a16z Bio + Health #15: The mRNA Revolution with Moderna CEO Stephane Bancel

The content here is for informational purposes only, should not be taken as legal, business, tax, or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any a16z fund. For more details, please see a16z.com/disclosures.

Thank you all for joining us today. Welcome to a16z's Bio Clubhouse room. where we cover the future of bio and healthcare in a loosely structured and hopefully interactive and fun format. For those of you who may not know me, I'm Jorge Conde. I'm one of the general partners here at a16z. And with me tonight are my fellow general partner colleagues, Benita Agarwala and Marc Andreessen. So today's guest obviously does need no introduction. Stephane Bancel, the CEO of Moderna, is joining us tonight. Moderna, as you folks know, is the company that so quickly brought us one of the mRNA vaccines against COVID-19. And Stephane himself has been the CEO of Moderna since 2011. Before Moderna, Stephane was the CEO of Biomérieux, which I always apologize to Stephane in how I pronounce the name of that company because my French is garbage. But Biomérieux is a French diagnostics company. And I have to say that on, on a personal note, I had the opportunity to have Stephane as a mentor and advisor on my own entrepreneurial journey when I was just starting my first bio startup, which was probably the low point of Stephane's career, but I think he's recovered nicely. So thank you for all your support over the years, Stephane, and a very big welcome. Thank you for being here tonight. Okay, so I guess the obvious place to start in this conversation is to say thank you. Stephane, obviously what we've all been through as a country, as a planet over the last, whatever it is, 17, 18 months with this pandemic has obviously had devastating effects on many people personally, economically, health-wise. And so again, the first thing, and I think the most important thing I think anyone can say is thank you for everything you and your team and the company and the industry has done To help us get to the other side of this pandemic.

Well, thank you, Rory, for the kind words. It has been—I'm sure we'll talk a lot about it—an incredible 17, 18 months. I've been blessed to have an amazing team. I've been blessed by all their sacrifice and the extraordinary teamwork that people have done and and what they have done literally 24/7, literally seven days a week. I've been at the factory where I'm with engineers, you know, at 11 AM and they have not gone to bed yet. You know, they just took a shower in the morning in the locker room of a factory before coming and meeting with me and they're still working. I mean, people have made so many sacrifices. Everybody has known that every hour mattered. And the incredible journey of going from sequence to shipping product to the CDC in 11 months, was something that I didn't think first was possible, and two, is very humbling to have been part of.

So speaking of every hour matters, as the story goes, Moderna had the design for this vaccine within something like 48 hours. So I guess 2 questions for you there, if we could start here. The first one is like, A, how is that possible? And B, can you just walk us through what that felt like, how those early days played out?

Sure. So let me first start by those of you that might not be as close to mRNA as others. I mean, the key piece to know is that mRNA is code. And so what we've done over the years is we developed all the technology bricks mRNA chemistry, mRNA technology, formulation technology to put around the mRNA, to drop it in a vial, manufacturing processes to make the mRNA, manufacturing process to formulate the mRNA. So all the tools to make vaccines and therapeutics. We had done 9 vaccines in clinical studies before COVID became a thing. And given that I've spent 25 years of my life working with infectious disease, and I was personally very touched and shaped by a big food poisoning outbreak, which happened in '96 in Tokyo, where I was working in my first job. And we were the only company at the time at Biomérieux who had the technology to do rapid testing in food samples and children were dying in Japan. And you can imagine as an entry-level person, I was a sales rep, I would literally carry machines in the trunk of my rented car to bring them to hospital where parents were crying next to their children who were dying, internal bleeding, you could do nothing for them. And so I've always kept an eye for infectious disease. I read, you know, every day the FT, the Journal, before I start my day. Spend a bit of time online as well. And I've always looked for pathogens. So when Zika happened, we moved very quickly. And so what basically happened is between Christmas 2019 and New Year, I was in the south of France with my family. And I read in the Journal one morning a small article about A new infectious agent in China who has pneumonia-like symptoms, and so literally you know on my old iPad I double-click on the the home button. I open the the email app, and I email Tony's Fauci team. We had been working with them for quite a while, and I send them an email. Hey, what is that pathogen? Is the bacteria or virus? And the question was important because we had done 9 vaccines against viruses. We are still working on the technology to make vaccines against bacteria. And so it was an important decision point or information to have to decide if there was a way we could help or not. And they replied to me a few hours after saying that they don't know yet. They are still awaiting news from China, but they are aware of the situation. And a few days go by and we learn it is a virus, so we can do something about it. It is not flu, it is not RSV, it is not SARS, it is not MERS. A couple more days goes by and we hear it's a coronavirus. That's new. And then we're told that the sequence will be put online, I think on January 10th. And so I work with my team to be ready as soon as the sequence comes online. That they design the vaccine. We had been fortunate that we had been working on the MERS vaccine with the NIH, with Dr. Fauci's team. And so we had studied a lot of different protein of a virus, as they are called antigen, a lot of antigen. And in everything we had tried in animal testing, the full-length spike protein So the whole thing, which is a monster, it's 4 kilobytes, was always giving us the best vaccine if you were to code for a full spike protein. And so the sequence comes, my team on the one hand, Dr. Fauci's team on the other hand, independently analyze the sequence. So for those of you again, who are not familiar with it, it's just, you know, 4 letters. That code for the entire genomics information of a virus. And they both independently come with a conclusion after doing a lot of kind of protein structure modeling and stuff like that. They both, both teams come to the same conclusion, which is they believe the best vaccine to do with our technology is to code the full-length spike protein of, at the time, the virus that had no name, since then SARS-CoV-2. And the reason it took 48 hours is everybody was freaking out because we literally designed the vaccine in silico. So we designed a software suite called the Drug Design Studio where you literally can copy and paste a sequence and then you choose in pull-down menus some attributes about the technology that you want, the toolbox you want to use, and then you click order like when you order something on the web. And here you go, you basically lock down the drug and it goes to a factory and it starts making vaccine in robots.

