Eli Lilly and Company (LLY) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by. Welcome to the COVID-19 neutralizing antibody program update media and investor call. [Operator Instructions] As a reminder, this conference call is being recorded. [Operator Instructions] And I will now like to turn the conference over to our host, Kevin Hern. Please go ahead, sir.

Good afternoon. Apologies for the delay and thank you for joining us for Eli Lilly and companies SARS-coronavirus-2 neutralizing antibody program update. I'm Kevin Hern, Vice President of Investor Relations. Joining me on today's call are Dave Ricks, Lilly's Chairman and CEO; Dr. Dan Skovronsky, Chief Scientific Officer; and Josh Smiley, Chief Financial Officer. 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 and media. It is not intended to be promotional and is not sufficient for prescribing decisions. I'll now turn the call over to Dave to provide some introductory comments.

Thanks, Kevin, and thanks to everyone for joining us today. The last 2 months have certainly been an exciting time for Lilly, with a string of high-quality pipeline readouts, which increase confidence in our long-term outlook. However, in the near term, here in the U.S. and around the globe, we remain focused on navigating the challenges of the COVID-19 pandemic. As this morning's news highlights, Lilly remains committed to bring the full force of our scientific and medical expertise to bear on the virus. And to benefit the global community. And we're delighted to have multiple positive readouts in our BLAZE clinical program, reinforcing the potential effectiveness of our neutralizing antibodies in combating COVID-19. As you can see on Slide 4, it has been an extraordinary journey from the start of our efforts in March, not even 1 year ago, to the impressive progress in the ensuing months, including our exciting news in the past several days. I am proud of the tireless efforts from all of my Lilly colleagues around the world who've delivered these outcomes. And I'm thankful for the contributions from those who have partnered along the way with us, and these partners include our partners at AbCellera and Junshi, the National Institute of Allergy and Infectious Disease in the NIH, the FDA and other global regulatory bodies and the investigators who helped us develop the data for the SARS-CoV-2 neutralizing antibody program and especially to the patients who've enrolled in the BLAZE clinical program. I am greatly encouraged by the partnership across our industry as we now move at unprecedented speed to answer the call to deliver solutions for COVID-19. The tireless dedication of regulators and health care workers continues to be encouraging as we work together to address the urgent challenges before us. Although at the moment, it may appear that the rate of scientific development is now exceeding the ability of governments and health care systems to absorb new information and implement new protocols and practices. But Lilly remains focused and committed to partner with everyone involved to solve these challenges. We have now tested our neutralizing antibodies in more than 4,000 patients, delivering significant evidence to reinforce our belief that they can be an important treatment for both the prevention and the treatment of COVID-19. As Phase III data from BLAZE-1 and BLAZE-2 show that early treatment with neutralizing antibody therapy results in a meaningful decrease in death from COVID with 0 COVID-19-related deaths in the antibody treatment arms. Since the FDA granted emergency use authorization for bamlanivimab for high-risk ambulatory COVID-19 in November, roughly 125,000 patients have been treated with bamlanivimab in the United States. Bamlanivimab remains widely available for patients throughout the United States with adequate supply at most hospitals. It is now authorized for use in several additional countries here in North America, across Europe and in the Middle East. Adding to the growing body of evidence for treatment with bamlanivimab, last Thursday, we shared exciting top line data from the BLAZE-2 COVID-19 prevention study, a first of its kind Phase III study, which showed that bamlanivimab prevented COVID-19 at nursing homes, reducing the risk by up to 80% for nursing home residents, some of the most vulnerable members of our society. We're looking forward to discussing these data with regulators to explore expanding the EUA for bamlanivimab to prevent the spread of COVID-19 in nursing home facilities. We had also submitted a request for EUA to the FDA in November based on the interim data from Phase 2 of the BLAZE-1 trial, which demonstrated that bamlanivimab and etesevimab together significantly reduced viral load, improved symptoms and reduced hospitalizations in ambulatory patients with mild to moderate COVID-19. This file remains under review by the FDA. This morning, we shared new data from the BLAZE-1 study, which further supports the request for EUA for the joint administration of our neutralizing antibodies and showed that bamlanivimab and etesevimab together reduced the risk of COVID-19 hospitalizations and death by 70% in high-risk patients and also significantly reduced viral load and accelerated symptom resolution. These data support our belief that bamlanivimab and etesevimab together can be an important treatment with significantly reduced hospitalizations and death in high-risk COVID-19 patients. Combination use may also allow efficacy against a broader range of variant strains of the virus. We also noted positive top line results from the BLAZE-4 study, focused on evaluating additional combinations and doses -- dose forms for neutralizing antibodies. Efforts with lower dose -- efforts with lower doses represent critical opportunities, which may extend the supply of neutralizing antibodies and offering the potential for broader patient impact. Lilly has a robust clinical supply chain in place to produce our neutralizing antibodies, and supply is expected to substantially increase in 2021. We are accelerating the manufacture of etesevimab in collaboration with Amgen and shared this morning that we expect to be able to supply 250,000 doses of etesevimab for administration with bamlanivimab in the first quarter and up to 1 million doses of the 2 antibodies for administration by mid-2021 for use around the world and await authorizations from governments so that we can make this available to patients. For bamlanivimab administered alone, we will have shipped approximately 1 million doses by the end of this month, and we'll have over 1 million additional doses available through mid-2021. Our goal is 0 cost to patients, wherever possible, for any patient receiving Lilly neutralizing antibody treatment. And we are working with regulators to update preparation and administration instructions in response to feedback from frontline nurses and doctors to enable infusion times to be reduced considerably to potentially less than 20 minutes and no longer than 60 in an effort to reduce the burden on the health care system. While we are excited about our meaningful progress, the surge in cases in the U.S. and abroad reinforces the significant need for increased utilization of authorized treatments and vaccines to combat the COVID-19 pandemic. With that in mind, I'll turn the call over to Dan to provide a more detailed update on the continued progress of our SARS-CoV-2 neutralizing antibody program.

