Pushkin from Pushkin Industries. This is deep background to show where we explore the stories behind the stories in the news. I'm Noah Feldman. As states begin to reopen, we're all desperately trying to figure out what is going to happen next. What are the possible courses of COVID nineteen, what are the possible outcomes if there is a vaccine, and what are the possibilities if indeed we don't get a vaccine or a therapeutic treatment that actually changes things. Joining me to discuss these issues is doctor Jonatan grad. He's an assistant professor in the Department of Immunology and Infectious Diseases at the Harvard chan School of Public Health. He's also an attending physician in the Division of Infectious Diseases at Brigham and Women's Hospital and Harvard Medical School. You know, Tom, thank you so much for being here. I really hugely appreciate it. And the first topic I want to ask you about is a paper on which you're one of the co authors, that was published in Science magazine. That is a model or a series of models, among other things, of what could happen under conditions of reopening if a vaccine is not yet available or potentially not available at all. Would you start by just describing at the core what predictions you and your co authors were able to make about the patterns of covid resurgence that are potentially out there. Sure, there really seems to be only a couple of ways in which a pandemic ends. One of them is elimination of the virus. If you're able to control the spread sufficiently towards the beginning of the virus emergence, you can perhaps contain it to the extent that you can eliminate its ongoing transmission. That I think is what happened with the experience of stars in two thousand to three. I think that this possibility for ending the pandemic now is extremely unlikely, given how globally we've seen the spread of SARS CoV two. The second way to end a pandemic is through population immunity or herd immunity, when enough of the population has acquired immunity to the pathogen that you don't see ongoing epidemic spread. This can happen either with a vaccine that can confer sufficient immune production or through natural infection, where that infection elicits immunity. A key to thinking about HERD immunity, then, is understanding what fraction of the population would have to be infected and recovered from infection with immunity in order for the pandemic to end. That gets to the notion of the basic reproductive number, a sense of just how transmissible the virus is. Estimates of this number for Sarryscoop two place it around three, although there have recently been some estimates that are considerably higher. There was one published last week at five point seven. But let's start with just the notion of this are not or the basic reproductive number of three. What this means is that on average, an infectious person will infect three other people. To prevent these ongoing transmission chains, then two out of three people would have to have immunity, right, so we would have to get to a point in which roughly sixty six percent of the population was immune in order to see that transmission would diminish. We then became interested in asking, given the broad based quarantine for communities that had been introduced through these social distancing measures and lockdowns, both in China and in elsewhere, what the impact would be of these social distancing measure of different durations and different effectiveness. We wanted to know if we were to have lockdowns or social distancing with effectiveness of say twenty percent so mildly effective, forty percent more effective, or sixty percent on the order of what was seen in China, what would happen and what would happen if it were four weeks, eight weeks, twelve weeks, twenty weeks, or for a long period. The key intuition here gets back to what I was describing about the fraction of the population that would have to be immune the extent to which we are successful in preventing spread. So that's the intervention. These lockdowns really diminished transmission, would maintain a susceptible population in the community, so that when you stop that intervention, when you lift the restrictions, now the virus would have the opportunity to spread again through the susceptible population. To the extent that we are successful, we will then see the resurgence of the virus. And that was really one of the main findings in this paper for that one time social distancing interventions. As there are susceptible people, the virus has no memory, it doesn't care what we've done. It just carries if there are susceptible people around that it can infect, and so we would expect to see a resurgence of the virus. So that resurgence is a version of what people call the second wave, and I take it the way you're saying is that that second wave is going to be a lot bigger than perhaps people might assume, given the success of social distancing measures that reduce the number of people who are exposed. So this is sort of the downside of flattening the curve is that fewer people have been exposed, and so when the time comes that they are exposed, potentially a lot of them could be exposed. Imagine a scenario and I think you do imagine this in the paper where then the government responds to that by saying, Okay, we have a second wave coming. We see it rising, we see the number of cases going up, in the number of deaths going up, and so now we're going to reimpose social distancing. Any reason to think that you wouldn't get similar effect of that social distancing in the second time as you did the first time, and then similarly the third and the fourth, and however many times you need to do this, that's right. So it gets to the question of why are we doing social distancing in the first place. If this virus were benign, we would just let it run through the population, right because there wouldn't really be a downside to that. But because this virus causes the extent of disease that it does, we worry about the impact it has on the healthcare infrastructure. So the reason for enacting social distancing and putting communities in lockdown is to not only try to save lives directly, try to diminish the spread so that fewer people are getting infected and dying from the disease, but also from the effect it would have on our healthcare infrastructure. What was seen in Uhan and then replicated in northern Italy and more recently in New York is that when we see this large fraction of the population infected all at once, it overwhelms the healthcare system, and then we see things like rationing ventilators, and also that people who would normally be able to come into hospitals to get care for their heart attacks or strokes were not coming in, so there would not only be death from the virus itself, but from our inability to properly care for people who have the virus and the inability to care for people who have other conditions, we would see extensive excess mortality. That led to this effort to flatten the curve. Once we stop