Moderna, Inc. (MRNA) Earnings Call Transcript & Summary

Hi, everybody, and thank you for joining us in the end of earnings day to actually participate in this conference call with Stephane Bancel, the CEO of Moderna. We'll try to cover 4 topics today a little bit by the logic of durability of the business of COVID-19 vaccines, the challenge of multivalent vaccines, curing additional diseases beyond immunogenicity, and a little bit about the experience of being a CEO of the company that attempts to change the world by -- while growing up at the same time. So Stephane, thank you very much for joining us today.

Thank you for having me.

So let's start with a discussion of the logic of boosting long term. So I think at this point, we kind of realize there is a logic to boost people once after getting the 2 initial shots. The question is the logic and evidence for the need for boosting every year. There seems to be some debate that we will be building immunity over time that we get a lot of plasma cells around this immunity will build with the second and third shot. And the question is do we actually have evidence that will tell us that we need a vaccination every year? Or is the verdict still out of this? Can you just discuss a little bit of scientific evidence, where do we stand?

Sure. So let me start by saying with biology we want to always be humble because there's a lot of things we don't know and don't understand as a scientific and medical community. So let me start by talking a bit about what we learn from nature. Because it is always a good place to start. Where there is a coronavirus that is separating the community, leading to hospitalization, is called OC43. It was actually the virus that caused the pandemic of 1890, that was called the Russian Flu, which we know now was not flu, but was a corona pandemic and -- because we have the same symptoms, and there was, of course, no [ PC ] at the time. So what do we know about OC43? So it circulated in the community in the U.S. and around the world. It is the course of around 10% of hospitalization in America. And if you look at the rate of reinfection and people having disease again from that virus, it depends by age. And I think this is where we're also going to be going with SARS-CoV-2 virus. If you look at the data around OC43 in the 50 to 60 years old and above, people get sick every year. It might be just like a cold-like symptom. People that are older or over comorbidity might be hospitalized, it is the 10% I spoke about. Some people will die from it. But in that older age group, it is not impossible if you look at the OC43 as people get reinfected all the time in the community. So I think that what happens with natural infection could be very similar to what we see with the virus, again, to be confirmed, speaking with humility. And then when you go down in age, you have people that get reinfected in the disease again on every 2 years. And then the younger people, I would say, 30 and less, it's every 3 to 4 years. So if I had to make a guess today, informed guess, I will guess that the vaccine may be with a similar schedule, which is 50 and above, maybe once a year booster. 30 to 50, maybe every 2 years. Maybe younger people every 3 to 4 years might be okay.

And I think the point is that OC43 is still with us. We just don't think about as an ongoing [indiscernible]. How is -- how does -- but we've never vaccinated against OC43. How does the vaccination interact with us? Essentially, you have a circulating OC43, to some extent because the population is not broadly vaccinated. We're now vaccinating people with vaccines like yours, which is better than 90% effective. Would that not change the dynamics you just discussed?

It's possible. But again, if you think about people getting infected every year with OC43, you will make a lot of immunity as one would think. But again, as the epidemiology data shows for this coronavirus all of people still get disease again every year. And that's the reason why it starts to predict precisely, but I think directionally, this is -- this might be similar in the case of mRNA vaccine than it is around natural infection by OC43.

And biologically, what happens there? You would expect that these plasma cells to stick around for a while and start producing antibodies upon introduction of the vaccine and the seasonal -- with the seasonality. Is this not happening, the plasma cells don't stick around? Is there something about this vaccination that leads to a lower memory?

Yes. So it's still a bit too early to know because, again, the vaccine has only been in the market for not even a year. So in virology, yes, it's not a very long time. And again, in the case of OC43, you will think that because people get natural infection, they get B cell, they get T cell memory and so on. And the fact that as people get older, as we all know, the immune system weakens with age, unfortunately, which is why older people tend to have more cancer and more infection than younger people, we might be thinking that you have the same thing with vaccination, which is a young person might be going on for 3, 4 years without needing anything and even if they get impacted 10 times, they might have 0 symptom. Whereas somebody who is 50 or 60 with some comorbidity factors, higher BMI other infections, you might be so sick of OC43 because you got the cold the week before from another virus and the one you got the week after when you go to your friends and you get OC43 through the dinner. That's the one that locked you down and that's the real life.

Okay. So the other question that comes out for this is the question of should you -- let's assume that I am over 50, I need to get boosted every year. Am I better off circulating between the Moderna, Pfizer and Johnson & Johnson vaccination? Or is the logic that I should get boosted with the same one? There seems to be at least some of literature suggested that being immunized and boosted with different vaccine actually improves your ability to mount a response when you get exposed?

