Coronavirus (COVID-19): Press Conference with Yonatan Grad, 05/07/20

You’re listening to a press conference from the Harvard T.H. Chan School of Public Health with Yonatan Grad, the Melvin J. And Geraldine L Glimcher Assistant Professor of Immunology and Infectious Diseases. This call was recorded at 11:30 a.m. Eastern Time on Thursday, May 7.

Previous press conferences are linked at the bottom of this transcript.


YONATAN GRAD: Hello, everyone. One of the things I wanted to mention is preprint that my colleagues, Daniel Larremore and Kate Bubar from Colorado just posted earlier today. I’m just going to put it into the chat. There’s a link there in which in this short letter we think about the impact of test characteristics and populations or prevalence on the immune passport idea. And basically the question we’re interested in here was when you have an imperfect test, something that doesn’t have perfect specificity, there will be a false positive rate. And that has implications for the immune past, for strategy, for opening up the economy in a couple of ways.

First, people who are false positives will enter into the population, assuming that they are immune. And then if they get infected, they will wonder what’s going on. How is it that someone who is supposedly immune could get infected again? And so this really can set up some kind of tension, since there is so much concern right now around whether immunity actually or evidence of antibodies indicates any – indicates immunity, you can really make more problematic that situation.

And then secondarily, again, assuming that serological positivity does correlate with with immunity, the false positive rate is dependent on the overall prevalence. So given an imperfect specificity, when you have low prevalence, the positive predictive value for the test is also low and that it could be that in fact, given sensitivity and specificity characteristics for the test, you could be in a situation where if those are poor enough, you’re actually introducing into the population people who you think are positive, but enough of them are actually negative that you could be below the herd immunity threshold.

So there’s an importance in thinking about seroprevalence as well as getting tests that are as good as possible if you want to pursue the Indian passport strategy. So again, there’s a link to that preprint that goes through it in more detail. And I think probably more eloquently than the brief summary that I tried to provide just now and I’m happy to handle questions on those items that Nicole mentioned. And if there are other things that can address, I’m happy to give those a shot too.

MODERATOR: Great. Thank you. First question.

Q:  I’m asking one you may not be able to answer, but I’m interested in finding out how important it is to know about testing, testing rates by all areas like a city or town or zip code. How important is that in terms of reopening, in terms of focusing your efforts on dealing with the pandemic. Right now in Massachusetts, you have total tests by state. We do not have any data on total tests by anything below the geography of the state.

YONATAN GRAD: Broadly speaking, I think having – okay, so we need testing for a couple of reasons.

First, we need virologic testing or basically testing to see if people are infected to help us understand where we are on the curve. And also to aid in that we need serologic testing, testing that informs on what fraction of the population has been infected. So a combination of both serologic and virologic testing. When we talk about testing, we need both things just to be clear about that. Now we need it to monitor not only where we have been, but where we are and the responses to different types of interventions. So when we start lifting the broad lockdown, for example, what happens? We need to be able to monitor.

How much does it matter, that testing at a state level or a local level, that gets to something that I think is quite important about this pandemic, and that is the pandemic is actually made up of very local epidemics. It isn’t one curve or at least we shouldn’t really think about it as just one curve for the world or for our country or even for a state. I think it’s important in terms of tailoring our responses to understand what the local trajectory is. We see that in Boston with evidence of what’s going on in Chelsea. So there was one study so far and I know more are underway that did really a pilot study, but it showed something like 30 percent roughly of the people tested in Chelsea showed evidence of past infection. So this is one of those serological studies that looked at antibodies to SARS-CoV-2. That is a high rate compared to what we believe is going on in many other communities across Massachusetts. So understanding the local prevalence, and so in other words, how many people have the disease locally, I think is quite important. So testing to monitor at a local level I think is going to be needed to help to tailor the responses and not only evaluate how current approaches to opening the economy go, but when we need to reintroduce social distancing or other types of mitigation efforts, can we do a better job of tailoring those interventions to the most at risk regions?

So rather than having a blanket and state-wide lockdown, would there be an improved strategy by having higher resolution, more granular understanding in communities or risk groups within which neighborhoods? So, getting back to the question, I would say it would certainly be helpful to be able to understand through testing at local levels what’s going on and what is and to be able to monitor and then tailor our responses.

Q: That wasn’t exactly an answer to my question, but I understand. I guess my question was how do you know what the prevalence of testing itself is at a local level?

YONATAN GRAD: How do you know?

Q: How do you know that? Some states that have that that information and they report it. Some state report it by county. Some states report it by zip code. But Massachusetts is not, I think from what I’m gathering from.