and see a resurgence of virus, as you say, we would expect that if we are still being guided by that principle of trying to maintain the healthcare infrastructure, then to the extent that it threatened by this resurgence, we would imagine reinstituting social distancing measures. This led to our consideration of, well, for how long would we have to do that? Assuming no other intervention, And that's the question I think everybody is so focused on. You know, again, assuming no other magic bullet solution, how many times are we going to have to run this same cycle of up and down and up and down. It looks like from our models that this could take a couple of years of cycles, and that may be influenced by a variety of factors. Those factors include whether we can expand our critical care capacity. So really what we were thinking about in this paper was can we titrate to some extent the turning on and off of social distancing to optimize the number of people were caring for in hospitals or in the healthcare infrastructure, and that really seems keyed to our critical care capacity. So the goal, then, from a intervention perspective, would be to try to find the line, which is how many cases we can handle in our critical care facilities in the hospitals, and then to just bring the number of infections up to that line as close to that line as possible, and then back down again, and to do that as long as it takes, And of course how long it will take depends on how high you can make that line, because that's the question of how many people you can manage to infect. But it sounds like making fairly reasonable assumptions about how high that line could be set. You're suggesting that we could end up having to do this again and again into twenty twenty two. That's right, And it could be that those numbers are adjusted by a variety of different factors. If we find an effective therapeutic that diminishes the need for hospitalization, or if a person is hospitalized that diminishes the need for critical care, that effectively increases our critical care capacity, right, So it's mathematically the same essentially as increasing our capacity, so that can help shorten the duration with which we'd have to go through these cycles. It could also be impacted by seasonality, which is another aspect of the projections that we evaluate in this paper. We don't really know yet whether there is seasonality in transmission of SARS CoV two. For some other respiratory viruses, we know, like, for example, for influenza, that seasonality influences transmission such that it is more transmissible in the winter than it is in the summer. We see as well for a couple of the other human coronaviruses that circulate in the US and cause common cold type symptoms, that it appears they also have seasonality. Using those as a basis for seasonality for SARS CoV two, assuming that whatever factors influence the other coronaviruses would also influence this one, we would see that there would be less transmission in the summer, and then transmission would pick up in the fall and peek in the winter. That could also influence the deboration of lockdown and release from social distancing interventions. So if COVID nineteen does turn out to be seasonal in similarity to these other viruses. That would suggest that we ought to try to reopen going into the summer rather than going into the winter. And that part sounds like good news insofar as most of the openings up that are happening now are going into summer months. That's right, And I think it does raise further concern for the possibility of augmented transmission in the fall and the winter. And I think it's of concern for two reasons. One, it overlaps with the increase in transmission of influenza, which we already know creates stress on our healthcare infrastructure. So to see both flu and COVID nineteen coupled together increasing together, I think that is something that we all worry about. The second reason is actually just one of domestic politics, right. The importance of this election in November is one where seeing a rise in cases and a second wave that overlaps with the election, I think is something that people have to be aware of and be concerned about. So, just to understand what you're saying, what you're saying very politely, but it sounds like what you're saying is you could see a moment where over the summer, as reopenings occur, we actually don't get a huge spike because the virus is more seasonal. But then starts getting colder, people start going inside. It's October, and the case rate might actually begin to go up pretty rapidly, and then boom. We have an election in the first week of November, as mandated by federal statute, and the president may think he can change that, but he can't. So it's going to happen in early November one way or the other. Is it cold enough in October to have this seasonal impact before the election? I mean, I hate to ask such a crudely political question, but it seems relevant. How does flu do or how do these other coronaviruses do in October? That's really the key timing. By the time you get to November, the election will have happened, right. I expect there to be an increase in transmission. The seasonality is not an on off process. It's one that fluctuates with the season. So as we start to move into the fall and winter, I do worry that there will be increased transmissibility and arise in cases. And I think even as this is an issue that we can't know for sure at this point, it is nonetheless one worth preparing for, just in case. It's concerning enough that being able to have the option to vote by mail seems critical for an election where there may be risks as we've seen in Wisconsin and the primaries to standing in line when there are serious respiratory viruses in circulation. There's a fascinating counterintuitive result of this, which is there's been a lot of talk about what Donald Trump tried to delay the elections. If Donald Trump were to listen to this podcast and read your paper, he ought to not, under any circumstances even think about delaying the elections, because if you delay the elections, there might be a lot more COVID nineteen cases as the weather gets colder, Whereas in the first week of November it's entirely possible that that might be going on, but we wouldn't have seen it yet. I think it's so tough to speculate even months out at this point. Frankly, I think with all of the variation in social distancing interventions across the United States, it's very hard to predict what the prevalence will be, and with the risks for recurrence will be by community. Different states and different regions have taken very different approaches to how to manage the response to COVID nineteen. Some seem to have adopted the Boris Johnson phrase take it on the chin. Others are trying to flatten the curve, and yet others are trying to crush the curve. And by crush the curve, what I mean is really prevent any transmission through a combination of social distancing interventions, contact tracing, and