The a16z Show - What to Know about Those Vaccines

Episode Date: November 20, 2020

A vaccine for COVID seems to be (almost) here… or is it? What’s hype/ what’s real beyond the headlines (and beyond the press release), when it comes to the announcement last week from Pfizer and BioNTech that their vaccine candidate was found to be more than 90% effective in preventing COVID-19 -- and relatedly, the most recent news around Moderna's vaccine candidate? Of course, this was just the first interim efficacy analysis — so how close or far are we? What’s the significance of the readout and case numbers? How do we put all this in context of all the other (458!) programs in development? And how much should/ shouldn’t we read into the news, given the buzzy excitement and penchant for evaluating "science via press release"? a16z bio general partners Vineeta Agarwala and Jorge Conde recently broke it all down in conversation with Sonal Chokshi on our show 16 Minutes: the math, the science, and the practical considerations — from “vaccine efficacy” vs. efficiency, from cold chains to distribution, from patients to the system… as well as why mRNA matters in the present future of vaccines. Stay Updated:Find a16z on YouTube: YouTubeFind a16z on XFind a16z on LinkedInListen to the a16z Show on SpotifyListen to the a16z Show on Apple PodcastsFollow our host: https://twitter.com/eriktorenberg Please note that 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. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Hi, everyone. I'm sharing one of our recent 16 minutes episodes here since it covers COVID vaccines, a topic that's top of mind for everyone. So we wanted to make sure that everyone had access to it in this feed where we've long covered the coronavirus pandemic since very early on. And we also answer some questions in this episode that actually applied to both recent vaccine announcements. If you haven't subscribed already, you can find 16 minutes from A6 and Z in your podcast player app and do so there. As a reminder, this is our show where we talk about the the news, tease apart what's hype, what's real in the headlines, or in this case, quote, science by press release. And we always cover where we are in the long arc of innovation beyond the news, why it matters now. In the episode that follows, we broke down the Pfizer and Beyond Tech News and specifically what the phase three interim trial means, the difference between vaccine efficacy and efficiency and the significance of mRNA vaccines now and in the future. For those who want to learn more about how vaccines work, how they're made, phases of trials and regulation, what trade-offs are and aren't being made for accelerated COVID development of vaccines, the where's and who's of distribution, you can find that in a past episode we did in this feed, or at A6NZ.com slash vaccines. In this episode, A6NZ Biogeneral Partners, Vinita Agarwalah, and Jorge Condi, join me to discuss the news right after the announcement of 90% vaccine efficacy.

last week. But we begin by orienting where we are right now, given all the other efforts out there. So it's important to put into context of Pfizer News. You just go to biocentury.com slash coronavirus. They have a great, always updated tabulated data set of all the clinical stage programs and preclinical programs in the fight against COVID. Even though they're the first to report phase three interim data, there are actually 458 clinical stage programs in development. Hundreds more that are not yet at the clinical stage are ready to be put into patients and trials. And of these, you know, this ranges from preexisting drugs that are being tested for COVID, preexisting vaccines that are being tested for their utility in preventing COVID.

In this set of clinical stage programs, there are actually 43 other clinical stage vaccines that might be effective for COVID and 12 that are in phase three clinical trials. or already approved in some other part of the world. There are lots of efforts that are ongoing. But it's really one of the first phase three studies to actually provide a readout from an interim analysis of efficacy, how well the vaccine appears to be functioning

and preventing infections. And that's a big deal and a big moment for the fight. As Vina points out, this is big news because the entire world is holding their breath and waiting for an effective vaccine. Another reason why this is important news because it's validation of two things. Number one, to the extent that this vaccine ultimately proves to be as effective, it's validation that the target of the vaccine, specifically the spike protein on the

virus, is in fact an effective target. And that's good news, not only for this vaccine, but also for many of the other vaccines that are in development, which are also going after the same spike protein target. The second reason why I think this is big news is it's validation of a new approach for making vaccines. In the case of the Pfizer-BionTech effort, this is based on MRNA vaccine technology, which is really a first proof point that MRNA vaccines can be effective against the virus, in this case COVID. And Moderna similarly is also developing their vaccine based on this MRNA vaccine technology. I should know I just saw, as we're going to record this, that Moderna announced that they

have enough cases now so that they can do an interim analysis. So we are likely going to see some follow-up news from Moderna on whether or not their vaccine is effective as well. Let's take a step back of the 12 vaccine candidates for COVID-19 that are in phase three trials. Four are actually existing vaccines like the MMR vaccine, BCG vaccine, other TB vaccines that people are testing for potential utility in COVID-19. The other eight are specific to COVID-19. and in the breakdown of the eight, two are inactivated virus, which is a really common vaccine mortality that we've all had. And then six are adenoviral vectors of various kinds.

