you for the question about the high-tau population. A significant aspect of the TRAILBLAZER trial is that we did not include patients with excessive tau in their brain. We believe that individuals with the highest levels of tau or widespread tau in their brain have likely surpassed the threshold for effective amyloid-directed therapy. As we analyze the data, which we hope to present soon, we will gain insights into how baseline tau levels may predict treatment response. This information could influence our enthusiasm regarding the inclusion of high-tau patients in the TRAILBLAZER 2 study. The percentage of affected patients varies based on the group considered. If we include all early Alzheimer's patients, mild Alzheimer's, and those with mild cognitive impairment, that's around 4.5 million in the U.S. and double that in Europe and Japan combined. Numerous patients in this group are amyloid negative, as we have previously indicated, meaning approximately one-third can be excluded. A small portion will be amyloid positive but have no tau, which we also did not include. A somewhat larger fraction falls into the high-tau category. After excluding those patients, it's estimated that about 30% to 45% of the mild Alzheimer's to MCI population would satisfy the enrollment criteria for TRAILBLAZER 1.

Chris, on 340B, what we said is if you look over the last 10 years, the 340B segment has been among the fastest-growing, certainly across the industry but for our business as well, and it rivals the size of Medicaid in our U.S. business, so about 10% of the business now. Of course, the change we made was to go back to the legislated intent and to provide the discounts to the actual hospitals that provide care and to exclude contract pharmacy that have grown over time. And when we look at that business, it's probably about half of the businesses in these contracted pharmacy. So that's where we've made the change, to not provide the pricing there. Now, we do have a process where those contracted pharmacy can apply and we've said for insolence, as long as they can demonstrate that they're passing on the entire pricing that they're still eligible to participate. When we look at all that together, we knew that that's where we implemented in September. We knew there would be challenges, and we're seeing those challenges come, but I think in terms of patient impacts, we haven't seen much yet. So, we're seeing the fact that the discounts are being provided as per our change, but we don't think it's impacting patient care at this point. So everything we're seeing so far is consistent with the decision we made. We knew it would be a difficult decision to implement. We knew there would be some customer concerns, we knew there'd be legal challenges, but I think what we saw in the fourth quarter is consistent with the guidance we've given for 2021, which is that we would expect this portion of the 340B program to moderate in growth and provide a two to three point price tailwind for us in 2021. So I think we're on track for that at this point.

So I just wanted to follow up on the one donanemab question, which is, Dan, can you just give us a little bit of color on the magnitude of blinding in this study, given the fact that patients obviously had almost absolute clearing of their amyloid plaques? And your confidence that the behavior of the placebo arm was consistent with the benefits of monitoring both tau and amyloid? And then the second question, just wanted to get a bit of an update on Verzenio and how Verzenio is tracking relative to your expectations.

Thank you, Seamus, for your insightful questions regarding the blinding of the study. As stated previously, we conducted TRAILBLAZER 1 with the high standards we typically apply to regulatory pivotal studies, which included a fully blinded, double-blinded design. You mentioned the possibility of unintentional unblinding due to amyloid clearance, but we don’t believe that’s feasible. Patients do not directly perceive the clearance of amyloid in their brains, and investigators would not be aware of the PET scan results. Another potential source of unblinding in Alzheimer's studies is ARIA E, as it can sometimes lead to side effects. We noted that around 6% of patients experienced symptomatic ARIA E, which is certainly an important aspect to analyze. In every Alzheimer's study, it is crucial to assess the drug impact on patients with and without ARIA, and we aim to have this analysis completed and ready to share in March. Finally, regarding the behavior of the placebo arm, that’s an excellent question. When we reviewed the data, my immediate focus was on this aspect. Occasionally, smaller studies may appear promising because the placebo arm performs worse than expected. However, we have expressed a high level of confidence and enthusiasm based on the data, which indicates that this placebo arm is not acting unusually. This does not account for the observed effect. We’ll delve deeper into this in March, but we are very pleased with how the placebo arm has performed.

Yes. As you noted, 2020 was another year of positive and clinically significant data for Verzenio. Consequently, our shares have continued to improve. Although no CDK is yet approved in the EVC setting, we believe this readout has particularly drawn attention to the market class. We are really confident that the current trends reflect our strong execution and focus on NBC, especially with the significant overall survival data, which a key competitor in the space does not possess. Comparing Q4 2019 to Q4 2020, we experienced remarkable growth for Verzenio. We saw a 6% increase in total prescription volume, 57% revenue growth globally, and 36% growth in the U.S. Thought leaders are repeatedly telling us that they see Verzenio as a differentiated agent. We've shared that in MEMS, we believe our CDK4 activity is higher than others, along with differentiated continuous dosing, a monotherapy indication, and the data from MONARCH 2 regarding the primary endocrine-resistant population. All these factors contribute to the growth we are seeing in the NBC market.

Thanks, Anne. Seamus, thanks for your question. Next caller please.

A couple of questions. First, on donanemab. On the issue of whether potentially donanemab could be submitted on the back of the Phase 2 data set that you have. Could you talk to your level of preparedness, particularly manufacturing, given the anticipated demand, but also to the P tau blood test that will potentially be used to define the patients? I'm just trying to understand where you're at because there is a scenario by which you could be on the market sooner than perhaps that many may believe. And that's an open observation rather than reflecting any particularly personal view.