other means. So it's hard for me to say exactly what the dynamics will be come September, October, and November, because those in part depend on the fraction of the population that remains susceptible, and that right now is just so hard to predict. We'll be back in a moment you're not on. One of the findings that really struck me in your paper was the finding that if immunity two stars COVID two is not permanent, is likely to enter into regular circulation, sort of like the flu comes around every year. There's something a little terrifying about that because the idea that we'll have stars CoV two with us in this form for a long time is kind of world changing. But hidden in there is a fascinating assumption that I think opened something that I did not understand at all, and I wonder if you would share with us, And that is what does it mean to say that if the immunity is not permanent? I tend to think I think a lot of non physicians tend to think that either you do become immune after you're exposed to something I have antibodies, or you don't become immune. But I'm sensing that that's not the right way to think about it. Immunity, we like to think of, is something that is lifelong, sterilizing protection that you get infected with something, you develop immunity to it, and you are protected against infection with that particular pathogen for the rest of your life. This is one of the concepts we think about for vaccines generally, and it's in some ways how they're pitched. Although you know from your own experience, if you go in for all of your recommended vaccines, as one should, you might have wondered why it is that for some vaccines you get several doses. Right, So every ten years, for example, you're supposed to get an updated tetanus shot right, So it gives a sense even just the fact that that happens, that immunity is not an all or nothing, It is not a one or zero. It is something that can change over time. And also that it is not necessarily as if you have some kind of invincibility shield. We can think of immunity as being in a few different categories of things. And does immunity prevent you from getting infected? Does immunity prevent you from developing severe manifestations of infection should you happen to get infected, And could immunity or different elements of immunity help prevent ongoing transmission so make you less infectious to others? These are also the different ways in which we think about the potential outcomes of vaccination. So for SARS CoV two, what we wondered about was whether immunity might wane over time. There was a study with one of the common cold type coronaviruses where military recruits were exposed to one of these viruses and then a year later exposed again, and they could still become infected, although they had less symptoms. So it seemed as though even if they were able to recover and develop an immune response to the initial infection that let them clear it That immune response was not of sufficient strength and duration that they could not a year later again become infected, although it was for many of them enough to limit the end of symptoms. We see this with many pathogens. One of the reasons why influenza is such the challenge that it is is in part because immunity to flu seems to change. Flu itself seems to change, so the pathogens themselves may evolve in response to the immune pressure from the human populations. We don't know whether this will be the case with sarscoby two. That is yet to be determined as well. So there are a variety of factors at play, both in the dynamics of the human population and individual level immune response, and in the evolution of the pathogen as well. So the takeaway then is that if exposure to the virus creates an immunity that is time bound or that is not capable of completely eliminating the chances that you get the disease later on, but just weakens the case that you get, or if a vaccine does either of those same sorts of things, then we're going to get very possibly an annual COVID scare and that could last. I take it indefinitely, because if we're not talking about total immunity, then herd immunity is in a sense a misnomer. The herd will never be completely immune because no one is, on this theory completely immune. They're just partially or immune in some time bound way. And then it will have to become something that we're going to have to manage going forward without particular end, even if there is a vaccine, again, depending on how the vaccine operates. Yes, so if we get a vaccine, and I want to add a note of caution here that it is not a given. Even with the recent positive news from the phase one of the Maderna trial, it does not seem to me a given that we will have an effective vaccine for sure. I just point out using the example of HIV, a very different virus to be sure, but nonetheless a cautionary tale. We have been trying to get a vaccine for HIV for thirty years and we still don't have one. So just to keep that in mind, I hope we get one for SARS CoV two, and get one quickly. That is highly effective and that does induce long lasting sterilizing immunity. But if we don't, then you're absolutely right. We are going to be looking at dealing with SARS CoV two as another of the seasonal respiratory viruses. It will join the ranks of influenza para influenzas once or for respiratories, and sitial virus, man aneuma virus, the other coronaviruses, rhinovirus. They're a whole panoply of respiratory viruses that cause everything from common cold to very severe outcomes including death, that come through the human population regularly, and it may be that SARS CoV two will join that group. Is there any way to know whether a particular type of community, whether it's naturally occurring or whether it's vaccine induced, will last over time other than waiting over time to see what happens? I mean, is there any way to prefigure that out? Yeah? No, not that I know of. I think it is something where it really takes time to figure it out. Part of it is, you know, what is the robustness of the immune response, and then watching what happens even over short periods of time to see whether antibody levels rapidly decay or are maintained. What am I not asking you? What are major points that you think are not being sufficiently discussed in public right now? About broadly speaking, the issues that we've been talking about. I think that there are a few key points that we haven't discussed that we should be thinking about. One, we've talked about vaccines, but we have not talked about therapeutics. I think therapeutics really give us the best and probably closest off ramp. If we had effective drugs that could diminish the possibility of severe disease, that could really treat and cure infection with stars CoV two, the need for a vaccine would go down tremendously. We could manufacture these drugs, especially if they were oral drugs. Of course, there are all the issues with resistance, and I'll just say the reason that I haven't asked you about that is that we've talked about this