Stephane, was this a software stack that had existed before that you'd already invested in over the last year? Or was this something that was spun up in kind of the heat of the moment as you were learning about a new pathogen coming online?

No, that's something we developed over the years. Because one day, when it must have been maybe 10 or 15 people still on First Street, the old building next to Rory's company, I walked into one of our scientists' office and he's literally designing an mRNA in Excel where every cell is a nucleotide. And I look at him like, what are you doing?

And that might still be happening amongst people listening to this conversation. That's still the norm.

Yeah, and I told him this is crazy. And he's like, why? And I'm like, because you just make one mistake in one cell and the whole drug you're designing is wrong. And we're going to wait 3 weeks for it to be made. And then you're going to be spending time and money to test it in animals. And in 2 months' time, we're going to get the result of an experiment. It's going to be failing. And I will have no idea if it's going to fail because the drug is not working or because the drug that we think was injected is the drug that was injected. So it's madness and we're never going to make it. We're going to go bankrupt. And so I got a bunch of extremely good, you know, IT employees that we hired. And I basically designed what we call the Drug Design Studio. For those of you that are curious, go online, look at our website in the research engine section. You will see a quick couple of minute video. And literally you design a drug on the computer, or scientists can do it from their iPad, from home, on the couch, or a computer anywhere in the world. It's like designing a Tesla. You pick the components you want, you click order, and you're done.

That's incredible. And by the way, for the record, that design that you had within 48 hours, was that the first draft? Of the vaccine, or was that the final version of the vaccine, or both?

So it was clearly the final version because after we click order, we never went back. We never changed one atom of that vaccine. The vaccine that many of you have have gotten was designed on the computer in 48 hours from sequence being online, and they took 48 hours because they were freaking out because they knew that the decision they were making.

So so 48 hours was too slow.

Yeah, yeah. I mean, they could have done it in 20 minutes or 10 minutes, because literally you copy and paste. When you know where the spike protein is in the sequence of the coronavirus, you go, you pick it up, you copy and paste with your mouse, you drop it in our software, you select 10 scientific or technology choices for the vaccine, and you click order and you're done. You could do it in 10 minutes if you had the guts to do it.

And by the way, just to flash forward a little bit, is this the future of how we get our vaccines? In other words, is that it? Like, from now on, let's put bacteria aside for a second. For viruses, you know, is it a question of sequencing, getting the sequence done, you know, designing the vaccine and hitting print?

Or maybe the corollary is, why didn't we have a flu vaccine Last year, let's say? Like, what was— what happened this year that was different?

Yeah, so what happened that was different is there was a pandemic, and the FDA and other agencies around the world did things that I never thought would happen in my lifetime. I mean, some days we will have 3 meetings in the same day with the FDA. That same process for normal products would be 3 to 4 months of back and forth and then using the law to tell them, oh, this meeting you're asking for, we will set you a date in a month from now. And so the velocity at which we could go by having in real time problem solving with the agency was incredible. And I would say the other piece is what the government did last year with Operation Warp Speed. This has been really something remarkable for me, that the government was willing to invest $10 billion to basically told— to tell, sorry, 6 companies where the government bet on 3 technologies: mRNA, adenovirus, protein. For each technology, they bet on 2 companies to reduce execution risk. And they basically said, you can have as much money as you want for clinical trial. But please go fast. Every day matters. And so this allowed us not to take clinical risk because we did a full phase 1, a full phase 2, a full phase 3. But this allowed us to do everything in parallel because we're just taking financial risk. And so, you know, usually you do your phase 1 when you have clinical data, both safety and antibody level. You will get your scientific advisors. When you have input, you will call the FDA. They will set up a meeting a month from now. And when you meet with them and they'll give you the green light to go into phase 2, then you will go and ask people in the factory to make the product for the phase 2. Because of course you don't want to make the product at risk if the FDA is going to say no. So all that things, I mean, I just described 6 months of work. But what we did here, because we had capital coming from the government because we're chasing this virus in a pandemic-like setting. We did everything in parallel. We made the phase 2 material before we started dosing people in the phase 1. And I told the FDA, as soon as we have safety data, we will not wait for efficacy data or antibody level, we'll go into phase 2 because we will not hurt anybody. We'll have safety data. The worst-case scenario is the vaccine doesn't make enough antibody. We will figure out while the phase 2 is already recruiting. But we're going to save a month and a half. And that's going to mean millions of lives at the end of the process.

Do you think what you just described is the government and the regulatory machinery taking a— unnecessarily taking a wartime footing, right, and being reactive and willing to move quickly and all of that? What aspects of this do you think remain in place as we move beyond the pandemic into more of a peacetime setting?