Yes. So I've seen data both ways. So again, in a humble way, I don't think I will be definitive at this stage. I've seen both ways, I'm sure you saw [indiscernible] data from the [indiscernible] team, where they did a [indiscernible] march in the U.S. It seems -- because it was a smaller study, it seems that the best outcome was a Moderna boost on Moderna priming was the highest level of antibody. The first study I've seen where it was the same assay used for the 9, arms, because when you do a 3x3 matrix, that's what you get. I've read in the literature that some people think mixing is better. I got my Moderna vaccine, I got Moderna booster. Just because -- as I look at the mass, as you know in the Moderna vaccine, you have a much higher mass than the Pfizer vaccine, and so I believe that leads to higher anti -- protein spike into antibody, and so that's important for duration of protection.

So the census is that scientifically we don't know if it's better to mix and match or not. However, the Moderna vaccination is given twice at a higher dose. And thus, you might be just better off getting both of that just because your immune response to both -- because that might be more important than the variation -- a little variation existing in the dose. Talking a slightly different angle about the coronavirus. There's been a lot of discussion that now that we have broad population which is vaccinated, we will start selecting for more resistant variants or did not have Darwinian advantage over the fast-propagating viruses. Do we have any evidence of this? Is this something that you expect? I think you probably -- the folks in Moderna have probably thought about this a lot more than other people. Where do you guys stand on that topic?

Yes. So we think that vaccination, especially with a vaccine that are highly effective with high neutralizing antibody, will play a role in slowing down the risk of mutation. And I think it's just a market problem, that's how we think about it, which is, again, I'm going to get vaccinate with Moderna boost. I have a pretty high level of antibody versus if I've got on a Chinese vaccine, let's say that, as we all know, have much lower available antibodies. So if I get infected tonight at dinner, I will clear the virus faster because I have a high level of antibody. So I'm going to clear the virus faster from my system. And so it just gives from a mass standpoint and probability standpoint, the less chance for the virus inside my body as it replicates, as it's been cleared that it has a chance to mutate because as we know, it's a mass number. It's a big number game. And so we believe that as we get people with high vaccination rate and high vaccination efficacy -- sorry, higher vaccine efficacy, which is the case of America, and more and more in the OECD countries, because if you look at it, people are very quickly realized that the adenovirus was not as effective. That was pretty well reported when the Phase III went out across the board. And plus, the concern around blood clotting and so on that emerged very quickly, everybody knows. I would say most of the OECD countries are only mostly on the mRNA vaccines now. And so that's because vaccines have high efficacy, it would have helped a lot. The piece that this clear is at this stage, is there's a lot of people on the planet not vaccinated still, even though, of course, every day, that number is getting smaller, which is great. And also, in the South, though, there's been a lot of Chinese vaccine in the South, as you know, as part of Chinese using that as a diplomatic weapon. And as we know in the start, there's a lot of people that are immunocompromised. And for those that listen but don't necessarily know, it was not surprising to us as we follow every mutation that is posted online on a daily basis, to see South Africa strain, to see the Brazil strain, to see the Indian or Delta variant. Because in those populations, you have a very high rate of immunocompromised people living with HIV. And this is a very fertile ground for the virus to replicate and then mutate because it has been reported that people that are immunocompromised sometimes keep the infection in their body for a very long time. And so we have a high viral load across a lot of time. That's the big area of [indiscernible] of copies of virus that is very, very good for mutation versus somebody who is healthy, has a high efficacy vaccine, as I described myself, and then you create that zone. And so I do worry about the South a lot. I do worry about delta, which has become dominant in almost every country in the world is going to keep evolving. And it's really hard to know what is the next one going to look like.

So that goes the math question, right? So you guys obviously track the speed by which mutation accumulates in the markets that are -- that have not had a high vaccination rate with the mRNA vaccines. And essentially, the first question is, what is your prediction here? Are we bound? Some immunologists think we are -- it's almost deterministic -- the mass is deterministic we will get another way with another partially resistant vaccine -- variant? Other says, well, the numbers are still overwhelming. Where do you -- what does your math tell you about the risk of us getting -- the odds of us getting at least a partially resistant variant to the existing formulation of the vaccines?

So it's a good question. So we think the probability is low. Again, at the end, it's going to be 0 or 1, obviously. But we think today, it's low, so less than 30% that we have a mutation that is programmatic with the vaccines, getting the mRNA vaccines, as we saw with Delta, which has quite evolved since the Wuhan strain. The efficacy of vaccine is still very high. And especially when you boost people, as we've shown, we've got the antibody fighting against Delta. And so the question is really a question of time, which is, are you talking about the first mutation away from Delta, like the AY42, I think, is the name of it. That has been recently reported in a few countries. Or are you talking of a couple mutation away from that? And then you start -- of course, the more you drift away, the more you have a problem. The pieces we should not forget that I think is quite unique in this situation about the amount of vaccine is they come from the full spike protein. And this protein is a monster. It's 4 kilobytes, it's 4,000 base pair, which make it a very, very big protein. And of course, the bigger the protein, the more epitope, which are part where the immune system will make a neutralizing antibody too. And so compared to a very small protein of a virus, where you might have very few antibodies to that protein, so if it evolves away from it, you might have nothing to grab to it. And I think the reason you still see very high efficacy, especially boosted against Delta is you have all those other epitopes that have not evolved since the one strain that are still around. And so I think the question is really a question of when in terms of evolution. So when, when in terms of people got their booster? Because the more time goes by, we're waning immunity. If we had a very nasty variant next week, where most Americans can just work at their CVS and get the whole repertoire of antibody, not one, but the whole super antibody backup, we might be fine. Just use this variant that has evolved a lot from Delta showing up, let's say, next June, where people maybe 6, 7 months away, where we will start to see breakthroughs and start to see all the people getting hospitalized again. Does that make sense?