And that's kind of another platform that actually hasn't produced that many human vaccines yet, but has been in vogue for a couple of decades now as potentially very promising. In this case, you deliver DNA for the spike protein. of COVID-19 inside an adenoviral vector. And only two out of the 12 are RNA vaccines. One is Pfizer's and one is Moderna's. Yeah, I think the high-level point to make is the reason why MRNA vaccines is a very promising breakthrough from a technological standpoint is essentially providing the body

with the instructions to make the protein that you want the immune system to react to. And so in this particular case, what scientists were able to do is essentially take the genome of the virus and they abstract out the pieces that they'd want the body to react to and essentially print those in the language of MRNA and administer those to subjects. And then what is up happening is the cells take in those instructions, make the pieces of that spike protein that the immune system is going to react to, and hence you get the immune response. And that's very different than some of the traditional ways of making vaccines where you have to grow up the virus itself, either weaken it to nature, use subunits of the virus, and have that be the basis for the vaccine.

What this sort of portends is a plug-in-play approach to making vaccines. That could be a much more efficient way for developing and scaling vaccines for future threats. I think that's a big deal. What I'm really hearing is this theme again and again that we always talk about, which is engineering biology and a place here where it can have concrete impact. So now let's talk about this specific news. Pfizer had, they were supposed to have 32. They ended up having 94 and Moderna was supposed to do their interim at 53.

And I just kind of want to understand what the interim is, why that matters, and what actually happened here where they suddenly went from 32 to 94. Yeah. So some things here that just off the bat should give everybody some reassurance is that the data from regulated clinical trials is typically analyzed by an independent external body. In this case, a data monitoring committee that actually takes a look at the data alongside the sponsor of the product and participate in the review of the data and the report. of it to the FDA. So in all the kind of headlines you read about Pfizer said this, Pfizer said

that, yes, but remember that there are checks and balances built into the regulatory system. The other check and balance is that for any product, whether it's a drug or a vaccine, there are predetermined points at which readouts are done. So for a COVID vaccine, the most important number is not necessarily how many total people you enroll, but how many people get COVID in each arm. So the COVID case count is one of the most importantly determined numbers. And that number will be a function of how many cases and how many vaccine arm patients you have and how many placebo arm patients you have and what you expect the effect size to be of the vaccine. But in this case, Pfizer had predetermined with the FDA that they would provide an interim

analysis when they had 32 COVID cases across their trial. But they then agreed upon 62. And by the time the analysis was actually conducted, the valuable case count had reached 94, likely in part due to the fact that we're seeing a surge in coronavirus cases across the country. And so the infirm analysis that was announced this week included more than what they initially anticipated they'd have at the interim readout, but based on those 94 cases, they provided initial estimates of vaccine efficacy. Yeah, and one of the things that's fascinating about this headline number of 90% is certainly on the higher end of the range of what people expected in terms of how well this vaccine seems to be performing. Right. Researchers have been saying that the vaccines will probably only be 60 to 70%

effective. Does this mean that one can extrapolate that the result is now likely much stronger than initially expected because we ended up with those 94 cases. Can you guys say more about how vaccine efficacy works? Vaccine efficacy is strictly calculated as the risk reduction from the vaccine. So you take the risk among those who are unvaccinated minus the risk among those who are vaccinated and divide that by the risk among those who are unvaccinated. So for example, if the placebo arm in the study shows that you're getting a 10% chance of COVID in the real world, which is not that high, thankfully yet in the U.S., but let's say it were 10%. And the vaccine arm showed only a 1% chance of COVID. You would say that the vaccine efficacy, as measured in a

randomized control study, is 10% minus 1% or 9% divided by 10%, which is 90%. And so that suggests that you have decreased in the vaccinated arm the risk of COVID-19 by 90%. You're 90% less likely to get COVID-19 if you have had the vaccine. And as you said, that is a number that is certainly on the higher end of what people were anticipating. I was going to add it's one minus the relative risk, right? So one minus the fraction of people that are vaccinated that get the disease over the percent of those non-vaccinated. get the disease, right? So if you're 100% protected, it's 1 minus 0 over 100, so that's 100% effective. But that's the dumb math version. I was about to say that both ways was not easy math,