Thank you, Dave. I'm excited to share our very recent data with all of you. But before I do that, let me start with an overview of the 2 antibodies, which we'll be discussing today. On Slide 5, you can see the 2 antibodies, bamlanivimab on the left, which was developed in partnership with our Canadian biotechnology partner, AbCellera, as well as in partnership with the NIH, the vaccine research center at NIAID. On the right is etesevimab, that was developed in partnership with our partner in China, Junshi Biosciences and the Chinese National Academy of Sciences at IMCAS. Both of these are fully human IgG1s, both bind the receptor binding domain of the spike protein of SARS-CoV-2, the virus that causes COVID-19. Each of these antibodies on its own is a very potent and fully neutralizing antibody against virus. Our theory here is that combining the 2 could allow efficacy against a broader range of naturally occurring SARS-CoV-2 variants as these new variants arise and spread around the world. On Slide 6, you can see the overview of our clinical program for our neutralizing monoclonal antibodies. On the left side is the ambulatory program. This is early treatment in recently diagnosed patients. And on the right, you can see our single Phase III trial in prophylaxis in nursing homes. The new data that we top lined today is from the BLAZE-1 treatment study. But just last week, we top lined data on prophylaxis in BLAZE-2. So let me start with the BLAZE-2 prophylaxis data. On Slide 7, you can see the design of this study. This was a remarkable clinical trial where when nursing homes were beginning to have an outbreak of COVID-19, we deployed our research staff to those sites through the use of mobile research units. We set up infusion centers and enrolled patients and staff into the clinical trial, where they were randomized to a single dose of bamlanivimab at 4,200 milligrams or placebo. We enrolled patients, residents and staff regardless of their baseline SARS-CoV-2 status. So some individuals that we enrolled in this trial were actually positive at baseline unbeknownst to the individuals. And they were put into a treatment cohort. The majority of patients were negative, and that is our prevention cohort. On Slide 8, you can see the top line results from this study. The Kaplan-Meier curve on the left shows a early and dramatic separation between placebo and residents that were prophylaxed with bamlanivimab 4,200 milligrams. The lines separate early and the separation continues to grow throughout the course of the study and offering protection for the full 2 months that we evaluated patients here. The odds ratio for protection of residents in any given facility was 0.20, which equates to up to an 80% reduction of risk with a p-value of 0.00026. So we were pleased to achieve this endpoint with protection of residents of the prevention population. We were also pleased to have had an impact on death due to COVID-19 in this population. You can see this highly vulnerable population, among residents who were randomized to placebo, there were 4 deaths due to COVID-19 versus 0 death amongst the 160 residents who were randomized to bamlanivimab. On Slide 9, you can see the death due to any cause results in the small treatment cohort in this overall prevention study. So these were the individuals who were positive at baseline and there were 24 such individuals in placebo and 17 in bamlanivimab. This is among the residents. And you can see there were 4 deaths on placebo versus 0, again, death to any cause on bamlanivimab. So we are excited about these results and going into the Phase III treatment trial, we had high hopes for efficacy. So Slide 10 just takes us back, we're going to now focus on the BLAZE-1 treatment trial. On Slide 11, you can see the design of this complicated trial, on the left side is the Phase II portion of the trial. We had multiple cohorts here, studying different doses of monotherapy as well as combination therapy of bamlanivimab plus etesevimab. Those results, in an interest in getting data out to the scientific and investor communities as quickly as possible, those were webcast October 7 and then published in peer-reviewed journals in the New England Journal and JAMA subsequent to that. Today, we're disclosing the data from the first Phase III portion of this trial. These are completely separate patients. It's a different cohort with different enrollment criteria and a different primary endpoint. Here, we're talking about over 1,000 patients. They all need a definition of high-risk either by virtue of age or obesity or other comorbidities, diabetes or high blood pressure. And the primary endpoint in this study is hospitalization or death, for any cause, through day 29. You can see there are other Phase III cohorts that have yet to read out. A 700-milligram plus 1,400 milligram combination cohort that is fully enrolled. We look forward to having that data soon and a subcutaneous cohort that is currently planned. On Slide 12, you can see the baseline characteristics for placebo and treatment arms. Arms are generally well matched. I just highlight here that this is a recently diagnosed population. It's just 4 days out from onset of symptoms. This was the time at which they received the infusion. And at that time, most of these patients still just had mild disease. So early diagnosis, mild disease still, but it's a group that has risk factors that suggest they're at a higher risk of going on to get more severe outcomes. On Slide 13, you can see a summary of safety and tolerability. This is adverse events in the study, which were similar between placebo and treatment arms and similar to past experiences with these monoclonal antibodies. I'd just point out that this is safety, excluding COVID-19-related events. COVID-19-related events are captured in the efficacy tables. Slide 14 shows the full listing of treatment emergent adverse events. And you can see these were all relatively uncommon, and we didn't note any differences between placebo and treatment that caused a particular concern. On Slide 15, I just remind us of the efficacy data that we had in our Phase II trial. This sets the stage for our design and analysis of Phase III. But what you can see on the left, this is the Phase II efficacy data for monotherapy. Here, the outcome of interest was COVID-19-related hospitalization or emergency room visit within 28 days after treatment, and we saw a 72% reduction in occurrence of that event versus placebo. We also noted on a post hoc analysis that these events occurred more frequently in patients who had risk factors for more severe disease, such as age over 65 or BMI over 35. These data formed the basis of our emergency use authorization with bamlanivimab. And in that emergency use authorization, we noted a 71% reduction in events in this high-risk group versus placebo. So when we started the Phase III, we set out to reproduce this kind of efficacy now focusing on this as the population, the high-risk individuals, focusing as hospitalization or death as the primary endpoint, not a secondary endpoint, and obviously, involving a large number of subjects. On Slide 16, then you can see the results from this Phase III study. And the study successfully achieved the primary endpoint in reducing COVID-19-related hospitalization or death from any cause. 