Sure, so I think for a company like Moderna, and I will not say that for all companies, given how well-capitalized we are now, and given we know the mRNA vaccine platform works beautifully, we are gonna be willing to take a lot of risk that a year or 2 years ago I could not afford to take. As you can imagine, an $800 million capitalized company where you don't really know how high the efficacy is going to be and how the safety is going to look like when you give a vaccine to several hundred million people, and by the end of the year, we're going to be at many more hundred millions, is our business appetite has increased 10x. Because as I know, all of our vaccines use the same 4 basic nucleotide to make the mRNA. Same chemistry, 100%. That the lipid is the same. The manufacturing process is the same. I believe my flu vaccine is going to work beautifully. I believe my CAV vaccine is going to be launched. I believe our Zika vaccine is going to work. The one that I have question mark on is our HIV vaccine, just because the virus is crazy complicated. And because the virus mutates at an incredible speed, even COVID, and I'm sure we'll talk about the mutant in a few minutes, the variant, sorry. And so our appetite to invest in vaccines is almost infinite now. Because if you think about the amount of capital I have to invest and the opportunities to improve lives and create returns, it's just How much of this, as we go forward—

and I want to come back to focus on COVID, but just looking forward for a second— how much of the future of vaccines will be— look, since it's just code, as you said earlier, we're just putting in whatever the right sequences of mRNA that we need to code for something to train our immune system. Will in a couple of years we just get a Moderna vaccine that has flu and COVID and whatever other common cold is around that season, et cetera, et cetera? Does this just become we're just getting patches of security software upgrade every year from you?

Is that the future? Yeah, so for respiratory viruses, yes. So we are working on the seasonal flu, which I believe will have a 90%, 95% efficacy. Unlike the 60% efficacy in a great year, you know, 20 to 30% efficacy in a bad year right now. And indeed, our goal is to combine COVID variant boost plus seasonal flu boost plus RSV plus a couple more that we have not disclosed yet, respiratory virus, where you get a single dose Moderna winter software update at your CVS every year. And you spend a nice winter without getting sick. So that's one product opportunity going after, which I believe is going to totally disrupt the seasonal flu vaccine. Because why would you want the old vaccine that doesn't really work if you can get the combination vaccine that covers many more strains of different viruses? Then if you look at the current space, I mean, it's interesting. I think there's a few buckets of opportunities for us. One is there are still a lot of viruses that are hurting humans for which there's no vaccine. One of the most exciting examples for me is our CMV vaccine. So for those of you who don't know, CMV is cytomegalovirus. It's the number one cause of birth defects in the US. Typical case is a woman who is CMV negative, I never been infected in her life by the virus, usually gets her first pregnancy going well. And usually the second or third pregnancy, one of the older child is at daycare, comes back with CMV, infects mom who is naive to the virus. And because she has no antibodies, transmits the virus to the baby in her womb. And it goes from brain damage, blindness, deafness, and death, including miscarriage. There is no vaccine on the market. It has been the number one priority of the National Academy of Medicine for 20 years. It's a very complex virus of the herpes family, and our vaccine, which is on its way to phase 3— and I'm talking weeks, not months or quarters— is a very complex vaccine. It has actually 6 mRNA molecules per vial because we need to make 6 different proteins in your body. And 5 of those 6 have to self-assemble in your cells to make a structure called the pentamer. And you need to make an antibody, a neutralizing antibody to the pentamer to protect humans. Nobody has been able to do that. Our phase 1 and phase 2 data make extremely high level of GB, the single antigen, and the pentamer, to the point that when I shared the data with Tony Fauci, human data, back in September 2019, Tony told me, I never thought in my lifetime I would see this type of human data against a CMV vaccine. And so There's a lot of viruses, Rory, that people are not aware of that hurt them for which we can do something. Another example I'm very passionate about is EBV. EBV is another herpes family virus, Epstein-Barr virus, which many of you might have heard cause mononucleosis in teens and young adults. But what is fascinating, Rory, is that if you look at epidemiology data at the national level in countries that have You know, very long longitudinal studies across population. You will observe that 90% of people that have multiple sclerosis, MS, are EBV positive and have had mono. 90%, 90. And so the question is, when you start to get the data in the clinic, do you go partner with a couple of those governments? You know, Switzerland, Sweden, Singapore. And you do a 10-year study for $1 billion, but when you have $8 billion going up, that's the type of investment you might be willing to make to go prove it's going to take you 10 years like Merck and HPV, but to prove to the world that, you know, every young teen should be vaccinated so that first they don't have mono, but two, more importantly, you know, 30, 40, 50 years later, they don't have MS. And the list keeps going on. Another thing about CMV that most people don't know is in Sweden they run a very big study and they show that people that have CMV, that are CMV positive, because CMV is a DNA-based virus, once infected you're infected for life, like EBV, like HPV and a few more viruses. So they show that people that are CMV positive live around 5 years less. Than people that are CMV negative. 5 years life expectancy at the population level. And they show that most of them die of cancers.

Not to mention post-transplant setting. Exactly.

Then you have, you know, pregnancy we talked about, birth defect, transplant setting, and then you have this in terms of life expectancy. So think about the power of vaccination in that setting. So there's a lot of vaccines like this that We are working hard to get to the market. Then you have a vaccine like seasonal flu, where you look at the current vaccine on the market and you're like, oof, it's not so good. Why don't we do this? Yeah, go a better one. Even, you know, HPV that Merck has, it's a very good vaccine, but it does not cover all the cancer strain of HPV. Well, you can go and do all of them. And then you can even think of a world down the road, which is not science fiction, Because in cancer, we already make, as we speak today, individual vaccines against your cancer. We do that in a clinic in 30 days based on the sequence of your DNA. That is something we're doing. It's not rocket science or science fiction. What if in 5 or 10 years' time, Rory, you go to your GP for your annual checkup, you do a blood draw, they check all your good stuff, And they also tell us, what is your current repertoire of antibodies? And we send to your GP or to your local CVS, you know, 36 years after, a vaccine for RORÉ that's going to boost all the viruses of concern in the environment in a vaccine that we make for you that has 10, 20 antigen in one dose that is designed for you. And there's a different one for your wife and there's a different one for each of your kids. That's something we know how to do at a very low-cost scale that we could individualize. We need, of course, to digitalize the whole PCs, to partner with people like LabCorp and so on. But if you think about the technology, all the bricks exist. It's all about integrating the whole thing.