That makes absolute sense. Two things that matter. So essentially, I think you're telling me it's up to you to prevent the next variance from happening, go out there and get your booster shot. And the other thing is the number you put out 30% kind of spiked my ears a little bit, because it's not 3%. I think what you -- the way I translate it, you're saying that this is a real -- it's not a likely case, but it's a realistic outcome for where we stand today.

Correct. And the good news about the mRNA vaccines, which again is what most Americas have got so far is we can move very quickly. As you know, we've -- already have in the clinic, a Delta-specific vaccine, we have a Beta-specific vaccine. We're testing a Beta plus Delta. And so if it was a totally new strain, we will first, try very quickly to test that strain with the blood of humans that have received the current vaccine and those new variants vaccine in clinic because we have all that blood, so we know very quickly, okay, are we far away or not? And if we were worried because this was not high enough antibodies, we could I think within a 100-ish-day time period get to develop a new variant adapted specifically to the strain where basically, you reset the vaccine for that strain. So I don't lose sleep at night about it because I think we have a way out. And the other thing that is very different from what happened, let's say, over the last 12 months is when both mRNA vaccine were authorized, we didn't have a lot because both ourselves and Pfizer were still in building up manufacturing capacity. I mean we pushed out to the CDC 20 million doses in December of 2020 in the 2 weeks from authorization to end of year. But if you think about the output each company has today come off if we were to switch to a new construct, the first time you would be authorized, you will have in a vaccine that same month for their authorization, to vaccinate or boost plus you will be only 1 dose per human versus 2. So you'll be able to boost the entire U.S. population of anybody who wants one, literally the week of a -- the weeks after authorization. So it would be a very different dynamic in terms of response. And I get people would not be naive to a virus, except those that are unvaccinated of course. But people that have been vaccinated and boosted now, which is why I tell anybody go get a boost now because you protect yourself and you protect the community, because if you get infected with a higher antibody, you will slow down transmission.

So the other thing that you kind of brought up is this issue of the risk in the South. And this have come up in your -- in the context of Moderna in terms of your choice to do your own manufacturing as opposed to out license the technology. Can you -- the logic -- as a scientist, the logic is obvious to me -- in the end of the day, the technology seems very similar to any construct you'll create. And obviously, you've got a business to run beyond COVID. But can you just talk a little bit about what is kind of like the elements of technology you're trying to maintain? What are the element of technologies that are not a problem to share? And a little bit about how you guys made the decision about how you will support the availability of the product in the developing markets?