7% of patients on the placebo group reached that endpoint, whereas only 2.1% on treatment, that equates to a 70% reduction versus placebo. On the right, you can see a noteworthy finding from the study, there were 10 deaths of any cause in the placebo group versus 0, no deaths of any cost in the 518 patients who are fortunate enough to have been randomized to treatment in this trial. I just pause for a second on this data to comment that the public health implications of these results could be enormous. A 70% reduction in hospitalization or death, if broadly applied, as well as a marked reduction in death for any cause, can have a profound effect on the course of the COVID-19 pandemic. Slide 17 shows how this drug works. It's no surprise to us. This is a drug that neutralizes virus. It is, after all, an antiviral mechanism and you can see very clearly statistically significant decreases in viral load at each of the time points. This is similar to what we saw in Phase II. Obviously, that was in a smaller population, and these results didn't always meet statistical significance. Obviously, they do here. On Slide 18, you can see a comparison of the Phase II data to the Phase III data, qualitatively quite similar in the magnitude of viral load reduction offered by combination therapy. We've also included in the blue line, the monotherapy results from Phase II for bamlanivimab alone. And as we noted before, bamlanivimab had a smaller effect on viral load, particularly at the later time points, presumably, due to the emergence of rare treatment-resistant variants at late time points in a few patients. However, we noted then and have confirmed now that there was not a difference in the efficacy on clinical outcomes that we observed for monotherapy as compared to combination therapy. As I said, in Phase II, the monotherapy showed about a 70% reduction in risk for those events and had very similar outcomes on symptom improvement as combination therapy did in Phase II as well as, as combination therapy did in Phase III. So the difference in viral load that doesn't translate to a difference in symptoms or outcomes between monotherapy and combination therapy. Slide 19 shows some of the symptom results that we obtained in the current study. Again, this was very similar to what we saw in Phase II for mono or for combination therapy with an early and sustained impact on symptoms. This particular graph is showing symptom resolution. It's a high bar endpoint. It requires an absence of all symptoms with an allowance for mild cough or fatigue. And you can see this reached statistical significance as early as day 3. Remembering day 1 is when they received the infusion. So 48 hours after infusion, already statistically significant benefit on symptoms that was maintained and even grew over the course of the study. The overall symptom assessment was time to sustain symptom resolution with 2 consecutive daily assessments showing absence of symptoms, and you can see that reached statistical significance with a p-value of 0.007. We didn't develop this drug to reduce symptoms, but it is a helpful marker that it is having an effect in the majority of patients. But of course, the effect that we're most focused on is the reduction in hospitalizations and deaths. Slide 20 reprises the COVID-19-related deaths in these 2 studies. So in the Phase III treatment study, BLAZE-1 on the left, there were 8 deaths on placebo arm that were deemed COVID-19 related, 2 other deaths in COVID-19 patients that weren't deemed that. But 8 COVID-19-related deaths on placebo and 0 on treatment. In the Phase III prophylaxis study in the prevention arm, 4 COVID-19-related deaths on placebo and 0 COVID-19-related deaths on bamlanivimab. Obviously, this data has really moved us based on results such as these 2 Phase III studies that are hitting statistical significance on their primary outcome measures and showing a marked improvement in prevention of death. Based on those results, we've made the decision that Lilly will no longer conduct placebo-controlled studies in high-risk COVID-19 patients. We now have enough data on monoclonal antibodies that we no longer have clinical equipoise, and we think it is doing patients a disservice and potentially harm even resulting in death, if they are randomized to placebo. So that's a change you'll see from us going forward. On Slide 21 is a summary of the results here that we found Phase III and the treatment population confirms and replicates our Phase II findings, obviously, with very robust p-values for a reduction in hospitalization, decreases in viral load and improvement on symptoms. These outcomes are very consistent with what we previously disclosed for monotherapy and are very consistent with the information that led to the emergency use authorization for bamlanivimab alone. The prophylaxis data to summarize 80% reduction in risk of symptomatic COVID-19 in residents that was highly significant and clinically meaningful and again, in treatment and in prophylaxis, is very significant impact on deaths due to any cause as well as COVID-19-related deaths. So what next? On Slide 22, you can see our progress here on making these medicines available for patients around the world. Bamlanivimab is currently authorized for use in early treatment of COVID-19 in a number of countries now in the Americas, in Europe, and in the Middle East, and we continue to work to expand its availability for patients around the world. The FDA is currently reviewing our emergency use authorization request for bamlanivimab and etesevimab at a dose of 700-milligram bamlanivimab plus 1,400 milligrams etesevimab. We also, now with this new data in hand, intend to begin global submissions for this combination therapy. Also with the new prophylaxis data for bamlanivimab alone, we plan to request an emergency use authorization for post-exposure prophylaxis in nursing homes for residents, where there is an outgoing -- ongoing outbreak. As Dave mentioned, in terms of supply, we've made a lot of progress as well. By the end of this month, we'll have shipped 1 million doses of bamlanivimab with another 1 million more anticipated to be shipped over the coming few months. We've partnered with Amgen to improve supply. And together with Amgen, we'll make additional 1 million doses of bamlanivimab plus etesevimab through just the middle of this year with many more to follow. Importantly, we have over 100,000 doses of the bamlanivimab plus etesevimab available today with 0.25 million available by the end of this quarter. We have, in parallel to this Phase III study, conducted a PK/PD bridging study called BLAZE-4. Results from that data support lower doses of the combination that were studied here. So for example, with 700 milligrams plus 1,400 milligrams, which is our proposed emergency use authorization dose of the combination therapy, we saw very comparable effects on both viral load and symptom improvements as we did for the 2,800 dose. Finally, we spent a lot of work trying to improve patient access and ease-of-use for medical professionals. As Dave mentioned, neutralizing antibodies are free of charge for patients wherever possible around the world. We have submitted data to the FDA supporting reduced infusion times. We know that the current 1-hour infusion protocol is burdensome for many health care providers. And we have data that hopefully will allow, pending FDA authorization, shorter infusion times as low as 16 or 20 minutes. And finally, working with various parts of the U.S. government, we now have numerous resources to connect patients with facilities that administer neutralizing antibodies. Slide 23 just highlights some resources for patients and physicians, starting with Lilly's 24-hour support line as well as a number of websites, covid.infusioncenter.org, which helps patients locate infusion centers near their home. Combatcovid.hhs.gov has a lot of information for patients from HHS on antibody therapy. And then on the right, the protect-public.hhs.gov therapeutics distribution website, which shows maps showing availability of all available monoclonal antibodies for patients. As you've heard, we're deeply moved by this data, and we're gratified to be able to bring forth such an important solution to what is surely one of the greatest public health crises that society has faced in many, many decades. With that, I'll turn it back to Kevin to moderate Q&A.