YONATAN GRAD: Right. I can’t speak to what that state is doing in terms of reporting. That’s that’s for the state to describe. I can tell you, it’s important to know and would be helpful, but I can’t tell you what’s guiding their decisions about what the overall prevalence of testing is important.

Q: So reporting at a local level is important and people in the community are actually [Inaudible]? 

YONATAN GRAD: I think monitoring at a community level is important. Yes.

Q: OK. Thank you very much. I appreciate that.

YONATAN GRAD: You’re welcome.

MODERATOR: Next question.

Q: Thanks for taking my question. This week, President Trump was asked whether workers returning to their jobs could get testy. Both antibody and viral testing. And he responded. They should have no problem. So my question is, can everyday workers get these tests now, particularly that antibody test? It’s one of the focus of this call. And if so, how does that work? What would they need to do to get antibody tests? And if not, will we see this happen where workers returning to work can go get a test and when?

YONATAN GRAD: I think the answer to the first part of your question, can people, can anyone get tested? The answer is quite clearly no. I don’t think that there’s, that we have that testing capacity, either virologic or serologic, to test everyone. It just doesn’t exist yet. I know many places are working to ramp it up, but it has been it has been quite difficult. Secondly, on the serologic testing, there are still – there are many tests out there of unclear sensitivity and specificity.

There are many, many people are working to characterize these tests that are out on the market and see how well they perform. And there are other people who are trying to make available essentially homebrew tests who are trying to take them the type of ELISA assay that was developed by Florian Kramer at Mount Sinai and develop their own version and and try to use robots and other types of methods to expand capacity. So there’s a huge amount of effort right now going into both trying to validate the tests that are flooding the market and trying to develop new ones. I say flooding the market, but in fact, test availability really isn’t that high yet.

So we don’t yet have the capacity to offer broadly either virologic or serologic testing. Will it happen? I certainly hope so. And again, I know many people are working hard at trying to scale up this kind of testing, but it’s essentially trying to create an infrastructure very, very rapidly where one for this test in particular hasn’t really existed. So and I think it will be extremely important to look to use these tests and to look broadly at the population. But we just don’t have the testing yet.

Q: OK. Just a quick follow up. I mean, any sense of how long it might take to get to the point where we have that kind of broad level of testing? Or do we just don’t know yet?

YONATAN GRAD: We don’t know yet because, you know, it depends on a wide variety of factors. In the same way that it has been very hard to predict the extent to which virologic testing can can be scaled up. Right. I don’t know that people would have anticipated we’d hit this plateau in the same way. So it’s it’s very hard for me to predict what the timeline will be.

MODERATOR: Next question.

Q: Hi, Dr. Grad. So I have two questions for this about the reopening policy. After the reopening, it’s  very important for the federal and state government to monitor the data and be ready to reintroduce new measures if needed. But my my question is, what’s a trigger of re-restricting? Do we need criteria or system for the whole country that we should be re-locked down or any new policy?

YONATAN GRAD: OK. So you’d mentioned two questions, but it sounded like one question. Was that just the first question and a follow up with the second one?

Q: Sorry?

YONATAN GRAD: You asked one question, but you introduced yourself by saying that you would have two questions.

Q: Yeah. I wonder if I could measure the second question after this.

YONATAN GRAD: OK. No problem. And so the first question, this question about what is the trigger for reintroducing mitigation efforts.

So, as we start to lift social distancing and other interventions that we should expect to see a resurgence in cases. This is the math of epidemics, that as there are susceptible people, if we have not hit herd immunity, we would expect to see a rise in cases. So to the point of what is the trigger, once we start to see this rise of re-instituting interventions and then what type of – I’ll extend the question and say what type of interventions? What puts the expectation for how those interventions will work? I think that’s an extremely important question and one that I have not seen well addressed.

And I think, you know what we’ll need to consider in developing the strategies for response to a resurgence includes better understanding of how successful the strategies we’ve used so far have been. Can we tailor those interventions as we spoke about earlier on the call, instead of using blanket kind of blunt instrument across a whole geographic region or a whole state, can we tailor the response particularly to those communities where we’re seeing a rise in cases? Can we use more refined measures? Can we combine social distancing of varying levels with contact tracing as well as quarantine and isolation? All of these – trying to figure out how to balance the types of interventions that we have available to us with the context in which we’re seeing a resurgence is going to be critical. And yes, having extensive testing, to pick up on another point already mentioned on the call, is necessary to really be able to monitor for the resurgence, to be able to then institute and have a trigger that will prompt us to reinstitute some type of mitigation.

So, you know, I haven’t seen many people describe what those triggers will be, but I absolutely agree that it will be important to have a strategy for when we start to see a resurgence.

Q: Well, thank you. Thanks a lot. And my second question is not related to today’s topic. It’s about epidemic models that we always need to report, different kinds of models that they have different results. So what factors should we focus on when reporting or reviewing these models or in other words, how can we say a model is accurate, is useful or not?