That's remarkable. And so, like, if we— so we have now, we have this technology, we have mRNA, we're on the verge of a revolution, right? Everything that you're describing. If you look back across the technology sector, and I know you're a student of technology, virtually every new technology, when it really impacts the world, is when it becomes broadly and widely available, right? And so everyone gets access to mobile phones. Eventually everyone gets access to laptops, or not everyone, but the vast majority of the world. What is sort of the What is the timeline and what is needed to sort of get the manufacturing capacity online for mRNA so that it is very widely available? Are we 1 year away from that? Are we a decade away from that? Like, when does this become ubiquitous technology?

So I would look at 2 dimensions, José, to answer your question. One is around manufacturing. Next year, we're going to make 3 billion doses. With a B.

Incredible. Of the COVID vaccine?

Of the COVID vaccine. But we can make anything with that capacity because mRNA is like code. You make mRNA for COVID or flu or cancer drug with the same raw materials in the same physical reactor, in the same room with the same people, with the same equipment as the previous product. It takes around 5 days to make mRNA. And literally at the end of a 5-day, which is a week, you throw out the disposable bag and the next day a team can start with a new fresh plastic bag and the same raw material. You just need a different template with a different order of nucleotide. Again, it's like software, zeros and ones, just change the order and you code for another piece of software. And so the manufacturing process is highly scalable. It is not easy, but we invested a lot of money over the years that we can scale it pretty incredibly. And so the manufacturing piece for me is just a question of money and how much we want to build. Because again, if you were to tell me, look, next year we need $15 billion of manufacturing, I need around 9 to 12 months lead time to build it, but it's highly scalable. The second piece for me is discovery. And so one of the crazy ideas that we have not yet implemented just for lack of time because we've been kind of busy with the pandemic, but that Stephen Hogg, who runs the science, and I have talked about for years, is we want to open source Moderna to any scientist in the world. And so what we are working on now is to develop a version of a drug design studio, which I spoke about earlier, that is simplified where a scientist in Senegal or in Palo Alto or in Beijing can log online, design 5, 10 candidates against one virus that they are passionate about, click order, a few weeks after gets a FedEx from Moderna on their bench, can go try it in cells, try it in animal models. And whenever they're happy with a candidate that they want to take to the clinic, they can come back and order. And 3 weeks after, I have a candidate like Fauci did last year, ready to go into clinical study in their country. So we are building that kind of open source app store so that anybody can design on our platform mRNA drugs. Because the best ideas, like always, are not residing in modern scientists' heads. And the more drugs we're going to be doing in parallel, the more we're going to create a network effect around modern technology, and the more we're going to create an impact on lives because there's going to be more and more drugs hitting the market.

Incredible. Let me ask, by the way, Stephane, I've brought on some of my colleagues here because, as you can imagine, there's lots of folks that are itching to ask you some questions if you're open to it.

Sure.

And so I'll start with some of my colleagues here who I'm sure have their own questions they'd love to pick your brain on. Before I shift it over to Judy and Venkat and Becky, 2 final questions from me. One is, when do you think, if you could predict or if you can say, that vaccines will be available for, say, a 5-year-old? And I'm asking for a friend.

So 5-year-old, I think, is going to be early fall, just because we have to go down in age very slowly and carefully. Also because of the mass of an infant or a young child is much lower than, you know, 180-pound adults. We need to start at a lower dose and dose escalate again like we've done in the adults last year. And so those 2 things require time. So I would anticipate data available in the September, October timeframe.

Excellent. And then my second question, which I guess is related, as we talk about sort of the technology, there are variants out there. So far, all of the data suggests that the variants that have emerged Your vaccine or the mRNA vaccines still seem to remain quite effective against them. As we as you come up with sort of booster strains or booster patch versions of the vaccine, will the testing be a lot faster? You know, in other words, can you just make the case now that look, we all the core components have been tested; they're safe. We've shown efficacy beyond I think anyone's wildest expectations. And so, as we test. Iterations against variants or versions against variants, that that should be a lower, maybe not a lower threshold, but a different threshold?

Yes, and actually the FDA has been very useful and helpful when the South Africa variant emerged and was causing a lot of concern because the adenovirus, i.e., J&J and AstraZeneca, don't look really great in them. The FDA issued very clear guidelines Saying if you can show you have the same antibody level in hundreds of people, you can get clearance for the new vaccine as long as you, to your point, Rory, use exactly all the same chemistry, component manufacturing process. So the only change is a few nucleotides between product 1 and product 2. So that's super useful. On your question about the variant, The answer is a bit complicated, and because it's biology and I cannot make it up, I think there are 3 things to think about in terms of boosters, when to boost, what will happen with currently vaccinated people with variants. And I think the 3 variables are, first, time. How long ago have you been vaccinated when you get infected? Because as we all know, Immunity wanes with time. If you look at natural infection by coronaviruses, because there are quite a number circulating and giving the common cold to kids or healthy adults, but some of them are getting the elderly in hospitals, like one of them called OC43, which is believed to be at the root cause of what was called in 1890 the Russian flu, which was not flu, which was a corona pandemic, by the way, which is why sequencing is always cool to get the facts right. So time is an important element in natural infection of OC43. Time to protection is 1 to 3 years, and this varies widely based on your age, And so, so time is really important. The second component is when you get infected down the road, months or quarters after your vaccination, by what are you getting infected? Because if you take an example of a current strain, the UK strain now called Alpha shows the same level of antibody protection than the original strain with mRNA vaccine, which is great news, which is why In this spring, the US people were vaccinated at no impact of 117 becoming dominant because the mRNA vaccine works very well and the mRNA vaccines represent 95% of vaccines in the US. But if you look at South Africa or Brazil or now Indian strain, you see a drop in antibody to those variants. The drop is such that you're still above protecting level right after vaccination. But what is really hard to tell is if you get infected 6 or 9 months or 12 months after your vaccination schedule by a nasty variant, like let's say the Indian variant, are you going to get sick or not, or how sick are you going to get? And the last component after time and how bad the variant is that infects you Is your your health and your age. So the example I take for people is look, my mom is 72 years old. She has leukemia, and so she gets an mRNA vaccine. She got an mRNA vaccine in January as she was you know high priority person in Paris. But do I want my mother to go through the next fall knowing she has a weak immune system and she's 72? And she might get an Indian variant or yet a more nasty one by then? Of course not. So what I told Tony Fauci and many health ministers in the last few months is we won't know until we know how long protection lasts. And I told all of them, I think we would rather be 2 months too early boosting than 2 months too late when we have hospitals exploding again. And back into lockdown, and people dying. We are all living this pandemic as a real-time experiment. We have never had in the world a vaccine against coronavirus authorized. We never had in the world mRNA vaccines authorized. So we still have to be very humble about the science in terms of what we know and what we don't know. And I think we're not done with this virus. I think a lot of variants are emerging. Look at the UK right now, seeing cases going back up despite a lot of vaccination. Again, despite this being mostly adenovirus vaccine, AZ vaccine. But I think we need to be very careful about next winter. I personally want my boost with a variant vaccine in the September timeframe because I would like to have a nice winter.