Thanks, Dan. We'd like to take questions from as many callers as possible. [Operator Instructions] Tammy, if you could please provide the instructions to callers for how to join the queue for the Q&A session, and then we're ready for the first caller.

[Operator Instructions] And our first question comes from the line of Ronny Gal with Bernstein.

Congratulations on a really impressive data you're showing us today. So thank you a lot for your effort on behalf of trying to solve COVID. I got 2 questions. First, I was wondering if you have been able to track the ability of carriers to infect others in the BLAZE-1 on a BLAZE-2. It sounds like household members could easily be tracked to see if they also got infected. So essentially, does your product beyond protecting the individual hospitalization also provide protection probability to infect others. And second, we kind of have to tackle the variant issue. There have been several articles about the U.K. and South African mutants and potentially their ability to scape those antibodies already available commercially. I was wondering if you can talk to us a little bit about what you're seeing in your experiments about both scape in vitro and in patients. And second, does Lilly plan to develop a second-generation antibodies to target some of those variants?

Thanks, Ronny, for those questions. We'll go to Dan for them.

Yes. Thanks, Ronny. 2 good questions. So the first one is a reasonable question, which is do monoclonal antibodies lower transmissibility of virus in patients who have been treated? We don't have the answer to that yet. But we do have important data that could inform that answer. So I showed viral load in the treatment study, which is dramatically reduced. I didn't show viral load in the prophylaxis study. But what we had -- what we did see is even in the fraction of nursing home residents and staff that still got infected despite being prophylaxed, their viral load was much, much lower. So the theory then would be if patients have a lower viral load, they're less likely to transmit the virus. Now that's still a theory. It's a very reasonable one. But we're trying to see if we can collect data to show that, that is true. Your second question on variants is also an interesting question. Of course, we're following that very closely. So let me start with the current situation in the United States. We -- according to all of the publicly available surveillance, bamlanivimab alone even, but certainly our combination therapy -- well, let me start bamlanivimab alone is effective or predicted to be effective against more than 99% of all strains that we see in the United States today. And bamlanivimab plus etesevimab should take care of the rest of them. That includes the U.K. variant, which is a very highly transmissible. Now there are new strains emerging in other parts of the world that we also have our eyes on. So of course, it's the South Africa variant or 501Y.V2, which is the one that is capturing the most attention and interest because it has so many mutations and changes in the spike protein that would be predicted to disrupt binding of most antibodies, whether those are monoclonal antibodies or antibodies that are generated in response to infection or even vaccine. So in response to that variant and also in anticipation that there will be additional future variants that are highly mutated. Emerging, we've developed next-generation antibodies, even just as quickly as this year, we've pulled forward an antibody that is very potent specifically against the South Africa variant, and that's moving forward quickly to human trials in case that variant becomes more widespread around the world.

Thanks, Dan. Ronny, thanks for your questions. [Operator Instructions]

Next question is from Sasha Pezenik with ABC News.