extensively on the show on other occasions. Remdesivir does seem to be producing some appealing results, but the numbers that are reported thus far, at least in the NIH study, were not world transforming. It was mortality from eleven point six percent to eight percent, which is meaningful but is not fundamentally transformative. And reduction of time and hospital by some number of days also great, but not formed of an I understand that this could be combined with other anti virals and potentially something could be cobbled together. So or not ruling it out in any way, but it seems as though some of the initial enthusiasm that we would find a therapeutic that would be the magic solution has waned in recent weeks. I think that enthusiasm should needs. It's too easily swayed and endurance and a continued investment is required. We should not expect that our first shot on goal will go in. We will need to continue that investment. If anything, I would say, there's encouragement that we were able to take a drug off the shelf and show that it has some anti viral activity that is clinically meaningful. Investing further in finding other drugs that have more specific activity or that are more clinically effective, I think is hugely important. What about monoclonal antibodies. We had a Kiko Iwasaki from now on the program and she said, you know, no one is paying sufficient attention to this, and she even expressed some puzzlement about why more people aren't focused on monoclonal antibodies. Yeah, I think monoclonal antibodies most likely given in combination, So not just one anti saris COB two antibody, but several that you could administer as a cocktail or in combination with desevere for example. That I think is a super interesting therapeutic angle. I think one challenge with monoclonal antibodies is that they're very expensive right now, and another challenge is that they are hard it seems to manufacture at large scale. So I think that it's an extremely exciting and important therapeutic direction, but I'm not sure that it's the answer for the broadly available therapeutic that would really give us an off ramp. So we should be talking a little bit more about therapies. What else should we be talking about. We should all to be talking about or really trying to get a better handle on the age distribution of infectiousness and susceptibility. How much of a role our kids really playing. I think this remains an open question. We've gotten some data, but we need a lot more. Our kids getting infected to the same extent as the rest of the population. Are they as infectious once infected as the rest of the population. This has big implications for going back to school and how we want our population to re engage and emerge from the hibernation we're in now. Do we try to restructure our communities in significant ways or are we able to let kids go back to school without concern because they don't play a big role in transmission right. I think this is from a variety of perspectives of real important question Clinically, I think there are still questions that are coming up. Why is it that this pathogen is causing hypercoagulable state where we're seeing clocks form both in the venus and arterial circulations. Why is it that we're seeing these unusual inflammatory conditions in some children. There are a variety of clinical questions that we're just really starting to pick up on and engage with. I think those are also going to be hugely important to understand. So there's the pathogenicity of the virus, the biology of the interaction between the host and pathogen, where I think we still have quite a bit to learn, and by understanding those features perhaps develop better interventions to prevent the severe sequila of COVID nineteen. Thank you for those steers. Those are all good stories for us. Continue to watch. Thank you for your clear analysis, and thank you for the work that you're continuing to do in this crisis. Thank you very much, Jonatan, Thank you for the conversation. It's been very fun. So many fascinating things came out of this interview with Yonatan that I tried to keep a running tally, and now looking at my notes, let me try to capture what I walked away with. First, in the absence of a vaccine becoming available very very soon, we're likely to see a recurring pattern of up and down, up and down of COVID nineteen. As the disease recurs, we respond with social distancing, and we lift the social distancing that could last at least into twenty twenty two. Second, COVID nineteen may well turn out to be seasonal, And if it does turn out to be seasonal, that will have a big impact on how the disease comes back and when. In particular, it's likely to recur in the fall, as it begins to get colder. That might turn out to be good news for Donald Trump. Third, depend on how immunity works, whether it's conferred by nature or conferred by a vaccine. It's entirely possible that we may see stars CoV two as actually turning into a permanently recurring disease that comes back again and again and again. Fourth, and perhaps I should have known this already, but I certainly did not. Immunity is not an on off switch. You're not either entirely immune to a disease or entirely not immune to it. Instead, it's a continuum operating on multiple different dimensions. How long it lasts can vary whether it's lifelong or partial. It can affect whether you transmit the virus more easily to another person. It could affect whether you get the disease to a lesser degree than you otherwise would. All of these factors turn out to vary from virus to virus, and from antibody reaction to antibody reaction. Last, but certainly not least, we're not going to know right away whether immunity that is conferred lasts or doesn't, and that will be true whether it's natural immunity or vaccine immunity. The only way to find out how well lasts will be to wait and to see. That's a lot to process and I'm still trying to process it. But as you know, here on Deep Background throughout the Corona Crisis, our goal is to bring you unvarnished the opinions of experts so that our learning process can be shared with you as well. Until the next time we speak, Be careful, be safe, and be well. Deep Background is brought to you by Pushkin Industries. Our producer is Lydia Jane Cott, with research help from zooe Win and mastering by Jason Gambrel and Martin Gonzalez. Our showrunner is Sophie mckibbon. Our theme music is composed by Luis Garat special thanks to the Pushkin Brass, Malcolm Gladwell, Jacob Weisberg, and Mia Lobel. I'm Noah Feldman. I also write a regular column for Bloomberg Opinion, which you can find at Bloomberg dot com slash Feldman. To discover bloomberg original slate of podcasts, go to Bloomberg dot com slash Podcasts. And one last thing. I just wrote a book called The Hour of Winter, a Tragedy. I would be delighted if you checked it out. You can always let me know what you think on Twitter about this episode, or the book or anything else. My handle is Noah R. Feldman. This is deep background