but I love that Benita has a pension for doing math on air. But remember, it's still an interim analysis. So that's the whole purpose of agreeing upon these case numbers with the FDA and making a statistical plan before you run the study, which is to say that you estimate what you think the effect size will be, you create interim readouts based on the number of cases you think you'll need to get confidence on a particular effect size. And then you have to kind of stick to that original schedule. And we're just over half the way there. The final analysis will be conducted when the case count gets to 164. And at that point, we'll have an even tighter readout on the vaccine efficacy. So Natalie Dean's an assistant professor of biostatistics at University of Florida,

who specializes in emerging infectious diseases in vaccine study design. She basically has a tweet storm called Vaccine Efficacy 101 that breaks down various topics. And she also wrote a paper with Mark Lipsich, who's the first author on that paper for Science Magazine last month, that answers the questions about how well does a vaccine prevent severe disease, how well does a vaccine prevent infection and contagiousness, how well does a vaccine work across different subgroups?

So I'll link to those in the show notes. Well, the other thing that might be worth highlighting is the fact that we're talking about a controlled clinical trial setting. And one of the things that's particularly tricky about this one is that this vaccine requires two doses. And so you're going to require people to get that first dose and then return to go back and get that second dose when it's indicated. And they're going to be less than 100% compliance in terms of people getting both doses. So this is probably going to have an impact on how the vaccine performs in the real world. I'm glad you said that because one of the questions I was going to ask you is what is the

difference between efficacy and effectiveness. What Jorge just described would be vaccine effectiveness, which is sort of the real world efficacy of a vaccine in contrast to what you measure in the context of a controlled trial. I totally agree that effectiveness, especially for a vaccine that requires two doses that are three weeks apart from one another, is likely to be lower than the measured trial efficacy. Folks might not realize, but other multi-dose vaccines that we think of commonly, typically not posing quite the imminent health risk that we all feel right now with COVID-19. But adherence in some studies to multi-dose hepatitis A and B vaccines is like in the 20% range. Patients just don't come back in time.

It's difficult to track them down. They forget that they didn't actually complete the series. So I'm hoping that the urgency around coronavirus is fundamentally different. But we do have lots of empiric data on multi-dose vaccine schedules. to be a little bit nervous about how this is rolled out. I was actually thinking exactly of the hep vaccines when you said the multidose and remembering that it's like, what is it, like six months delay, it's actually a long period between doses. It's not even just the number of doses, but the distance of time between them.

Which definitely makes it harder. So I think three weeks is better. Like it's top of mind. You just got a shot three weeks ago. You're supposed to go back. That's, you know, a lot easier to remember. So I'm hoping that is absolutely a lower bound, but just worth noting that the battle's not over. Oh, and then one more topic. When we're talking about even efficacy at preventing infection, remember that patients in both arms of the study, they didn't get like daily COVID tests, right? So we actually don't know the efficacy at preventing asymptomatic transmission or infection. There might be infections in both arms of the study that we just don't know the distribution of because they're symptomatic. That's a great point, Benita, because I assume if you have a vaccine that

that is effective at preventing disease, but not effective at preventing infection, you don't protect other people by getting the vaccine. You just protect yourself from getting sick. Exactly. Okay, so one of the things that I wanted to ask you guys now is really about what's hype, what's real, when it comes to really how excited we should or shouldn't

be about this news and how to kind of put that in context when it comes to practical implications and what it means. Yeah, look, there are several open questions that still will need to be addressed. I think there are four Ds to consider here. The first one is durability, which is how long will protection last? Will this be protective for vulnerable populations like the elderly?

Number two, are there any dangers remaining here? Are there any longer-term safety concerns? We need the vaccine to be effective, of course, but we also want to make sure it's safe because it's going to be broadly applied to the general population. And so they're going to look at safety data over a longer period of time. I think we're already approaching the two-month mark in some of the trials that are already ongoing. The third D, I would say, is distribution, is when will it be widely available?

There are a lot of challenges that will come here to distributing this broadly. One is it has to be manufactured at scale. Two, some of these have, especially in the case of the Pfizer vaccine, have very complex coal chains that will need to be managed if we want to get this vaccine broadly distributed and widely available. And then the fourth D that I would say is, what's the risk of denying it? nihilism here. There's been a big concern about what folks call vaccine hesitancy, where people are skeptical or scared of taking no novel vaccine. And so to the extent that the general population