Okay. So that's a lot of questions. So I'm going to try to answer them, and if I forget a piece, please ask again. So first, when we started, it's important to remember because people don't always remember our history, is we had no manufacturing infrastructure. We had the 1 site in Massachusetts, there was a development site. It was built to do a lot of small clinical trial lots. And then there's a pandemic. When we set our goal in March 2020 with the Board and the management team, we said we're going to try to make 1 billion dose in '21. Just for context, the entire flu market across 5 companies annually is 500 million doses. So that was a crazy gigantic number, okay? And instead of taking 3 to 4 years to build it, which is what people do during Phase III, we had 12 months to build this. We thought 1 billion dose was going to be great because there was Merck with 2 vaccines in development and Sanofi, who was 1 of the biggest vaccine company, with a vaccine, and AZ. We have vaccine in J&J, with vaccine in Pfizer, BioNTech, of course, and Novovax and many more. And so when we thought about the 1 billion, and looking at the other players, we like, look, if we make 1 billion will contribute to the world. It is a crazy number for us to get to. And we didn't do it alone. We could not have done it alone. We partnered with Lonza in the U.S., and in Switzerland we partnered with Catalent and Baxter and Rovi and Sanofi and Recipharm and Samsung. So we partnered with a lot of people to maximize the output to get to those numbers. Then in February, March 2021, this year, we realized that the assumptions we had for what was going to happen, we are massively wrong, because Sanofi was 1 year late. I did not anticipate or plan for Sanofi to be a year late. Merck had 2 vaccines that failed. GSK didn't do anything towards the pandemic. Novovax was late. Sanofi was late. And so when you look at the world where it was -- and then they had vaccines at low efficacy versus mRNAs, which we didn't know back in March 2020 when we sized the capacity for manufacturing. And then there was a blood clotting issue for safety that really turned off, obviously, a lot of people. And so if you had told me in March 2020 that all those things were going to happen, first, I would not have believed you. If you're to tell me, Stephane, you're going to get a vaccine next year [indiscernible] efficacy. And by the way, because of your higher dose versus BioNTech, Pfizer, efficacy is going to last longer. And most of our companies are going to fall, and you guys are going to be the 2 companies standing with a vaccine that everybody wants, I would have not believed you. But even if I had believed you, the number that would have had to build for was most probably 3 billion doses. And so what -- when we realized early in 2021, that this was the world from a probability standpoint, that was happening, the scenario was happening, we went right back to the board, and at the board, we need to increase manufacturing, because while we don't need it for high income country, we're going to need it for low-income country. So we decided to go from 1 billion doses to 2 billion to 3 billion doses a year. So why do I say if we'll focus only on high-income country, we will not have needed it. It's pretty obvious, because I'm sitting here in Q1 2021. By 2022, in high income country, this was going to be a booster market, 1 dose per human, pays less than 1 billion people in OECD countries. So even if you assume 100% of people want a vaccine, which is not true, and we get 100% market share, which is not true. I see we'll have too much capacity just for the rich country. So we could have sat on 1 billion doses and be happy. And we said, no, we need to go to 2 billion to 3 billion doses, and we're investing, as we speak, several billion dollars to get the capacity to that point. And so that's what we're doing, and we're doing for a low-income country. Then the issue we've had in '21 is because of the commitment we made to the U.S. to protect Americans into Europe and a couple of countries like Japan and Canada and so on, is just the U.S. year-to-date has taken 60% of our output, 60%. So if you do the math, I could not -- we've -- assuming only 1 billion dose at 2 dose per human, I could not covered the planet. The math just never worked. Because again, in March 2020, when we decided for capacity, we did not believe that all those other vaccines were going to fail and we will end up with the best vaccine.

Fully understand. I guess the point I was aiming for was the question of the idea of licensing out of technology versus simply to continue to increase your own manufacturing. You obviously had to think through that when you decide to go from 1 billion to 3 billion. What was the consideration that you would go to do it that way?

So it was just what's the fastest path. We've always been obsessed since the pandemic started. As you know, we are the first company in the clinic in March 2020, and we were approved a week after Pfizer, which is only at the time 100x our size as a company. Now it's only 50x our size as a company. And so we're obsessed by speed. And so when we looked at it, we said, okay, how do we get to 2 billion doses as fast as we can? And the fastest as we looked at licensing and adding capacity in the sites where they already knew how to make mRNA, the answer was because we don't have a big team is -- and you need to [ detect ] transfer, and it's not easy to detect transfer, because the people who can transfer to know zip about making mRNA, they have never made mRNA before. It's not like recombinant or small molecule like Merck just did, which is great, they licensed the technology. You cannot find companies that have mRNA capacity sitting either. It does not exist. It's not the industry. And so where they going to teach knows nothing about mRNA. And so you need a lot of support like we did to Lonza in 2020, to get them up and running. And so as we solve for the fastest time to get the capacity up, we said, the best way is to have capacity in Norwood Moderna and in Lonza. And that's what we are doing now.

Got it. Makes sense. So let me transition away from the COVID-19 epidemic and talk a little bit about multivalent vaccines. So some of the discussions that were early in epidemic, and I actually asked Merck that on an open -- on their quarterly call is, can you multivalent mRNA vaccines and still get efficacy? And the answer is, well, there are some technical challenges associated with this. Can you discuss what are like the technical challenges associated with creating multivalent vaccines? You're obviously creating 1. 2, as a matter of fact, in the clinic today, 1 against influenza and 1 combination pan-respiratory vaccine. Can you talk about how those challenges are? How are they resolved? And what we still have to resolve to get those things to the [indiscernible] patients?