Pushkin from Pushkin Industries. This is deep background to show where we explore the stories behind the stories in the news. I'm Noah Feldman. As states begin to reopen, we're all desperately trying to figure out what is going to happen next. What are the possible courses of COVID nineteen, what are the possible outcomes if there is a vaccine, and what are the possibilities if indeed we don't get a vaccine or a therapeutic treatment that actually changes things. Joining me to discuss these issues is doctor Jonatan grad. He's an assistant professor in the Department of Immunology and Infectious Diseases at the Harvard chan School of Public Health. He's also an attending physician in the Division of Infectious Diseases at Brigham and Women's Hospital and Harvard Medical School. You know, Tom, thank you so much for being here. I really hugely appreciate it. And the first topic I want to ask you about is a paper on which you're one of the co authors, that was published in Science magazine. That is a model or a series of models, among other things, of what could happen under conditions of reopening if a vaccine is not yet available or potentially not available at all. Would you start by just describing at the core what predictions you and your co authors were able to make about the patterns of covid resurgence that are potentially out there. Sure, there really seems to be only a couple of ways in which a pandemic ends. One of them is elimination of the virus. If you're able to control the spread sufficiently towards the beginning of the virus emergence, you can perhaps contain it to the extent that you can eliminate its ongoing transmission. That I think is what happened with the experience of stars in two thousand to three. I think that this possibility for ending the pandemic now is extremely unlikely, given how globally we've seen the spread of SARS CoV two. The second way to end a pandemic is through population immunity or herd immunity, when enough of the population has acquired immunity to the pathogen that you don't see ongoing epidemic spread. This can happen either with a vaccine that can confer sufficient immune production or through natural infection, where that infection elicits immunity. A key to thinking about HERD immunity, then, is understanding what fraction of the population would have to be infected and recovered from infection with immunity in order for the pandemic to end. That gets to the notion of the basic reproductive number, a sense of just how transmissible the virus is. Estimates of this number for Sarryscoop two place it around three, although there have recently been some estimates that are considerably higher. There was one published last week at five point seven. But let's start with just the notion of this are not or the basic reproductive number of three. What this means is that on average, an infectious person will infect three other people. To prevent these ongoing transmission chains, then two out of three people would have to have immunity, right, so we would have to get to a point in which roughly sixty six percent of the population was immune in order to see that transmission would diminish. We then became interested in asking, given the broad based quarantine for communities that had been introduced through these social distancing measures and lockdowns, both in China and in elsewhere, what the impact would be of these social distancing measure of different durations and different effectiveness. We wanted to know if we were to have lockdowns or social distancing with effectiveness of say twenty percent so mildly effective, forty percent more effective, or sixty percent on the order of what was seen in China, what would happen and what would happen if it were four weeks, eight weeks, twelve weeks, twenty weeks, or for a long period. The key intuition here gets back to what I was describing about the fraction of the population that would have to be immune the extent to which we are successful in preventing spread. So that's the intervention. These lockdowns really diminished transmission, would maintain a susceptible population in the community, so that when you stop that intervention, when you lift the restrictions, now the virus would have the opportunity to spread again through the susceptible population. To the extent that we are successful, we will then see the resurgence of the virus. And that was really one of the main findings in this paper for that one time social distancing interventions. As there are susceptible people, the virus has no memory, it doesn't care what we've done. It just carries if there are susceptible people around that it can infect, and so we would expect to see a resurgence of the virus. So that resurgence is a version of what people call the second wave, and I take it the way you're saying is that that second wave is going to be a lot bigger than perhaps people might assume, given the success of social distancing measures that reduce the number of people who are exposed. So this is sort of the downside of flattening the curve is that fewer people have been exposed, and so when the time comes that they are exposed, potentially a lot of them could be exposed. Imagine a scenario and I think you do imagine this in the paper where then the government responds to that by saying, Okay, we have a second wave coming. We see it rising, we see the number of cases going up, in the number of deaths going up, and so now we're going to reimpose social distancing. Any reason to think that you wouldn't get similar effect of that social distancing in the second time as you did the first time, and then similarly the third and the fourth, and however many times you need to do this, that's right. So it gets to the question of why are we doing social distancing in the first place. If this virus were benign, we would just let it run through the population, right because there wouldn't really be a downside to that. But because this virus causes the extent of disease that it does, we worry about the impact it has on the healthcare infrastructure. So the reason for enacting social distancing and putting communities in lockdown is to not only try to save lives directly, try to diminish the spread so that fewer people are getting infected and dying from the disease, but also from the effect it would have on our healthcare infrastructure. What was seen in Uhan and then replicated in northern Italy and more recently in New York is that when we see this large fraction of the population infected all at once, it overwhelms the healthcare system, and then we see things like rationing ventilators, and also that people who would normally be able to come into hospitals to get care for their heart attacks or strokes were not coming in, so there would not only be death from the virus itself, but from our inability to properly care for people who have the virus and the inability to care for people who have other conditions, we would see extensive excess mortality. That led to this effort to flatten the curve. Once we stop