I guess I want to continue on the variant themes here. How much exploration of the various mutants within these trials, did you discover? I'd love to hear a little bit more about the antibody that you have in trials, specifically against the South African variant. And then in regards to stopping placebo trials, I would love to hear a little bit more about the thought process behind that. Certainly, there's an ethical component as well as a practical one.

Yes. Thank you for all 3 of those questions. So the first one is what are we seeing in our clinical trials, we're seeing a prevalence of variants that matches what's in the general population. So these kinds of variants are still rare in our trials, which were enrolled in the United States. Of course, who -- we don't see -- although we don't have full sequencing data yet as this is a very recently disclosed trial. For example, in the nursing home prevention trial, putative resistance variants were extremely rare in both the placebo arm and the treatment arm, and I don't think are affecting efficacy at all. In this combination trial, I would be surprised if we see a variants here that are of concern. In some sense, it seems to be going down the incidence of these variants versus what we saw in our first trial. We don't fully understand that yet. I just pause there for a second to say that the data from our 2 Phase III studies now, both of which drove to their final outcome of the main populations, based on that data, we know that these monoclonal bodies can work today, and we have them today. So the most important thing is getting access for patients today for these antibodies because the data show they prevent hospitalizations and prevent death. So your second question that was just on the South Africa variant and making the new antibody. This is just something that we've gotten good at, and we can move quickly now into clinical trials. I think the regulatory framework and the data generation question is important and that's unresolved. How much data will we need each time we make a new antibody, now that we really understand the preclinical to clinical translation, which appears to be extremely strong for these antibodies. Your final question was on placebo controls. This is something that comes up frequently in drug development, where once you have a drug that you feel confident is having a meaningful effect on patients, you no longer use placebo. In this case, we're pleased that we've reached this point with 2 adequate and well-controlled Phase III trials completely done. So we have a lot of data. We've shown part of the data that we've analyzed already. And now we can use these antibodies as control groups for future trials.

The next caller is Carolyn Johnson with Washington Post.

I was wondering if you could explain how fast you could launch and scale manufacturing, if you did have to generate a new antibody. And also whether you would anticipate that it would be backwards compatible to the current strains of COVID or whether you have to do like a 3-part cocktail?

Thanks. Dan?

Great. Two good questions. So we don't know how fast we can scale. I think last time we did it in, we started in March. And by October, we had that Phase II data, and we're ready to scale. I think we have the potential to go even faster in the future. So we're talking months to reach that stage. Of course, manufacturing scales from there. And that's why we need to start working on variants as quickly as possible so that manufacturing can be scaled before they're too widespread. Your second question was whether antibodies are -- can neutralize both South Africa strain and wild-type strain or any new variant and wild type. Right now, it is possible to generate antibodies that can do both. And certainly, that's what we have in our labs. In the future, as this virus mutates more and it's likely it will in response to normal immunity as more and more people have had the virus as well as vaccine immunity, it may be true that we need antibodies that are variant-specific and no longer bind to the original strain of the variant. And then you can imagine cocktails or combinations, I should say, of antibodies that go into 3 or more antibodies together. That's also feasible. And importantly, that's also a factor in why we're trying to drive dose down through some of our dose-ranging studies.

Maybe just to build on that with the manufacturing just set expectations. And one of the reasons Dan raises the question of having a clear regulatory framework. Because one of the ways we can make up time is by beginning manufacturing before we have all the clinical data. We're pretty confident in the preclinical to clinical translation that we see now. But the minimum time is realistically 5 or 6 months because you need to grow cells and cells divide at a certain rate. And that takes some time to get to scaled cycles in monoclonal antibody production, assuming you can free up the capacity and you define the process. But just to set expectations on that. The wave of variant we're really worried about now is the B117. That's the Kent, U.K. one. The good news is both mono and combo fully neutralize it with equal potency. So what's in front of us in the next few months from a public health perspective, we think this is a key tool to get in the hands of every clinician. What may await us later, we need to prepare for now, and that's the discussion about the regulatory framework.

Next caller is Navin Jacob with UBS.

One for Dan, if I may, and one for Dave as well, please. Dan, the antibody you're binding to RBD, I believe, but some of the variants have significant changes to the N-terminal. I was wondering how you're thinking about that? I mean, you've kind of alluded to that. But just as we see more changes to the terminals, does that affect how you're thinking about efficacy? And is that something you would want to react to in terms of developing new antibodies? And then maybe for Dave. Dave, obviously, an amazing accomplishment in such a short period of time, especially considering Lilly doesn't necessarily have a background in infectious disease. It's truly a very quick turnaround from you guys. Is this -- should we think about this as an entryway for you guys to have a new vertical for the company in infectious disease? Wondering how you're thinking about that, whether it's something you scale their internal processes or look externally to expand that vertical?

Thanks, Navin. Dan, first to you.

Yes. Thanks for the question. You're noting that our efforts here are focused on the receptor binding domain, which is correct, and that's mutating. Originally, when we started this project, working with the VRC at NIH, we set aside some really quite potent N-terminal domain, NTD binders anticipating that they could be used in case of RBD mutations. But it turns out that some of these newest variants, including the 501Y.V2 have deletions in the end terminal domain. And so the virus has sort of evolved to escape natural immunity against the NTD as well as the RBD. So probably those antibodies won't be as good. On the other hand, we do have antibodies that bind the spike in other ways that don't interact with the ACE2 surface. But yet the sort of side binding antibodies can still show potent neutralization, and these might be more resistant to mutations. So we can take those forward as well. Dave?