Yes. So I will -- I have a very different answer from what you just mentioned. We think all those programs are very manageable. We work our technology and our capability from a technology and scientific standpoint it is today. We have actually even one more than the one you mentioned. We have a Phase III now, so we got FDA okay to start the Phase III, for CMV vaccine, which is actually a very complex vaccine. It has 6 mRNA per dose. So 6 mRNA molecules in each vial, 1 coding for GB, and then this is science fiction for anybody who is not recombinant: 5 coding for each component. So 5 protein of what then has to come together into a pentamer. So it's a very complex protein structure where you have 5 proteins that have to come together as a very big complex. And if you don't have that, you want to have the right antigen, and you present it to the immune system, and you want to have the right antibodies. And we proved in our Phase I and our Phase II by taking the broad of humans that we have the right functioning antibodies, neutralizing in humans to the pentamer. And so we were able to even play with the ratio that we needed between GB and the pentamer. We've done it in flu with a quadrivalent flu vaccine. And we've showed actually in animals. But we've shown over the last 7 years that we've done vaccine in the clinic, this is now maybe we for 12 or 13 vaccine in a clinic with our technology, that animals with our technology always translate to humans in infectious disease vaccine, I will not say that of cancer, because cancer is a totally different beast from a biology standpoint. But what we showed in animal is that we've shown a vaccine with flu, RSV and COVID in the same vial, and we show that you can make the neutralizing antibody against all those viruses and [indiscernible] for [ fluids 4 ]. At the same level, if you go monotherapy as if you go into a combination. Because it's a very important piece from an efficacy standpoint, obviously, is that if you combine those things, you don't want to lose neutralizing antibody versus what you had with monovalent. And we've shown that.

So what is the trick there? So I would expect that if you drop from 100 microgram to 25 micrograms of 4 different antigens, you -- and assuming you're not changing the construct much, you will get 1/4 of the antigen being presented to the immune system. What is the -- can you talk a little bit about technology advances that allow you to get there?

Sure. So it's a very good question. So the first piece is on COVID. We are now at 50-microgram booster. And the products -- the multivalent, the combo product, the vision has always been a booster product, 1 dose per year, in respiratory and you're set for the winter. If you think about it through, we all have been affected all of our life, and that's why you can get with a single boost of shot per year. And so we anticipate the dose. We don't have the human data. They're going to come in a few weeks from now, now it's very soon. But we anticipate that you'll be able to get a much lower dose, COVID now is at 50%. And what is very interesting that we have shown is our lipid is biodegradable. That's one other thing that I don't think a lot of people realize is that why could Moderna get to 100-microgram vaccine, and Pfizer could not? And there's a lot of reasons linked to the lipid. Because if you go back in the Phase I data, Pfizer tried 100 microgram, and it was not tolerated at all. And there's also a manufacturing process plays a role. And so we think that by able to use 50-microgram for COVID, so we'll have to see where we go. But some vaccines, we've tried in a clinic given 150-micrograms. So we think without lowering the dose because, to your point, if you put a lower dose, you get less neutralizing antibody. We'll get to a point where we could have a very effective vaccine against COVID, flu and RSV at a dose that is well tolerated.

Can you talk a little bit about what makes the reactogenicity high. So essentially, I got the Moderna vaccine as soon as it happens, it will knock me out for a day, and the question is what drives that reactogenicity? And can we essentially reduce this on an ongoing development basis?

Sure. So we think there are several factors. There is a lipid that you get, which is why all lipid being biodegradable we're talking a couple of hours of half-life in your body is a big deal. The other piece is the antigen. We always believe that Moderna that -- if we can give you a high dose that you can tolerate, you'll make more antigen. And if you make more antigen, you make more antibodies. And we have shown that around many vaccines now. And we think that's a very important point to provide both protection but duration of protection. Because if you get from -- you start from a much higher base of antibodies and they go down slowly over time, of course, you'll be protected longer above whatever cutoff you need to be productive. So those are really the pieces. We have -- so a lot of the years the process plays a big role, how you make the mRNA, the purities you have. And this is kind of part of, I would say, the 3 big lever: The lipid, the quantity of antigen and the manufacturing process.

And is there like a continuous -- where are we in terms of moving away from where we were we designed your original COVID vaccine to where we are today? And I'm sure you've got a lot of other things to do this year, but I'm sure reducing reactogenicity was 1 of the efforts. And how far have you gotten essentially in the last year, 1.5 years in terms of reducing reactogenicity?

Yes. So we've done a lot of improvement on the process. And again, some of the reactogenicity we see, we believe is linked to the antigen.

So we can reduce it and get it to lower?

Correct. That's why the booster, the 50-microgram, again, it's still almost twice the Pfizer dose. So you get a lot of antibodies. But because you don't get a crazy amount of antibodies because you already primed and so on, that you get very nice protection as you should, the 50-microgram give you 30% above the level of antibody that you got after your second dose. So basically, you get your second dose antibodies, you go down with time, and then get you back 30% to 1.3x above where you were after -- well after the second dose.

Got it. Can you talk a little bit about your HIV program? Obviously, there, the problem has been very different than the problem that you're facing with a virus that is not hypermutagenic. Can you just talk a little bit what is the problem there that you're seeing? And where you guys -- what is your approach to trying to address HIV and where you are?