That makes sense. Thank you, Stephane.

Stephane, I want to jump in on the idea of open sourcing your platform because I think this is just an incredibly visionary idea. At a16z, we invest in tech platforms that we see as these machines that generate assets. And so having other people outside of your company generate drugs using your machine, that's like the ultimate demonstration of that concept. So I'm curious to hear a bit about how you think that might shake up biotech in general. mRNA is one modality, there are many others. I'm curious if you think other companies will take this idea of open sourcing and run with it?

So, Judy, I think what you need to be able to do such a model is an information molecule. I don't think you could do this model with small molecule or recombinant. It's too slow. It's too expensive, not really scalable. Those are fundamental aspects of the technology that you need. So could you do this with gene editing? Yes. Could you do this with gene therapy? Yes. RNAi? Yes. So that's why I go back to the information nature of nucleic acid medicines, which is if you really industrialize, to Ori's question earlier, how to make it, how to scale it, if you have the entire infrastructure software-wise that you can have an infinite number of people designing, you know, thousands of mRNA at the same time, then you can open source the technology. If you need a bunch of physical chemists going and inventing the synthetic route for every molecule you put in the system, it's totally unscalable.

Well, maybe I can push a little bit. You know, what if, you know, there's a software, a magic oracle software where you can do structure-based design and synthesis predicted through whatever algorithm you have? Do you think that that starts to sort of open up the possibility for external innovation on a platform like that?

It would help it again on the silico part. Then you will have to believe that your silico model predicts toxicity in humans. I think when you start to get out of true biology platform, the biggest risk you have is that you might feel super cool about what the computer is telling you. There's still a lot that we don't know about the human body. You need the molecule that you generated to bind to one molecule or protein in your body to have toxicity. And this is really hard with what we know of biology to anticipate and know in advance today. What you think about mRNA as a platform is you design a vaccine from your house tomorrow once I've open sourced the technology. The chemistry we're going to use for the mRNA is the same one as the one for COVID-19. The lipid is the same one, the manufacturing process is the same one. So your risk of tox goes down by, in my opinion, a couple logs.

So Stephane, we've talked a lot about vaccine applications for mRNA and how that really has been the killer app. But I'm curious, what do you think is the moonshot application outside of vaccines that if it works would completely upend therapeutics as we know it today?