and see a resurgence of virus, as you say, we would expect that if we are still being guided by that principle of trying to maintain the healthcare infrastructure, then to the extent that it threatened by this resurgence, we would imagine reinstituting social distancing measures. This led to our consideration of, well, for how long would we have to do that? Assuming no other intervention, And that's the question I think everybody is so focused on. You know, again, assuming no other magic bullet solution, how many times are we going to have to run this same cycle of up and down and up and down. It looks like from our models that this could take a couple of years of cycles, and that may be influenced by a variety of factors. Those factors include whether we can expand our critical care capacity. So really what we were thinking about in this paper was can we titrate to some extent the turning on and off of social distancing to optimize the number of people were caring for in hospitals or in the healthcare infrastructure, and that really seems keyed to our critical care capacity. So the goal, then, from a intervention perspective, would be to try to find the line, which is how many cases we can handle in our critical care facilities in the hospitals, and then to just bring the number of infections up to that line as close to that line as possible, and then back down again, and to do that as long as it takes, And of course how long it will take depends on how high you can make that line, because that's the question of how many people you can manage to infect. But it sounds like making fairly reasonable assumptions about how high that line could be set. You're suggesting that we could end up having to do this again and again into twenty twenty two. That's right, And it could be that those numbers are adjusted by a variety of different factors. If we find an effective therapeutic that diminishes the need for hospitalization, or if a person is hospitalized that diminishes the need for critical care, that effectively increases our critical care capacity, right, So it's mathematically the same essentially as increasing our capacity, so that can help shorten the duration with which we'd have to go through these cycles. It could also be impacted by seasonality, which is another aspect of the projections that we evaluate in this paper. We don't really know yet whether there is seasonality in transmission of SARS CoV two. For some other respiratory viruses, we know, like, for example, for influenza, that seasonality influences transmission such that it is more transmissible in the winter than it is in the summer. We see as well for a couple of the other human coronaviruses that circulate in the US and cause common cold type symptoms, that it appears they also have seasonality. Using those as a basis for seasonality for SARS CoV two, assuming that whatever factors influence the other coronaviruses would also influence this one, we would see that there would be less transmission in the summer, and then transmission would pick up in the fall and peek in the winter. That could also influence the deboration of lockdown and release from social distancing interventions. So if COVID nineteen does turn out to be seasonal in similarity to these other viruses. That would suggest that we ought to try to reopen going into the summer rather than going into the winter. And that part sounds like good news insofar as most of the openings up that are happening now are going into summer months. That's right, And I think it does raise further concern for the possibility of augmented transmission in the fall and the winter. And I think it's of concern for two reasons. One, it overlaps with the increase in transmission of influenza, which we already know creates stress on our healthcare infrastructure. So to see both flu and COVID nineteen coupled together increasing together, I think that is something that we all worry about. The second reason is actually just one of domestic politics, right. The importance of this election in November is one where seeing a rise in cases and a second wave that overlaps with the election, I think is something that people have to be aware of and be concerned about. So, just to understand what you're saying, what you're saying very politely, but it sounds like what you're saying is you could see a moment where over the summer, as reopenings occur, we actually don't get a huge spike because the virus is more seasonal. But then starts getting colder, people start going inside. It's October, and the case rate might actually begin to go up pretty rapidly, and then boom. We have an election in the first week of November, as mandated by federal statute, and the president may think he can change that, but he can't. So it's going to happen in early November one way or the other. Is it cold enough in October to have this seasonal impact before the election? I mean, I hate to ask such a crudely political question, but it seems relevant. How does flu do or how do these other coronaviruses do in October? That's really the key timing. By the time you get to November, the election will have happened, right. I expect there to be an increase in transmission. The seasonality is not an on off process. It's one that fluctuates with the season. So as we start to move into the fall and winter, I do worry that there will be increased transmissibility and arise in cases. And I think even as this is an issue that we can't know for sure at this point, it is nonetheless one worth preparing for, just in case. It's concerning enough that being able to have the option to vote by mail seems critical for an election where there may be risks as we've seen in Wisconsin and the primaries to standing in line when there are serious respiratory viruses in circulation. There's a fascinating counterintuitive result of this, which is there's been a lot of talk about what Donald Trump tried to delay the elections. If Donald Trump were to listen to this podcast and read your paper, he ought to not, under any circumstances even think about delaying the elections, because if you delay the elections, there might be a lot more COVID nineteen cases as the weather gets colder, Whereas in the first week of November it's entirely possible that that might be going on, but we wouldn't have seen it yet. I think it's so tough to speculate even months out at this point. Frankly, I think with all of the variation in social distancing interventions across the United States, it's very hard to predict what the prevalence will be, and with the risks for recurrence will be by community. Different states and different regions have taken very different approaches to how to manage the response to COVID nineteen. Some seem to have adopted the Boris Johnson phrase take it on the chin. Others are trying to flatten the curve, and yet others are trying to crush the curve. And by crush the curve, what I mean is really prevent any transmission through a combination of social distancing interventions, contact tracing, and