Sure. So again, let me start with a caveat, which is this is very risky and complicated biology, as you said. And we thought it was 1 of those public private partnership that makes sense for us to do, because this is funded by foundations like the Gates Foundation. And so we think it's a good use of our platform to make the world a better place. The challenge here, as you said, is the -- again, for those that are not as familiar with the biology, we have the challenge that HIV mutates very quickly. If you think of COVID mutates quickly, then study HIV, and you're going to see it's a different ball game, because it mutates within the same person's body over time. And so one of the feature of the technology we are trying to leverage in a way to help people is to adapt the vaccine to your own mutation. And we're leveraging the technology that we developed for personalized cancer vaccine, that is now in Phase II, awaiting data fully enrolled. So it's [indiscernible] therapy and in the other arm over it's [indiscernible] or plus cancer vaccine of Moderna. So how does that work? We basically take a biopsy of your cancer cell, next-gen sequence it, next gen sequential healthy cell, send both full genome to the AWS, Amazon, compare all the mutation to see where this mutation, there is an algorithm to pick the most important 30 mutation in your DNA, and it comes down to a robot or team build that looks like a big American fridge where we make a vaccine for 1 human being at a time. And we do this needle-to-needle of biopsy to injecting you back your own vaccine in 60 days. And we're still working to shrink this down. So one of the things that is being tested on the HIV front is could we do the same on HIV based on the virus that's in your body right now? So the idea here, I think it's really around compliance. As you know, the issue -- because it's great antivirals against HIV. I mean, what has been done by the industry over the last 40 years is just wonderful. But the issue that happens in a lot of countries, especially developing country is people do not take regular liver antiviral. And so we don't work very well. It's like people who don't take insulin every day with diabetes or don't control HA1C very well. They ended up having a lot of disease happening because they don't control diabetes or, in that case, HIV. And so the vision that we have with the clinicians that are working with us is could you do a vaccine so that you inject people every few months, every quarter, whatever, again, frequency to be determined in the clinic. But we -- basically, we make our vaccine for you that keeps your viral load below detectable levels, so they don't hurt you. And that if it mutates, we just do and do another vaccine for you again, and we can adapt the treatment as your virus evolves if you keep it in check, it's going to, of course, evolve slower and mutate slower. But the idea will be that over the year last time, we might make 3, 5, 10 iteration of a vaccine. But if we can do that, and then for a long time, you don't have compliance issue because, let's say, you get once a quarter, picking a quarter for your shot. There's no compliance issue, you stay in good health. So that's kind of the concept being developed. Again, it's a bit science fiction because it's a product per human being. But the beautiful thing about this technology is as we've shown with COVID, we can scale up very quickly. But also, we can go down and individualize product, which is in that science fiction to small molecule or large molecule.

It's a really cool idea. Where are you right now in terms of -- where are we? I'm assuming you're starting with patients who are resistant to the common antivirals we have today. But where are you now in terms of operationalizing this?

Yes. So it's going to be in the clinic very soon. Very, very soon.

So this is not a vaccination. This is the treatment through -- that's cool. It's a really cool concept.

[indiscernible].

Let's look away from talking about some -- introducing antigens to the immune system to using the mRNA delivery system for therapeutics that are not vaccine-generating, but rather look to create a protein that will have a functional role within the cell or the body. So can you just talk a little bit about the boundaries of this approach? Where is this actually better than giving an IV? I mean -- or a subcutaneous injection. I kind of -- my ideal drug is basically the IL-23, right? One quarterly injections and you get an efficacy for -- 90% efficacy against your psoriasis. But as we think about approach like this, the mRNA approach, what are the kind of the ideal places where this would work? And where would not work, which is probably more interesting now?

Yes. So let me take those 2 good questions. So we are not interested in doing me-too products that can be done by recombinant. What we tried since the beginning is to develop medicines that where we do things that you cannot do the recombinant or small molecule. So where is that space? It is intracellular protein, proteins that are used to make a cell function that you need the protein inside the cell. This we don't know how to do this using a recombinant, because if you make IV or subcu recombinant, you just want to go in your blood and circulate, it won't get inside your cells. But if the disease root cause of a mechanism is inside your cells, you need the protein inside your cell. But we get the amount inside the cell to make the protein inside the cell while it's there. So intracellular protein -- sorry, been speaking all day long. So intracellular protein, we love that space. The other one is cocktails of proteins. As you know, some pathways are 1 protein, 1 disease. But as we learned the hard way over the last 50 years for recombinant, there is not 50 million target that you can go where it's 1 protein and 1 disease. Because as we know, and we learn more and more about pathways and biology and feedback loops into the body, there are drugs where we believe the right approach is a cocktail of mRNAs or cocktail of proteins versus cocktail of mRNAs versus one. We have 1 example in oncology where we have in 1 drug, 3 different mRNAs calling for 3 different protein to help the IDEs a recombinant checkpoint approved given IV, plus the mRNA, with free mRNA product to improve the response rate of a monotherapy. So that's another example. The other example is protein that are secreted but have to act locally. And the best 1 we've announced today that the data should come on November 15 at the American Heart Association with AstraZeneca. It's a Phase II study, of which we're going to give an update where the drug is VEGF, not antibody VEGF, but VEGF protein. Where the idea here is if you survive a heart attack, can we inject in your heart very quickly, like around 48 hours window max of your influx, mRNA in your hearts coding for VEGF to grow new vascularization to revascularize the tissue of your heart that was damaged during the influx. Because as we all know, if you survive a heart attack, you will die of heart failure because your heart is so damaged from the influx. So that's a good example because VEGF was tried with recombinant, but it was extremely toxic. Because if you give VEGF IV, it go everywhere. And it is not good biology in nature like when you cut yourself and you have to grow new blood vessel because you cut the pipe and there's blood coming out and you're bleeding, your body makes VEGF locally for a short amount of time so that it induce stem cells to make new blood vessels. And so this was tried for recombinant, was massively toxic in animals, and then it was tried with gene therapy. And with gene therapy was also awful because you could never turn it off. So it kept making VEGF. And if you look at the animal model, actually, the animal on gene therapy VEGF were dying faster than placebo. Because just a lot of VEGF nonstop is not a good thing, it doesn't happen naturally. And so when you think about the mRNA that stick around for a few days, make a burst of protein and then goes away. That was felt by the scientists to be a very elegant way to leverage mRNA in a way that you could not use a recombinant or gene therapy or small molecule.