Yes. And on the strategy question, I mean, it's not something we're going to be definitive about now. Our task in front of us now is to help the world overcome the pandemic. We got here via a strong skill set in the modality of monoclonal antibodies. And I think if nothing else, we've demonstrated we're good at that, both discovery development and scaling up manufacturing, that will remain true. So I guess at a minimum, we would say, yes, we'll probably preserve some effort to surge when there's a global health crisis. Clearly, there's benefits well beyond the P&L of Lilly and we view it as a societal contribution that we should step up and make. Whether that makes, in nonpandemic times, for a productive business that competes for capital inside the company and makes sense externally, remains to be seen. We'll get through this wave of where we are, help define a path for variant mutation and then probably step back and reevaluate the whole thing at that point.

Next caller is Sasha Pezenik with ABC News.

1 or 2 follow-ups. The data here is really reflecting pretty high efficacy. So I'm curious what your thoughts are on why the uptick across the nation is still so low versus how much has been allocated. And then number two, President Biden has -- that he wants to invoke the DPA on a lot of different pathways to combat the pandemic from testing to vaccines. Have you been in talks with the Biden administration on whether you would be involved in a DPA effort for monoclonal antibodies and how did those go, if so?

Thanks, Sasha. We'll go to Dan first and then Dave, you can weigh in as well on her questions.

There's a number of factors that contributed to the uptick that was lower than we hoped initially. Of course, probably compared to any new drug that changes practice, uptake in adoption have been faster in this case. It's just that in a pandemic, we set expectations extremely high. And yet, there's still significant obstacles. We've talked about infusion centers, which needed to be set up and hospitals and providers needed to do that. And that's -- there's been a lot of progress there, information getting to patients. And then importantly, information getting to physicians who are often trained, especially academic physicians trained to be skeptical, and that's great in normal times and sometimes good in pandemics. But in this case, we've had to react quickly, often with partial information. That's what we asked people to do based on our Phase II data. Now I think we've closed that gap. This is a complete set of information that's highly compelling. And that one final obstacle, I think, is removed here, and we expect to see utilization increase quite dramatically based on this data set. I certainly hope that's what will happen that the impact on public health will be profound.

Yes. Maybe just to add on the question of how to get this out more. I mean, I do think we've suffered from 2 key constraints. One is information, which Dan touched on, and do leading experts sort of endorse this? Does it become the expectation for use? And that wasn't the case on Phase II data, of course, data takes time to develop, and we don't have time in the pandemic, so there's a bit of a paradox there. Perhaps we can reflect and learn from that. The other thing I'd mention, though, because we see a variety of execution across the country. There are some hospital systems, hospitals you haven't heard of and in places that are not famous for practicing medicine who have really excelled, and the -- some of the more well-known prestigious institutions have yet to give their first infusion. And that's not lost on us. And I think the observation I would make is that we have a lot of capacity in the U.S. health care system, it's not very flexible. And the more prestigious the institution seems to be the less flexible it seems to be. And what we need to do to respond both for vaccination and antibody treatment is move resources around rapidly to maximize treatment impact. And that has just not happened. At the scale that's necessary. Hopefully, on the first point, this data is convincing, a mortality benefit is always convincing in any disease state, but particularly now. And we're, as I said in my prepared comments, prepared to partner arm in arm with whoever wants to work with us to share best practice and allow the mechanics of more infusions to happen. So more lives can be saved. That's what we're talking about now. Secondly, on the DPA, we have -- I'm not aware of discussion we've had directly with the Biden administration on this. We did work with [indiscernible] folks without a DPA on freeing up input precursors to the manufacturing process. There are some constraints there on kind of lower technology items, but that are -- become rare in this kind of setup and may be used across vaccine and antibody production. They helped us get that prioritized. We've worked closely with the regulatory agency as well to move our own capacities around, that's been most helpful. If the idea is make more antibodies by invoking DPA, it won't work. Because the capacity for monoclonal antibodies is relatively fixed and the time to develop new capacity, a new plant is probably 3 or 4 years. We don't have time for that. We have to use what we have. The most impactful thing we could do is find the least effective dose, the lowest effective dose range and begin using that. And as Dan articulated, we have our BLAZE-4 program really focused in on that problem. Once a dose that could spread the available supply by a factor of 2, 3 or 4 to have the maximum human impact. That's what we need to do now in our view.

Next question comes from Seamus Fernandez with Guggenheim.

So just 2 quick ones. Dan, the prospect of simplifying delivery you guys are working to get down to a 60 minute infusion. Is there any possibility of bringing these antibodies as they stand today into a subcutaneous formulation. And then the follow-up question is actually in terms of studying a range of beta coronaviruses, not just SARS-CoV-2, is there a possibility that -- or do you see the possibility that a more highly conserved antibody or target on beta coronaviruses could actually result in something that is a worthwhile antibody that could be stockpiled per se? So just trying to see where Lilly might be sort of taking this approach.

Thanks, Seamus. Dan?