So actually, a lot of applications. Becky, the way I've always thought about Moderna, this is a company which over time, and I'm talking, you know, another 5, 10 years, not a few weeks, should have 10 to 15 verticals. And the verticals we call modalities is basically getting mRNA into a cell type. Vaccine is our first modality. So today as a company, we have 5 more modalities on the therapeutic side of things. 2 in oncology, 1 in cardio where we get mRNA in people's hearts, autoimmune disease, rare genetic disease in the liver. I mean, there are 30 to 50 rare genetic disease in the liver that if we get 1 to work, we could literally crank 30 to 50 drugs within 12 to 24 months because the biology risk is very low for rare genetic disease because it's the wrong instruction in the DNA of those kids from mom and dad. And so you know exactly what you need to go fix. And if you don't need twice, well, because mRNA is like code and you have all of the tools we talked about before, you can scale very quickly. We're working on the lung, how to get via an aerosol delivery into the lung mRNA with Vertex. It's starting to look really good in higher species like monkeys. We announced 2 weeks ago at our annual science day that we find a way via an IV injection to get mRNA into stem cells and to be able to act on cell differentiation inside your body. We can use gene editing using mRNA to code for a gene editing enzyme. Like many of you know Cas9, but there are many other systems to code that into the mRNA. So we can do that as well. And so it's tough now to tell me what's a killer app because I don't know what I don't know about all the innovation that my team is going to generate. I don't know what I don't know about new protein function that are going to be described in 2 weeks or a year or 5 years by all the academic labs in the world that are trying to work to identify You know what all the proteins in our body do and what are the system interaction of all those proteins. But one, for example, as an example that I find super cool, is we have right now a program in phase two with AstraZeneca in cardiology where we inject mRNA coding for the protein VEGF into people's hearts after a heart attack. And what we have seen in pigs, and as many of you would know. Pig is a very good model for cardiology. You know, I would never get excited by curing a few mice of cancer because a lot of scientists have cured a lot of mice with cancer, but not as many humans. But in infectious disease, the right animal model is a good predictor to human, like in cardiology with a pig. So what did we show in those pigs? Well, we show that if you inject mRNA as one-time intervention, It's between 7 to 15 injections that are done within 10 minutes in your heart where your heart got damaged by the heart attack. You've shown in pigs that we can restore ejection fraction, the ability of a heart to pump blood, to very similar level to pre-heart attack. So think about this. I mean, if you survive a heart attack, You will most probably die of heart failure because your heart muscle is so damaged through the lack of oxygen through the heart attack. But if you could within 48 hours of an infarct, like, you know, as you know, when somebody has a stroke, getting them to the hospital ASAP is critical because literally every minute counts for their recovery. If you could get within 48 hours, which is what we believe, somebody injected with that 10-minute intervention done in a hospital. You could potentially restore people's heart function because what the VEGF molecule does, as a highly simplified version— so the biologists on the call, please forgive me— is basically you inject mRNA coding for VEGF in the heart muscle. The heart cells are going to produce the VEGF protein on demand because again mRNA equals code. And that VEGF protein, what basically it's going to do in your body like when you cut yourself and you bleed, is going to call your cells, your stem cells to come into your heart to build new blood vessels. And those new blood vessels are going to basically revascularize your heart muscle to get back the heart function. And that's what we have seen in pigs. The phase 1 of that molecule looks very exciting. The safety was pristine. There was really no tox identified, no liver enzyme, no cytokine. And we were able to show in people by doing a control in every subject. So every patient got in their arm the drug mRNA coding for VEGF and also got placebo in the same arm. So it's kind of the best control that you can do from a biology standpoint. You don't have patient-to-patient variability. And we showed that in every subject, we saw an increase in blood flow at the location where you injected the VEGF mRNA around a week after the injection. And you showed no increase of blood flow at the placebo site. And the only thing that could explain an increase in blood flow is more blood vessels. So again, the proof is going to be in the pudding. The Phase 2 study is ongoing. But because of what we have seen in the pig, Because of what we have seen in humans in phase 1, and because I've seen over the last 10 years of the technology, I personally believe that the phase 2 is going to be quite interesting, and I cannot wait to see the data. So that's the type of thing that I think is going to be really kind of making our heads spin in terms of what you can think of doing. Because if you think about what I just described, this is a regenerative medicine in your own body. With one medical intervention of 10 minutes.

That's incredible. I want to hand it over to Venkat. I think, Venkat, you had a question, and then we have some other guests that have joined us that we want to make sure we have time for.

Yeah, definitely. We'll make sure that Rajeev, Jim, and Ethan have time. Stephane, this has been amazing. We hope to have you back, and maybe an episode dedicated to the stories where you're Jorge's board member. Those stories would be fantastic for a follow-up session.

I won't be attending this.

Exactly. So, so, Stephane, just, you know, thinking about— I was looking at some of the talks you gave at TED in Boston, and it's kind of amazing. You were describing mRNA as a software of life, and you said, I have this crazy bold vision of serving 100 million patients, which, looking now, as you quote, the billions, is pretty remarkable. We talk a lot about category creation, you know, in Silicon Valley and around the world. And category creation takes different forms in different verticals. Healthcare and life sciences has its own sort of friction points, manufacturing, regulatory environment, et cetera. Having built this entire new category, what advice do you have for entrepreneurs who are looking to do just that in this very regulated, you know, high-friction environment? Sure.

I think the first advice is don't forget manufacturing, because many times I've seen biotech companies with great science believing that manufacturing is easy. And with some technology like small molecule, manufacturing is indeed easy because we have been doing small molecule for 150 years at the pharma industrial scale. But I've seen companies literally going bankrupt because they were using new technologies and they had no understanding because the team was mostly scientific leading the company, no understanding of the difficulty of scaling up process development, of developing analytical methods to prove to FDA that your process is in control, that you have product purity, all those things that if you have been in the business for 20 years, you would know by heart in your sleep. But if the only thing you have done is science and you have not industrialized a pharmaceutical product, you might be making out of naivete and, you know, the typical unknown unknown to go after something and to not do things in parallel. And then at the end, being stuck with very long work that might take you in a position where you run out of cash, it becomes very difficult to finance the enterprise. As you guys know, what is really hard for entrepreneurs is to build a business with everything in parallel, because of course, whatever is going to delay you could be a kiss of death on the financing side of things. And that was the big realization I had, because this was the first time in my life I was an entrepreneur, you know. I've only worked in cash flow positive companies before. As I've told many students in many business schools who have been invited, the pressure of running a cash flow negative business is extreme compared to running a cash flow positive business. Because if you are late, you'll survive. You might look bad to your management, but you'll survive. In a small company, you might die or destroy extraordinary amount of value and product on the way. And so, so I think for entrepreneurs is to make sure you surround yourself with either your board and/or advisors and/or people on your team. We have seen enough movies of developing products, maybe not in your science or in your biology, but are going to be there to help you to start investing early in those other things you need to do so that you manage the entire drug, which a drug is a system, it has a lot of components. All those components need to make sense together. As we always say, you know, a chain is as weak as its weakest link. And that's what I think is a kiss of death in drug discovery and drug development that many science-based leaders just miss and sometimes at very, very steep cost.

It's good advice. It hits close to home. Thank you, Stephane. I want to shift over to our friend Rajeev Venkaya, who is in your neck of the woods. You folks may know each other. Rajeev leads Takeda's vaccines business unit. Rajeev, you have a question, comment?