is hesitant is going to obviously limit its impact. Now, my hope, of course, is that to the extent that the data holds up, that that hopefully will minimize some of the vaccine hesitancy concerns. But there are real concerns, and those among the other ones, we're going to have to be addressed. I have a question I want to probe on about the cold chain. So specifically, you talked about the stringent requirements. Apparently, the vaccine must be shipped and stored at minus 70 degrees Celsius, which is minus 94 degrees Fahrenheit. And of course, for me, immediately, I thought about my extended family relatives in India and how it's actually quite hot there. And I really can't see like the coal chain at that level necessarily playing out because of the refrigeration and the power consistency required, et cetera, et cetera. Can you say more about some of the cold chain. of the complexities about the cold chain here and the supply chain logistics involved on the ground? I think it's worth contextualizing. Most fridges are like roughly four degrees and most hospitals have fridges. Then below that, there's a minus 20 degree storage scenario, which is where a lot of molecular biology reagents are stored, often in labs. And so typically universities,

most academic medical centers will have access to minus 20 degree storage. What we're talking about here is minus 70, which is just like a whole other range. Most hospitals, even academic medical centers in the United States, do not have access to this kind of freezer. So there's got to be something where the manufacturer of the vaccines would actually deliver and distribute not only the vaccine, but also ultra-low temperature freezer boxes in which to hold the vaccine, what they've described is that you might be able to take the vaccine out of those boxes. And so, you're talking about, and keep them in a regular fridge temperature for up to potentially five days, but they're trying to get to more flexibility on all of these parameters.

I think it'll mean that it's going to be somewhat centralized distribution where you're going to have specific centers that have the infrastructure, and then we have to figure out how to get people to the centers versus getting the vaccine to the people, which of course make it much easier to distribute. Of course, this makes me think of Zipline and what they've been doing in Africa and even in the U.S. now with figuring out ways to kind of create this layer that can reroute existing infrastructures via drones. I think from a systemic standpoint, it's important to keep in mind what an incredibly ambitious

mobilization effort it will take for us to all get access to an effective vaccine. You know, there was an article in the Wall Street Journal on how the ingredient that they use for delivery is this liquid nanoparticle. They rely on the sourcing for that to come from a small, family-owned company in Austria. So this is a very delicate supply chain that's going to require a lot of people coordinating to make sure that we can make enough doses of this vaccine.

And then it's very different than the majority of vaccine campaigns, which are either targeting one patient population or one group of people. This is going to be something that should be at some point available for everyone. So this is going to require an incredible amount of mobilization across the entire healthcare supply chain, for manufacturers to provider systems to governments and beyond. Let's talk a little bit more about who gets it when.

We did talk about this idea of vaccine nationalism in our previous podcast with Rajiv Benkaya, and people can listen to that for more there. But I'm actually particularly interested in what happens with high-risk groups getting it first. The National Academy of Sciences, Engineering and Medicine laid out sort of the guiding framework for how we would phase in. Who gets access to the coronavirus vaccine? I'll try to link to that in the show notes. Yeah.

Healthcare workers and people living in nursing homes will likely be at the top of the list to read the first phase of vaccine distribution. The purpose of the phase three study and the endpoints that FDA and the sponsors have designed is precisely to assess not only efficacy but also safety. Most of those studies endeavored to include a fairly representative, diverse patient population. In Pfizer's case, they have enrolled patients over 55. They haven't yet published all the breakdown of exactly how many patients were in each age category. We're hoping and very much expecting the phase three study to provide data that is representative enough to apply to

a wide range of people. So what are future implications when we think about the direction of what this means for engineering biology and specifically vaccines? This is a harbinger of a revolution in our ability to make vaccines programmable. Traditional vaccines that were inactivated virus particles were very specific. All parts of the manufacturing were very specific to the virus that the vaccine is intended to target. In this context, the nucleic acid RNA, which is easily programmable, is being delivered in a universal particle called a lipidano particle. And so it does paint the vision to be able to put all kinds of different viral or pathogen sequences into this universal delivery particle to be able to spin up vaccines against different diseases more quickly than we have in the past.

I would say the big jump at the risk of oversimplifying it is we used to have to grow our vaccines and now we can print them. And that's going to be a big change. So I'm going to actually quote something which is an op-ed that Walter Isaacson, famous author, et cetera, wrote he was actually a volunteer in this particular trial. And he wrote, quote, it is another wondrous miracle from a biotech revolution in which knowledge of genetic coding will become as important as digital coding and molecules will become the new microchips. So now I'd love you to bottom line it for me. What is the impact of this news specifically in the broader context as well of vaccine development and COVID? We've been able to develop what looks to be like an effective vaccine in less than a year. Traditionally, vaccines would often be measured in decades in terms of when they're developed.

that's a remarkable step forward at the risk of sounding hokey. I think this has been an incredibly inspiring big win for science. We've tried lots of efforts, many in vain, to try to bend the curve. It's possible that these innovations will help end the curve. And I think that is something to be very optimistic about or at least hopeful. All right, you guys. Thank you so much for doing this episode. Thank you.

Thank you.