other means. So it's hard for me to say exactly what the dynamics will be come September, October, and November, because those in part depend on the fraction of the population that remains susceptible, and that right now is just so hard to predict. We'll be back in a moment you're not on. One of the findings that really struck me in your paper was the finding that if immunity two stars COVID two is not permanent, is likely to enter into regular circulation, sort of like the flu comes around every year. There's something a little terrifying about that because the idea that we'll have stars CoV two with us in this form for a long time is kind of world changing. But hidden in there is a fascinating assumption that I think opened something that I did not understand at all, and I wonder if you would share with us, And that is what does it mean to say that if the immunity is not permanent? I tend to think I think a lot of non physicians tend to think that either you do become immune after you're exposed to something I have antibodies, or you don't become immune. But I'm sensing that that's not the right way to think about it. Immunity, we like to think of, is something that is lifelong, sterilizing protection that you get infected with something, you develop immunity to it, and you are protected against infection with that particular pathogen for the rest of your life. This is one of the concepts we think about for vaccines generally, and it's in some ways how they're pitched. Although you know from your own experience, if you go in for all of your recommended vaccines, as one should, you might have wondered why it is that for some vaccines you get several doses. Right, So every ten years, for example, you're supposed to get an updated tetanus shot right, So it gives a sense even just the fact that that happens, that immunity is not an all or nothing, It is not a one or zero. It is something that can change over time. And also that it is not necessarily as if you have some kind of invincibility shield. We can think of immunity as being in a few different categories of things. And does immunity prevent you from getting infected? Does immunity prevent you from developing severe manifestations of infection should you happen to get infected, And could immunity or different elements of immunity help prevent ongoing transmission so make you less infectious to others? These are also the different ways in which we think about the potential outcomes of vaccination. So for SARS CoV two, what we wondered about was whether immunity might wane over time. There was a study with one of the common cold type coronaviruses where military recruits were exposed to one of these viruses and then a year later exposed again, and they could still become infected, although they had less symptoms. So it seemed as though even if they were able to recover and develop an immune response to the initial infection that let them clear it That immune response was not of sufficient strength and duration that they could not a year later again become infected, although it was for many of them enough to limit the end of symptoms. We see this with many pathogens. One of the reasons why influenza is such the challenge that it is is in part because immunity to flu seems to change. Flu itself seems to change, so the pathogens themselves may evolve in response to the immune pressure from the human populations. We don't know whether this will be the case with sarscoby two. That is yet to be determined as well. So there are a variety of factors at play, both in the dynamics of the human population and individual level immune response, and in the evolution of the pathogen as well. So the takeaway then is that if exposure to the virus creates an immunity that is time bound or that is not capable of completely eliminating the chances that you get the disease later on, but just weakens the case that you get, or if a vaccine does either of those same sorts of things, then we're going to get very possibly an annual COVID scare and that could last. I take it indefinitely, because if we're not talking about total immunity, then herd immunity is in a sense a misnomer. The herd will never be completely immune because no one is, on this theory completely immune. They're just partially or immune in some time bound way. And then it will have to become something that we're going to have to manage going forward without particular end, even if there is a vaccine, again, depending on how the vaccine operates. Yes, so if we get a vaccine, and I want to add a note of caution here that it is not a given. Even with the recent positive news from the phase one of the Maderna trial, it does not seem to me a given that we will have an effective vaccine for sure. I just point out using the example of HIV, a very different virus to be sure, but nonetheless a cautionary tale. We have been trying to get a vaccine for HIV for thirty years and we still don't have one. So just to keep that in mind, I hope we get one for SARS CoV two, and get one quickly. That is highly effective and that does induce long lasting sterilizing immunity. But if we don't, then you're absolutely right. We are going to be looking at dealing with SARS CoV two as another of the seasonal respiratory viruses. It will join the ranks of influenza para influenzas once or for respiratories, and sitial virus, man aneuma virus, the other coronaviruses, rhinovirus. They're a whole panoply of respiratory viruses that cause everything from common cold to very severe outcomes including death, that come through the human population regularly, and it may be that SARS CoV two will join that group. Is there any way to know whether a particular type of community, whether it's naturally occurring or whether it's vaccine induced, will last over time other than waiting over time to see what happens? I mean, is there any way to prefigure that out? Yeah? No, not that I know of. I think it is something where it really takes time to figure it out. Part of it is, you know, what is the robustness of the immune response, and then watching what happens even over short periods of time to see whether antibody levels rapidly decay or are maintained. What am I not asking you? What are major points that you think are not being sufficiently discussed in public right now? About broadly speaking, the issues that we've been talking about. I think that there are a few key points that we haven't discussed that we should be thinking about. One, we've talked about vaccines, but we have not talked about therapeutics. I think therapeutics really give us the best and probably closest off ramp. If we had effective drugs that could diminish the possibility of severe disease, that could really treat and cure infection with stars CoV two, the need for a vaccine would go down tremendously. We could manufacture these drugs, especially if they were oral drugs. Of course, there are all the issues with resistance, and I'll just say the reason that I haven't asked you about that is that we've talked about this