YONATAN GRAD: I think there had been a fair amount of discussion about models and where models are useful and and how to think about models, and that that could be an entire course, not even an hour long discussion or even a short answer. So I may refer you to some recent discussions about the utility and interpretation of models rather than try to try to tackle that one of my own, other than to say it’s important to keep in mind that that models for epidemic’s are intended to be projections to help inform thinking and decision making.

They aren’t intended to be forecasts in the same way as you would forecast the weather. With a weather forecast trying to anticipate at, for example, when and where a hurricane might hit, there is little that we do to change those parameters, meaning the hurricane does not respond to our actions. For all of our modeling efforts, it’s really important to keep in mind that they are these models and the projections from these models are dependent on what we do, right. So they aren’t so much forecasts of what will happen. They should be viewed as tools to help inform our intuition for what might happen under different conditions or different types of interventions.

Q: Thanks a lot. 

MODERATOR: Great. Next question.

Q :I also have a couple of questions. The first is the one that I sent over about – so New York City is seeing looks like two thirds of their new admissions to hospitals are people who say that they were shut in, that they were staying home. And nonetheless, there they are. In Massachusetts, we’re recently seeing an uptick, actually just a day or two. But do you have any explanation for what’s going on?

YONATAN GRAD: So I haven’t seen those data. I did see your question just before coming on this call, but I haven’t seen the report, so I can’t speak to the specifics of that report. But what I can say is that households transmission is going to be the way in which transmission takes place when most people are at home and there isn’t opportunity for community transmission. So how does that happen? Well, it’s not that the virus is coming in through the walls. It’s almost certainly the case that it’s from other members of the household who may not be shut in. So from other people who are still out in the community.

What I’d be curious about is in what communities are we seeing cases appear and what is the structure of the households in which these people who are presenting with infections are coming from. So if it’s the case that they’re from communities with high prevalence and from households where there are still members of those households who are working or going out because they’re essential workers or have other responsibilities that take them outside of the house, then it could be the case that people are – that this is just evidence of households transmission and even the people who are trying to stay at home are getting exposed by other people who are in their homes or in their buildings.

So I think it’s gonna be important to look at what the household density and structure are in those communities where we’re seeing households transmission. And to what extent do we still see the people in those communities or in those households, buildings are still mobile are still going out of the house. So, you know, to to the point that risk of infection has to do with not just what you’re doing, but what the people around you are doing.

Q: Yeah, that totally makes sense. Thank you. And my other question is about serologic testing. There are quite a few serological tests that are being offered to consumers now, and there is certainly quite a few consumers who felt crappy two months ago or one month ago or three months ago and would be interested probably in buying them. What would be your advice to people who are considering at the individual consumer level getting serologic testing and possibly even businesses too, I would imagine? What what what would you say to those people?

YONATAN GRAD: Right now I would say wait until you have good confidence in your test. I think that there are many tests out there that would fall in that category that I was describing earlier of being, of not high enough specificity such that in, particularly in communities where there is low prevalence, that false positive rate is going to be high. Right. So they would have low positive value.

And additionally, we still don’t know the extent to which antibodies in exposure to SARS-CoV-2 indicate immunity. So one of the reasons why people are so enthusiastic, I think, in taking these tests is the hope that it would indicate that they are immune privileged, that they have had the infection and now they’re immune and they can reenter the communities, but we don’t – while it’s likely the case, we don’t know that for sure. And because there are false positive rates for these tests, I would hesitate to trust the information and to act on it. So I think it behooves people to understand that many of the tests out there are not great tests.

There was a New York Times piece on this actually work done by a lab at Mass General and the lab at University of California, San Francisco or UCSF, looking at a variety of these tests and showing that many of them had poor test characteristics. So not only are there issues with the tests themselves, but there is an issue with the interpretation that I would caution people about.

Q: So then quick follow up. So then the the idea of immune passports is actually premature, very premature at this point?

YONATAN GRAD: Yes, I would say so. And indeed, there are a number of other even social concerns around using immune passports. And I would point you to a very nice piece in The Lancet written by Alexandra Phelan, P-h-e-l-a-n, recently on a number of the concerns around immune passports. And then I think we we cited, I put up in the chat a link to the preprint that we posted. I think it was yesterday, but it has a number of references to these issues in there as well. So but I think, you know, immune passports are a little premature and are problematic for a variety of reasons, both from the test side and then also from more ethics and sociology.

MODERATOR: Thank you. Next question.

Q: So could you comment on how the antibody tests so far, aside from being not perfectly accurate, are essentially binary? In other words, don’t we really need to know how many neutralizing antibodies someone has? And then I guess secondarily, isn’t it also true that antibodies aren’t necessarily the whole story, like how much are killer t cells is likely to be part of the necessary immune response?