Yes. Seamus, your point on subcu delivery is a good one. Look, we're encouraged by the data we've seen from BLAZE-4 so far. These are extremely potent to antibodies. And we've been using them at extremely high doses. So I think there's a lot of room to go down. I don't think we're looking for a solution that involves a 10 mils of subcutaneous injection. That's probably 2 onerous for patients. But if we can get this down into just a couple of shots of more traditional volumes for patients, that would be a huge advance and could allow administration, for example, in primary care offices. So that's what we're working towards, and that's going to be our final Phase III arm there. In terms of stockpiling antibodies against a highly conserved region of beta coronaviruses in anticipation of future pandemic. It's one approach. I think flexibility in discovery and manufacturing, as Dave was describing, it could actually be more likely to play out as a success. It's hard to predict what the next pandemic will be and what the gene structure will be and what kind of antibodies we need. But if we retain that capability to engage quickly and we're surveilling the world for new strains that might bear out to be a pandemic and if industry and governments are willing to invest early at risk in partnership, then we could have new stockpiles and monoclonal antibodies developed quickly.

The next question comes from Gregg Gilbert with Truist Securities.

Sorry if I missed this, but when would you be in a position where supply would not be an issue? I think you've commented on 2021 specifically, but at what point would supply not be an issue sort of regardless on the existing antibodies. And my second question is about vaccination. Some consumer research we've done very recently suggest that a pretty large portion of the U.S. population is not ready to get vaccinated, even if there is supply. Whether that's true or not or changes or not, how do vaccination rates shape your thinking about how these antibodies will be taken up and used and maybe as importantly, how you should study antibodies in a vaccinated world in the future?

Thanks, Gregg. We're going to…

I'll start with the manufacturing supply question, and Dan can talk about the partially vaccinated world question. Gregg, I think so far, I mean, we had predicted we would have less supply the demand from the from the jump here. That didn't happen, not because there aren't patients, but for other reasons, which we've commented on here. We hope those reasons will fall down and at the same time, fewer patients be contracting COVID-19. So at some point, presumably, the lines will cross. But our capacity is rather fixed, as I mentioned before. The global capacity for monoclonal antibody production is finite. We can't create more during this pandemic. We could create more for the next pandemic. That's a different discussion. But for this pandemic, we sort of have what we have. We've assembled across 5 or 6 sites now bulk production. The next increment of capacity to go get is smaller than the smallest one we have so far. So it becomes quite piecemeal. The leverage is on dose. Because how could we go to treat twice as many, 3x, 4x as many people, it's studying lower doses that could have similar, although maybe not identical, similar effect. That's the path we have in front of us now. Presumably, we'll reach a point where fewer patients are being diagnosed, and we have enough supply to meet that, assuming perfect knowledge in the health care system and perfect delivery. We're not close to any of those things happening right now. So we'll keep making it. We'll keep shipping it to governments who want to approve it and buy it. And hopefully, those mechanical issues get sorted out over the course of 2021. That's kind of our view. For the next pandemic, we can talk more about that, but we need standby capacity that's much larger than what's available today. Maybe just one factoid to put this in perspective, the amount of protein we're making in response in the form of monoclonal antibodies to this COVID-19 pandemic is more than we usually make in a given year for all other therapeutic areas. So we've ramped up substantially and we could do a little bit more, but mostly we need to spread it across more patients with dosing.

Yes. Thanks, Gregg, your second question on vaccine. Maybe before I get on that, one final comment on supply is just to reiterate that bamlanivimab 700 milligram, which is the emergency use authorization for treatment of COVID-19, is widely available today in the United States. That figure I showed, the map. Each of those dots is a place where a patient can go and get an infusion of this drug. That doesn't mean it's always easy. I think patients and providers, in some cases, have to be persistent to navigate our health care system as is, unfortunately, often the case in different diseases to get optimal therapy. But for patients who want it and qualify and providers who think it's right for their patients, it is widely available. So getting that usage is the top priority in improving health. With respect to vaccine uptake, I think there are 2 separate issues at work here, one is how quickly the vaccine is rolled out and people use it and get vaccinated. And then the second is the durability of the vaccine response, particularly in the face of what is obviously an evolving viral threat here. Both of these factors impact our thinking. For example, in the nursing homes where our staff are working every day, doing clinical trials still today, we see variable uptake of vaccination, even among these most vulnerable members of society. In some cases, vaccination rates are low, less than 50%, in many nursing homes. And the disease rages on, and there are deaths every day in these facilities. That motivates us to be aggressive about prophylaxis, where we think we can have a quick impact during this sort of uncertain mixed vaccination period. Then the second factor is, will there be true second waves of disease in a post vaccination world, where people are getting either infected or reinfected with new variants. And that probably important our thinking on next-generation antibodies and the need to stay ahead of us.

[Operator Instructions] The next question comes from Vamil Divan with Mizuho Securities.

Again, thanks for the strong efforts here against COVID. So 2 questions for me if I could. So one, just as we sort of wait for the -- I know this is sort of new data, but while we wait for the full publication, can you comment at all if you noticed any differences in the responses you saw based on the patient demographics, based on maybe elderly patients or different ethnic backgrounds or anything along those lines that you could comment on? And then second, just a quick one, if you can remind us in terms of the economic setup between yourself and your 2 partners on these antibodies, the royalty that you have there? And is there any difference what -- if you're paying for -- if you're using the products in monotherapy or combination that anyway would change the economics on those things?

Thanks, Vamil. We'll go to Dan for the first question and then Josh for the second.