extensively on the show on other occasions. Remdesivir does seem to be producing some appealing results, but the numbers that are reported thus far, at least in the NIH study, were not world transforming. It was mortality from eleven point six percent to eight percent, which is meaningful but is not fundamentally transformative. And reduction of time and hospital by some number of days also great, but not formed of an I understand that this could be combined with other anti virals and potentially something could be cobbled together. So or not ruling it out in any way, but it seems as though some of the initial enthusiasm that we would find a therapeutic that would be the magic solution has waned in recent weeks. I think that enthusiasm should needs. It's too easily swayed and endurance and a continued investment is required. We should not expect that our first shot on goal will go in. We will need to continue that investment. If anything, I would say, there's encouragement that we were able to take a drug off the shelf and show that it has some anti viral activity that is clinically meaningful. Investing further in finding other drugs that have more specific activity or that are more clinically effective, I think is hugely important. What about monoclonal antibodies. We had a Kiko Iwasaki from now on the program and she said, you know, no one is paying sufficient attention to this, and she even expressed some puzzlement about why more people aren't focused on monoclonal antibodies. Yeah, I think monoclonal antibodies most likely given in combination, So not just one anti saris COB two antibody, but several that you could administer as a cocktail or in combination with desevere for example. That I think is a super interesting therapeutic angle. I think one challenge with monoclonal antibodies is that they're very expensive right now, and another challenge is that they are hard it seems to manufacture at large scale. So I think that it's an extremely exciting and important therapeutic direction, but I'm not sure that it's the answer for the broadly available therapeutic that would really give us an off ramp. So we should be talking a little bit more about therapies. What else should we be talking about. We should all to be talking about or really trying to get a better handle on the age distribution of infectiousness and susceptibility. How much of a role our kids really playing. I think this remains an open question. We've gotten some data, but we need a lot more. Our kids getting infected to the same extent as the rest of the population. Are they as infectious once infected as the rest of the population. This has big implications for going back to school and how we want our population to re engage and emerge from the hibernation we're in now. Do we try to restructure our communities in significant ways or are we able to let kids go back to school without concern because they don't play a big role in transmission right. I think this is from a variety of perspectives of real important question Clinically, I think there are still questions that are coming up. Why is it that this pathogen is causing hypercoagulable state where we're seeing clocks form both in the venus and arterial circulations. Why is it that we're seeing these unusual inflammatory conditions in some children. There are a variety of clinical questions that we're just really starting to pick up on and engage with. I think those are also going to be hugely important to understand. So there's the pathogenicity of the virus, the biology of the interaction between the host and pathogen, where I think we still have quite a bit to learn, and by understanding those features perhaps develop better interventions to prevent the severe sequila of COVID nineteen. Thank you for those steers. Those are all good stories for us. Continue to watch. Thank you for your clear analysis, and thank you for the work that you're continuing to do in this crisis. Thank you very much, Jonatan, Thank you for the conversation. It's been very fun. So many fascinating things came out of this interview with Yonatan that I tried to keep a running tally, and now looking at my notes, let me try to capture what I walked away with. First, in the absence of a vaccine becoming available very very soon, we're likely to see a recurring pattern of up and down, up and down of COVID nineteen. As the disease recurs, we respond with social distancing, and we lift the social distancing that could last at least into twenty twenty two. Second, COVID nineteen may well turn out to be seasonal, And if it does turn out to be seasonal, that will have a big impact on how the disease comes back and when. In particular, it's likely to recur in the fall, as it begins to get colder. That might turn out to be good news for Donald Trump. Third, depend on how immunity works, whether it's conferred by nature or conferred by a vaccine. It's entirely possible that we may see stars CoV two as actually turning into a permanently recurring disease that comes back again and again and again. Fourth, and perhaps I should have known this already, but I certainly did not. Immunity is not an on off switch. You're not either entirely immune to a disease or entirely not immune to it. Instead, it's a continuum operating on multiple different dimensions. How long it lasts can vary whether it's lifelong or partial. It can affect whether you transmit the virus more easily to another person. It could affect whether you get the disease to a lesser degree than you otherwise would. All of these factors turn out to vary from virus to virus, and from antibody reaction to antibody reaction. Last, but certainly not least, we're not going to know right away whether immunity that is conferred lasts or doesn't, and that will be true whether it's natural immunity or vaccine immunity. The only way to find out how well lasts will be to wait and to see. That's a lot to process and I'm still trying to process it. But as you know, here on Deep Background throughout the Corona Crisis, our goal is to bring you unvarnished the opinions of experts so that our learning process can be shared with you as well. Until the next time we speak, Be careful, be safe, and be well. Deep Background is brought to you by Pushkin Industries. Our producer is Lydia Jane Cott, with research help from zooe Win and mastering by Jason Gambrel and Martin Gonzalez. Our showrunner is Sophie mckibbon. Our theme music is composed by Luis Garat special thanks to the Pushkin Brass, Malcolm Gladwell, Jacob Weisberg, and Mia Lobel. I'm Noah Feldman. I also write a regular column for Bloomberg Opinion, which you can find at Bloomberg dot com slash Feldman. To discover bloomberg original slate of podcasts, go to Bloomberg dot com slash Podcasts. And one last thing. I just wrote a book called The Hour of Winter, a Tragedy. I would be delighted if you checked it out. You can always let me know what you think on Twitter about this episode, or the book or anything else. My handle is Noah R. Feldman. This is deep background