Good example.

So that's the type of things we're trying to figure out. So what doesn't work yet is -- well, we don't know how to get mRNA in your brain. So there might be a fantastic target against a very important CNS disease in your brain, where we could make the mRNA tomorrow morning in our sleep. But if we cannot get it in your brain, which we cannot today, we cannot design a drug. The thing we're trying to do at Moderna is to invest in science to get more and more applications, more and more cell types, where we can bring the mRNA. There was a big news in the last couple of days between Vertex and us. We've announced that now we're in GLP talks to get an inhaled mRNA in the lung for cystic fibrosis. And so that's a good example where now the lung is going to be accessible to us with an inhaled form, whereas before it was not. And so we continue to go after a new application of delivery to get a nucleic acid to a new cell type, we can go after new disease.

So there are 2 questions to follow-up. One of them is this issue of chronic administration. And the question is, is there a limitation here? Does the people develop after 10 administration a sort of antigenic or allergic reaction to the lipid? Is there anything that prevents those kids from taking the mRNA inhaler for the rest of their lives? What do we know on this question?

Yes. So what we have done in both nonhuman primate and published and now in humans is we've done more than 10 administration in a row. Some of them a couple of weeks apart. We've no problem observed in terms of safety or loss of protein production, which would have been one of the worries. We had issues in the early days of the company because the technology was not well refined. But as we've learned and improved the technology, we go to a point now where we don't think it's an issue. And so it was an issue before. Remember, we don't use viruses like AAV or whatever. And so it's a different system. But still, we engineered all the issue we observed that was sometimes leading to loss of protein production. We [indiscernible] those things out. So what went on today, we don't believe it's an issue with our current technology in mRNA.

And the fact that you're modifying your RNA a little bit, it's not native RNA, it's not a problem? I would think that...

No, actually, it's a massive benefit, that's why we modify it a little bit. So what we modified is the uridine, we use uridine analog. But the good news is that uridine exist in every one of your cells. That uridine is natural. It actually exists in your tRNA, and you have pathway in your cells to keep in balance the concentration of uridine that is used in new natural mRNA in your cells to the uridine analog that is used in your tRNAs. So your cell has a mechanism to [indiscernible] to keep the good homeostasis. And so the uridine modifier will bring, with our mRNA, first, is a tiny proportion of all the mRNA you have and all the nucleic acid you are floating in your cells. And then it has a system to be managed nicely. They are naturally occurring. They are in every one of your cells and they have been in humans forever that have been human on the planet. So we think it's not an issue at all, but it's a huge benefit, because it reduces the immune response. TLR7 and TLR8 get activated by uridine. That's how our body protects itself against viral infection. And so we had to treat the human body so that we can get an mRNA in your system without your immune system thinking it was a virus.

That toll receptor analogs are -- have got a role in antiviral, yes. The other question that came out from this, in addition to the safety, is the question of getting a consistent level of expression. So you can think about many, many things that will complicate that across individuals, and even within the same individual, it depends on how much cholesterol there is in the -- or did they have steak last night or not, right? You can think about many ways where this might have an effect. Have you kind of proven to yourself that people will get consistent level of expression every time and across people? And if this is the case, then why is it the case? Because naturally, you'd think that there will be variability.

Yes. So it's interesting because we ask ourself the question in the early days. And the most telling data on the therapeutic side because on a vaccine as you've seen in the clinical studies, you have -- if you remember the study, we had a very consistent antibody titer across people of different age, BMI, ratio, ethnicity and so on and so forth. The piece that was the most shocking to us in a good way of shocking is our chikungunya antibody program. So that's a program that was not a vaccine, where we made 2 mRNA in 1 lipid to make a heavy underlying of the antibody to self-assemble into an antibody. We try this in human. And when you look at the data, the CV between patients was tighter than recombinants.