Yeah, thanks, Jorge, and thanks, Stephane, for these phenomenal visions of what the future holds for the platform. And just for disclosure, Takeda, whom I work for, has a distributor partnership with Moderna. We're bringing Moderna's vaccines to Japan. In fact, that just started in the past couple of weeks, which is very exciting for the population there. Stephane, you have really hit on the fundamental— one of the fundamental challenges of traditional vaccine manufacturing, which is getting your process sorted out, under control, getting your analytical methods qualified and validated, which, you know, these things can set you back years if you don't get it right. I'm not telling you that. I'm telling the audience that because It may not be apparent. So that's one of, I would say, 3 key challenges in vaccine development and the one that people don't necessarily appreciate. The other 2 are around demonstration of efficacy and demonstration of safety. And when it comes to prophylactic vaccines for infectious diseases, they each carry their own challenges. Again, I'm saying this for the audience's sake before I come to my question. So in the case of safety, it's the need to do very large clinical trials to pick up rare events. Like what we've seen with clotting disorders for certain platforms and vaccines. And in the case of efficacy, because you may not be able to predict incidence of particularly episodic infectious diseases, you need very large trials usually. That was not, of course, quite as much of a challenge in COVID because there was so much virus circulating in communities. So given the— so you've nailed innovation in manufacturing for sure. What are you thinking about on the clinical and regulatory side? Side to expedite safety and efficacy monitoring. If you could fast forward, are there things that you're hoping will happen or that you'll be able to offer up to regulators that will make that process faster as well? Sure.

Well, Rajiv, good to talk to you. And indeed, delighted that we have now the Moderna vaccine in Japan thanks to great collaboration. I think 2 things. So let's teeth those, those pieces apart. On the safety front, I believe that the regulator over time, it might not be only after one vaccine authorized or approved, but will get comfortable with a platform as long as we, you know, keep the same chemistry for the mRNA, the same lipid, the same manufacturing process. And this has happened in the preclinical setting and the IND setting. You know, in the early days, We used to have to make GLP tox studies for every vaccine we took into the clinic. And after the 9 vaccines, we had a big holistic review with the FDA of all the clinical studies we had run around the world—Australia, Europe, and U.S., of course—all the tox studies for those vaccines. And we showed to the FDA that there's never been in human across 9 studies on 3 continents. A safety signal that was not seen in animals before, and that the tolerability profile within a range of dose was very similar. And so what the FDA did to us last year, which I always told my team would happen one day and they all said I was crazy, is we were allowed to go into the clinic with now several vaccines without doing GLP-1 studies. So we do non-GLP tox study to make sure there's nothing funny with the antigen, but we don't do GLP tox anymore. They agree with us that it's sacrificing animal for nothing. And that's if you want a big step forward in the preclinical to clinical transition phase. And given you know the industry well, you know it's a big deal. And so I believe the same thing will happen on the back end over time. And I can of course not guess if it's going to be Another year, another five years. But I think if you think about the corona vaccine now, we're gonna have you know five hundred million people dosed by the end of a year, maybe another billion next year as we are planning to do three billion doses. So those are incredible large number sets for pharmaceutical products. And so I think over time with a couple vaccine authorized. The FDA is going to realize what we've realized in the transition of preclinical to clinical, for the transition of clinical to commercial, that this is really truly a platform. And that, as expected, when you put the same lipid as a little ball of fat in a syringe for an IM injection and the lipid gets into the human cells and, and deliver an mRNA molecule that is made of exactly the same 4 chemicals than the previous mRNA, vaccine, that they behave the same way from a safety standpoint. And so this is where I think we're going to go. On the efficacy front, because we believe that mRNA is a platform, because we believe that most of our vaccines are going to work, and because we don't have to build dedicated manufacturing plants, It's really a question of money and risk tolerance for investment because, as you say, different viruses have a different attack rate, but you can solve a problem backward by throwing capital to the problem. And so would we have the appetite to run another 30,000, 50,000, 100,000 people study if we thought that this was the right way to go to approval quickly. If we thought we had the right cocktail of neutralizing antibodies based on the design of a vaccine and the challenge model we would have run in non-human primates before that we think would be relevant biology, I think that given the cash balance that we have and what we believe in the platform, I would have the appetite to invest even a couple billion dollars against the right vaccine and not to get $200 million of peak sales. But to get the right vaccine, because I really believe that the probability of success of a vaccine entering phase three is in the eighty plus percent. Like I think the flu shot is going to be ninety five plus percent probability of success. The HIV vaccine, as I said, because the biology is so complicated and the virus mutates so fast, I want to be very humble about this one. But I think that. in a world where you believe your vaccine is going to work with a high strike rate, your appetite to invest in the clinic, also because you do not need to invest in manufacturing, it's the same manufacturing engine. So as you look at the asset, I think you would have much more appetite to accelerate time and to get to efficacy by just throwing money at a problem. Does that make sense, Rajiv?

Yeah, that's really great to hear, Stephane. And I think it validates this thesis that by doing what you've done with the platform to reduce timeline, capital investment, and risk, you're making this more attractive as an investment option for people that could deploy their— for entities that could deploy their capital in many ways. And you, as the owner of substantial capital at this point, demonstrated that you're willing to put capital against this, given the de-risking that you've done, I think is Really valuable for the community to hear.

Great. Thanks, Roger.

Fantastic. So, Stephane, I want to be sensitive to your time. I know it's late where you are, so we're going to wrap up quickly. If you have the time, Stephane, 2 more questions for you.

Sure.

Is that okay?

Yeah.

Great. So I'll first go over to Jim O'Neill. Jim, thanks for being here. What do you have?