Yes. Thanks, Vamil. I have to say we don't have all of the different possible subgroup analyses completed. But what we've seen so far in this study and across different studies is relatively consistent. Which is to say that it's patients who are at highest risk for severe outcomes can get the most benefit from the treatment. That's not to say it doesn't work in people who are low risk. It's just that the absolute risk of achieving a negative outcome is low. And therefore, the benefit provided by antibodies is, on an absolute basis, a bit lower. But for example, in the nursing home study, we saw the most positive results among the residents and then intermediate results among staff members with the staff members who are at higher risk for severe complications, having more benefit that they could obtain from the antibody treatment. We saw the same thing in Phase II, as I showed in the treatment paradigm and now replicated in Phase III. So that's why we're focusing our efforts here on people who are at the highest risk of bad outcomes.

Thanks, Dan. Josh?

Thanks. On our partners, both Junshi and AbCellera have similar types of structures in terms of double-digit royalties on end sales. We also then have as we're producing the product, as Dave mentioned, we have a partnership with Amgen, and we have some sharing, some cost-sharing and profit sharing there. We haven't announced the price for the combo yet. So it's hard to sort of give you too much more on that other than to say at a manufacturing cost perspective, the combo will be more because we're not only sharing royalties with the 2 partners, but also some economics with Amgen. I think overall, though, we're -- these data support the outlook that we've given for the year, and we want to ensure that we get the right product to the right patient. And over time, that looks like that's combo. I think the economics will be good for all parties involved.

Is with Geoff Meacham with Bank of America.

I just had a few. Dan, another variant question. When you look at the range of variants that's going across the world today, does it reveal epitopes or even segments of the virus that are more optimal and perhaps ones that weren't originally considered when you developed your combo? And then the second question, more regulatory on the EUA for the combo, are you or the agency waiting on any additional data? I'm just curious kind of what the gating factor there is, especially considering the current infection rate trends?

Thanks, Geoff. Dan?

Okay. Geoff, your first question was on the variants, are there any well conserved regions. So it turns out, and this may not surprise you, that there's sort of an inverse correlation. So the parts of the virus that neutralizing antibodies work the best on are the ones that are mutating the most. And of course, the reason for that is because the virus is mutating in response to people's own immune systems. People make great antibodies, hopefully, in many cases. And if they are variants that have mutations that escape those great antibodies, that those are the ones that have survived. The other factor here is variants that improve transmissibility of the virus. So that's the other way the virus evolves is to get better at spreading. 501 is an example of that. And those kinds of epitopes certainly become attractive if you can stop the those kinds of spread. With respect to the emergency use authorization of combo, no, I don't think there's additional data. Of course, that submission was based on the data set that we saw in Phase II, which was a relatively smaller population. This confirmatory Phase III study certainly should give us all more confidence in the efficacy that we saw Phase II was, in fact a true signal.

The next caller is Louise Chen with Cantor.

So with the data that you have today, where do you see your combination treatment fitting into the standard of care for hospitalized COVID patients? And then second question I had was on your guidance. You've given some antibody sales guidance in 2021. And how does the combination treatment data today and the potential use of that impact that guidance, if at all?

Thank you, Louise. So we'll go to Dan for the first one, and then Josh for the second.

Yes. Thanks, Louise. Just a quick clarification is that these monoclonal antibodies aren't for hospitalized patients. But I think you were asking -- this is for recently diagnosed, generally mild to moderate COVID-19. You were asking though which patient segments might be for mono or combo. Right now, on an individual patient basis, we haven't seen any differences. The efficacy is extremely comparable across monotherapy and combination therapy. We're just anticipating a future state where there could be resistant variants that are more common over time. And so on a societal basis, it just makes sense to be prepared with an insurance policy of having the second antibody there so that we're prepared when those variants come. So right now, there's not any particular guidance about who you get one or the other. Over time, usage will initially be mixed if the combo is granted emergency use authorization. And then over time, as our supply of etesevimab fully ramps up, we would eventually just phase out the bamlanivimab alone. That will happen over the course of this year.

Louise, it's Josh. Thanks for the question on guidance. Just to remind everyone, what we've said for 2021 is that we should expect somewhere between $1 billion and $2 billion of sales of the antibodies. I think this -- if you look at what we've talked about in terms of supply of 1 million more doses of the mono and then 1 million of the combo by the middle of this year. If we sold all of that, then it depends on what the price is for the combo and where it goes, U.S. or otherwise. But you clearly could be headed towards the high end or above that guidance. But I think for what we know now, what we need is -- we don't have an EUA yet, so we need an EUA. We need government contracts on pricing for the combo. And then we need to get patients who need it, using it. I mean, of course, there's -- as we talked about, there's more patient need than supply, but there's a lot of things that need to happen here. So I think for now, we should stick with the $1 billion to $2 billion range, and we'll update it as we get approvals and contracts and otherwise.

Thanks, Josh. Thanks for your questions, Louise. Thanks, everyone for their question. We'll go to Dave for the close.

Yes. Thank you. In closing, we remain resolute in our commitment to battling the COVID-19 pandemic. We're encouraged with the speed with which we have been able to advance neutralizing antibody treatment options for patients, either used alone or together. And it addresses what is clearly a significant unmet medical need. We look forward to continuing to leveraging our capabilities and resources to sustain progress in the fighting against the virus as well as other diseases. I appreciate your participation in today's call and your interest in Eli Lilly and Company. If you have follow-up questions, please contact, respectively, the Investor Relations group or our communications team. Enjoy the rest of your day. Thank you.

Thank you. That does conclude our conference for today. Thank you for your participation and for using AT&T Event Conferencing. Thank you.