Is it -- you get more persistent by giving the RNA than you get with the protein itself?

Right.

Interesting. Okay. Two last questions, and then we'll close for the day. The first one is, I kind of looked carefully at your pipeline, and you do not seem to have mRNA knockdown approaches. You are putting RNA into cells, and one of the question is, can you use this approach to knock down a particular protein's RNA level? And you're not using that approach. And I'm wondering...

Correct. So we are using mRNA to make protein for a gain of function, unlike [indiscernible], we use RNAi to turn things down. So I'd like to say us and [indiscernible] are the yin and the yang from a biology standpoint. And I think it's just because -- I mean, you could do an antibody inside the cell. If you wanted to take something down, you could make an antibody that has no signal peptides, that antibody does not leave a cell. That's an approach that we played within animals, but we never had a program that made sense to take to a clinic. We've been putting a couple of different approaches. So far, there's nothing that has been exciting enough for us from a biology and patient impact. If we find one, we will, of course, take it forward. So it's not impossible for us to do, but there are so many targets to do where you get gain of functions. That's a natural way to do it. So at this stage, it's all gain of function.

Okay. So the point is we just haven't found a reason to go that way.

Right.

The last question is about the direction of what Moderna will take. So you're now in an interesting position where you are a commercials company, you're a big research agent with a commercial company. And the question where do you want to take that? The company is obviously going to have a lot of capital. You can become a global infrastructure company. you can stay a research engine and think about your future products as being something that you license out. You can be somewhere in the middle. And as you -- as CEO, as you and your Board think about what Moderna will become post-pandemic, are we looking at like an integrated pharma company? Or of course, a biotech company, but robotic company that market cells have all those function well developed? Or are we staying primarily a research house that licenses out mature products to the commercial organization exists today? How are you thinking about this?

Yes. So we have already decided to build our commercial infrastructure. We already have people on the ground in U.S., Canada, the big 5 in Europe and Switzerland. And this year, we're opening -- we already have hired company managers in Japan, South Korea and Australia. And we're looking at adding more countries in the countries where we're not direct, we're going to use distributors. We think there's a lot of value for us to own commercial as well in terms of building the expertise, not only on the product, but also the technology. Because it's already same technology, it's a platform. We also believe that to increase the quality of launch, investing early is critical. I mean a good example is CMV, Cytomegalovirus vaccine. We believe this is a $2 billion to $5 billion annual peak sales vaccine, it is in Phase III now. If we're to partner this with the company -- there's always a question of when do you partner to maximize value for your shareholders? Too early, you leave most of the money going to big pharma; and too late, the problem you have is you don't educate the market. And I really believe having spent time in commercial when I was at Lilly, that investing early for this awareness, especially if you are the only player. Like CMV, there's no vaccine on the market.

I want to follow that, and vaccination, to some extent, is a very different commercialization process than others. But we just talked about our cardiovascular product, and we talked about oncology product. You have an interesting hammer, it potentially will find a nail in many, many different parts of medicine. And the question, you're not going to spread yourself across 8 therapeutic areas in the next 5 years, is the idea, okay, we will be a vaccine commercialization platform, leave other things to others? Or the ambition is broader than that?

So I think the ambition is broader, but it's not everything to your point. Again, cardio, the VEGF program is an AZ partner program, so they are a good cardio company. They're going to take it to commercial and that's great. Rare genetic disease, as you and I know, you don't need a lot of people on the ground. The biology is very straightforward, and you have patient association because those kids and parents have been waiting forever for a drug. We're not going after me-too drugs in rare disease. It's always about first-in-class in that therapeutic. So I don't think it's a heavy lift, again, speaking with commercial experience. We do a lot of sales force, and some heavy lift and so on. And so we should do it. The question is going to be oncology and does it make sense for us to do it so long? Or does it make sense to, let's say, I mean just as [indiscernible] vaccines partnered with Merck, we have a drug partnered with AZ, I mean, with [indiscernible], which is AZ's drug. So we're going to be practical at the end of the day. I don't think it's religion to be commercial for every product. But I think having a commercial presence and the commercial culture is important for any company. I think staying only focused on science and technology is great. But at the end of the day, we are here to deliver value to deliver products that are used in the marketplace. And so I think having a commercial muscle or commercial culture is important. If it means we are vaccine because the prelaunch is very important for market awareness increasing to a consumer, rare disease is pretty straightforward. I think I see that as base camp of what we will do for sure, and then the rest will be very pragmatic people.

Very good. Bancel, this is a nice place to end the conversation. So really appreciate your time here, and good luck. We're always very much hoping to see some of your successes looking going forward.

Thank you so much for the invite.