Hey, thanks for having me. So My question kind of builds on Rajiv's question about regulatory speed and in particular regulators' tolerance for reasoning under uncertainty. So early last year, a lot of economists were calling for challenge trials for vaccines. Since you developed the vaccine in January, challenge trials could have compensated people for some risk and probably led to much swifter approvals. Thereby saving tens of thousands of lives. Any thoughts on that and whether that should be an idea we should discuss for next time?

So we spent a lot of time thinking about it since January of last year. We had a couple even meeting with ethicists, infectious disease docs, and so on. The piece that always gave us a lot of pause of doing it It was such a novel virus that we were not sure of the ability to make in a dose-controlled way GMP quantity of a virus, because you need to make sure that what you give to healthy participants is the dose you think you want to give them, and do that following all the right processes of safety and good manufacturing practices. So that's something— sorry. So that's something that really worried us because we were worried about if that part of a challenge study in terms of getting the right strain at the right dose, and nobody really knew how much you get when you get a natural infection. There's numbers all over the place. And we were worried that we could have the wrong data in the clinical study because there were too many scientific unknowns about the virus and GMP unknowns, that given there were so many infections in the population anyway, that we were able to run a successful phase 3 in a matter of a few months, which is just what happened. You know, we started the phase 3 July 27th. We could not have started the pivotal study earlier. And as you know, by mid-October we were done. So I don't know, given it was a 2-dose vaccine, you know, prime and boost a month apart, I don't know if we would have saved so much more time. And given the uncertainty, including putting people's health and life at risk, we were already struggling with it.

Thank you. Thank you, Jim. Over to Ethan. Ethan, I think you may have the penultimate word here. Thank you so much, Stephane, for a wonderful room. I just have a quick question on the regenerative medicine side of the portfolio and whether you are thinking of applications for acute spinal cord injury, so delivering a cocktail of trophic factors and maybe other gene products that might help in that critical window. Obviously, this applies to stroke as well, but I was wondering in the spinal cord injury case in particular where there's such an acute window and such an unmet need.

So it's a great question, Yifan. So we have nothing in development right now. We have been playing with, with medicine in the lab quite a lot. We looked at going upstream, you know, and indeed doing different cocktails of proteins to see what we can do. The place where I think open sourcing Moderna to the academic world could accelerate tremendously the ability of experts to apply their depth in biology across a disease or a path, because we cannot be good at everything at the company. And so I think that's something that requires much more work. And I really hope down the road that we can help get, get people back after, you know, trauma onto literally their feet. something that we're interested in, it's too early for us.

Fantastic. Thank you, Stephane. So look, I know we're over time here. I just, I thought we could ask one last question, maybe in 2 parts, and then we will call it a night. The first one is, I'd be very curious to know how you and the company celebrated When you when you got the news from the FDA regarding the vaccine, because that that has to be such a transformative you know moment for any company, right? It's such a rare thing in our industry. So I'd love to know how you celebrated that. And the second part of that question, which is probably the most important question you're going to get asked tonight, is when they inevitably make a movie about this story, who plays Stefan Bunsell in the movie?

So I'm gonna pass on the second one because I couldn't hear the rest. On the first one, I might surprise you. So, you know, the FDA advisory committee, the VRBPAC committee who reviewed the vaccine, met on the first day and there was a vote around 5 or 6 PM, which was a unanimous vote for approving the vaccine. And at that time, we knew the authorization by the FDA was in the bag, given there was a pandemic and there was such beautiful data on the vaccine and such a strong endorsement by the independent advisory board. And so unfortunately, it was December something like '17 or '18 or '16, I forgot now. And we were all in lockdowns in Boston. So we couldn't get together. And the next day, we were expecting the EU authorization and the formal notification by the FDA. And I was so tired because, you know, we have been working 7 days a week since January of last year. And, you know, the psychological exhaustion when you go through a process. That I think around 7:30, I told my wife, "I'm going to bed," and she's like, "What?" It's like, "Yes, I'm going to bed," and she's like, "But you don't have your authorization yet." Say, "Look, it's gonna happen in 10 minutes, or it's gonna happen in 2 hours, or in 6 hours overnight. But it's happening. I can do nothing to change it, and I'm barely functional. So I'm gonna go to bed, and I'll see you in the morning." And so I woke up the next morning at four o'clock in the morning to go for a run, and when I opened my email. I saw the email from my team that we got the EOA. That's how it happened. I was sleeping. Incredible.

Well, thank you. I mean, I can't imagine what that must have felt like. Also, foreign morning run in Boston in December, it's its own kind of special feeling. Yes. So I want to say, speaking of sleeping, I want to say thank you and good night. Thank you so much for being here with us. Thank you for spending the time this evening. And going late into your evening to talk with us all. Thank you for my team for joining as well as as our guests Rajiv, Jim, Ethan for asking great questions. We hope to have you back soon, Stefan. And again, at the risk of repeating myself, more importantly, thank you for everything you, your team, Moderna, and the industry has done, and all the scientists have done to get us through this pandemic.

Well, Rory, thank you so much for the invitation tonight. It was fun. To catch up with you, and thank you for everybody for the great questions. I think we are living in an amazing time in the life of biology. I really believe, you know, the 20th century was the century of silicon and software, as Mark and many have shown us the way. I really believe the 21st century is the century of biology across many, many industries, not only for drugs. And so I wish you guys all the best in all the investments you are making, in all the companies you are building, those of you listening, because directly or indirectly, without you knowing today, like I didn't know 10 years ago, down the road you might have a huge impact on your friends, on your families. And so every day matters and let's all keep pushing hard because we can do, make the world a better place. So have a lovely evening and speak to you all soon.

Thank you, Stephane. Good night, all.

Thank you.

Thank you. Bye-bye, guys.

Thank you, Stephane.

Thanks, Stephane. Thanks, Jorge.

Thanks. This was great.