Andrew, you raised an excellent question on all the work that needs to be done to prepare for ganitumab. We actually initiated manufacturing preparedness before we had this data. So, this is something we do at Lilly. We always prepare for success. And so at the same time, that our manufacturing colleagues we're ramping up the anti-COVID antibodies, we asked them to also prepare for donanemab. So that is well underway, and I like what's going on, the progress we're making in manufacturing there. The diagnostic ecosystem also needs work. We started that both ramping up imaging and proceeding along with opportunities to bring the phospho-TAL blood test to more patients. That's both underway, again, started before we had this data and then, of course, is useful whether we're successful in coming to market or whether any other anti-amyloid drug comes to market. So, those are preparations that we're certainly taking.

Thank you for the question regarding 305. It's important to understand that our work in the MSL population involves multiple aspects across our portfolio, not just 305. We have other assets in development that could be relevant to this area. Therefore, there's no need to overemphasize this point as it’s part of our routine operations to ensure comprehensive coverage of our portfolio. However, we do receive numerous inquiries about 305 and the anticipation from treating physicians. Additionally, we have several large Phase 3 trials set to begin. Our field force plays a crucial role in identifying high-quality sites for these programs. Regarding CLL, the narrative is very promising, and we believe we have a strong contender in this space.

Thanks, Anne. Andrew, thanks for your questions. Next caller please.

On the COVID antibodies, Dave, I realize the latest data is quite fresh. But to the extent you're worried that there could be a disconnect between the power of the data and the speed of uptake, is there anything that Lilly plans to do proactively to help move this along? I was intrigued by your comments on the prior call about some well-known hospitals being slower than other hospitals to get trained and ready. And I think that was even from before this latest data set. And then maybe for Josh, on that call, it sounded like the upper limit of $2 billion in your guidance is not necessarily set in stone if demand picks up. What's the practical limitation from a manufacturing standpoint as it relates to 2021 for the antibodies? Thanks.

We've been collaborating with healthcare systems and physicians nationwide to promote the use of the antibodies. This situation showcases the outcomes when there isn't a typical commercial approach to deployment, as this was facilitated through government channels under an EUA. We've noticed significant differences in adoption rates, and it appears there is an inverse correlation between advanced medical facilities and actual usage. A major barrier is the confidence in the data. I'm pleased with the recent data we shared, which should bolster that confidence. Additionally, there is a broader class effect at play, with other antibodies showing potential in various situations, further supporting our findings and building trust. We also anticipate changes from two key endorsements: one from the NIH and the other from the Infectious Disease Society. However, there are several practical challenges to scaling up treatments for COVID-19, which have been navigated by some health systems but remain unresolved by many others. We’re collaborating extensively with state health departments to disseminate best practices, which include flagging charts for at-risk patients who test positive for COVID-19, scheduling appointments, and establishing dedicated facilities and teams for efficient antibody administration. In locations where these practices are in place, we’ve observed an uptick in usage and a potential decrease in healthcare utilization. We are conducting an experiment in New Mexico to validate this, alongside our efforts with UnitedHealthcare. I'm hopeful that we will witness an increase in usage rates, which rose from about 20% in December to approximately double that now, and we anticipate further progress moving forward.

Thank you, Greg. For this year, we have provided a guidance range for COVID antibodies of $1 billion to $2 billion, though there is considerable uncertainty. Based on our announcement today regarding the new agreement with the U.S. government, we have a strong likelihood of reaching $1 billion in the first quarter from existing purchase agreements. Additionally, we anticipate about 500,000 doses of bamlanivimab available in the first half of the year for monotherapy, along with one million combo doses. If we sell all of that, primarily in the U.S. or high-income countries, it could exceed $2 billion. However, we currently do not have an emergency use authorization for the combo from the U.S. government or any other, and there remains uncertainty regarding vaccines and the usage of these products. Looking into the second half of the year, we expect to maintain a production pace of over a million doses per quarter, and manufacturing shouldn't pose a significant challenge if we continue on our current path with vaccines. Thus, it's possible we could exceed $2 billion for the year, but uncertainties remain, and we will monitor this and provide updates as more agreements and approvals come in.

Congratulations on nice result, and I will continue on the tradition of asking one question about donanemab and one other. So on donanemab, I guess the bar that you can see from the FDA is one kind of well internally correlated trial and data support elsewhere. When you look at your program, it looks like the support you could provide is from removal of plaque from earlier trials. And I guess the question for you is, is the understanding that plaque removal is related to clinical benefits solid enough that you can use that as early approval based elsewhere?

Yes, Ronny. It's a really good question that you're asking, which is how strong is the biomarker evidence here? And how can that be used to support regulatory decisions? I certainly think that's the direction the field is going. So in other words, when I look across all of the trials that have read out, again, this is the first plaque removing antibody that has had a positive study that hit its primary endpoint on the pre-specified statistical analysis. But even the other ones that didn't do that, when you look across the totality of evidence, you get a sense that there's a correlation between plaque removal in general and better cognitive performance. Here, we have probably the most significant plaque removal, the fastest, deepest plaque removal and a very meaningful clinical effect. So it's another point in this sort of dose response curve comparing amyloid plaque removal and cognitive changes across different trials.

As I mentioned earlier, we anticipate a two- to three-point price advantage in 2021 due to the changes we made regarding the 340B program. We have clearly stated that the purpose of these changes is to provide discounts that patients currently do not receive. We believe that the program needs reform and improvement. Therefore, we expect changes to occur in 2021. Additionally, we have mechanisms in place for contract pharmacies to access these discounts. Our goal is to ensure that patients receive the pass-through discount on their insulin. In our guidance for 2021, we are assuming that there will not be a complete financial benefit for the entire year, as we expect modifications throughout the year, either to help patients receive the discounts or due to some administrative or legislative changes.

So this concludes the prepared remarks, and I'll turn the call over to Kevin for the Q&A.

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. Tony, please provide the instructions for the Q&A session and then we're ready for the first caller.