YONATAN GRAD: Alright. So there are a couple of factors here. So immune responses are dynamic, so they take time to appear and then over time they wane. So we know this from a variety of other infections and it is the case here as well. So there was recently some work by Florian Kramer and his group at Mount Sinai, where I believe that they had a preprint on medRxiv that came out earlier this week on this topic where they followed people after their diagnosis. So they’re known by the virus test to have been infected. And then they looked to see what happened over time.

So we see that, to your point about, you know, binary and not binary, many of the lateral flow assays, the kinds of tests that are basically like pregnancy tests – you know, does the line appear or does it not – are interpreted as binary. And I think those are perhaps best understood as indicating whether you have exposure. What we can say is that over time, those will, you know, the likelihood goes up. Right. So the immunoglobulins, the antibodies themselves, will increase over time. And that’s what Florian and his crew showed. And again, we know from other types of infections. So the longer you wait after an infection and recovery, the more likely you’re going to see what positive signal. Basically, the longer you wait after an infection and recovery, the more likely you’ll see the antibodies, right. They’ll grow, right, or they’ll increase in concentration. So there’ll be more.

Q: Yeah. Yeah.

YONATAN GRAD: So we see that. Right. And that’s something that that Florian and his crew showed for SARS-CoV-2 as well. So getting to your question about neutralizing antibodies and the type of immune response, finding the immune correlates of protection, I think is always, you know, it’s a challenge for any type of infection and trying to say, you know, do you have enough antibodies to generate an immune response that, again, is binary rising what is probably a non binary relationship. So, for other types of infections, you know, you can imagine that the amount of virus that people are exposed to will vary. Right. So, you know, that the immune response or the extent to which you can be protected may be related to the amount of the pathogen that you’re exposed to.

So given some low amounts, perhaps, you know, your immune response can protect you. But given a very high exposure, perhaps the immune response would be overwhelmed. Right. So there is probably some relationship there for for most pathogens and SARS-CoV-2 as well.

So this gets to know about while we try to draw a line and say above some number you’re protected, below some number, you’re at risk, it probably is the case that that’s a simplification as well. It’s often a useful one because it may be the case that the exposure tends to be of a particular amount. Right. So if you know that the amount of virus you get exposed to is within some amount, some concentration, then having an idea of this, the relationship between protection and not protection will be useful. But if the amount of virus you get exposed to varies widely, then it may be harder to infer correlates of protection.

Q: OK. But so is there any evidence yet that perhaps neutralizing antibodies are perhaps only part of the story and that you really need, you know, the so-called cellular T-cell prefund to be part of it? I mean, I know that different with different viruses, like how much each one is important.

YONATAN GRAD: Yeah, I haven’t seen – I don’t know that people have really characterized what an effective immune response that clears the infection. What exactly is necessary? So I haven’t  – I don’t know. I haven’t seen that really characterizes what a fully effective immune response or an effective immune response looks like yet.

Q: Thank you very much.

MODERATOR: OK. Next question.

Q: Hey, thanks for doing this. I’ve got to wonder, though, wouldn’t it be a near-certainty that you get immunity since that’s normally the case. I mean, can you cite viruses where it’s not the case? Are there many of them? And secondly, and I have a follow up after this. Secondly, if we don’t get immune from the exposure to the virus, then how on earth could a vaccine help? Which is kind of a modified version of exposure to the virus. I mean, why do we even have any hope for that?

YONATAN GRAD: Right. So I think it is certainly the case that we expect the people who recover do so because they mount an effective immune response. And so I would say, at least in the short term, the expectation is that people will have some measure of protection. It is not clear whether many of these antibody tests are really measuring neutralizing antibodies. So we don’t know how much protection now we should expect people to have on the basis of many of these tests yet.

Similarly, we don’t know how long or how good that immune response is. So, for example, in one of the alpha coronaviruses, one of that seasonal coronaviruses that causes common cold like symptoms, there was a study in military recruits, as I recall, where they basically got infected and then were challenged with the same strain a year later and many of them were able to be infected again. Right. So it suggests that at least in those cases, the duration of protection from whatever response was generated was fairly short lasting. For other infections, we know that the immune response generated is quite long. For measles, for example, we think that it is lifelong or nearly lifelong. So, yes, I think it it varies by virus. And for some of the other viruses in coronavirus family, it can be apparently short. So I think that’s one of the concerns.

For a vaccine, you’re absolutely right. You know, the ability to come up with a vaccine that generates, that we believe would generate is an immune response, you know, if you can’t come up with an immune response to the infection itself, it absolutely raises the question, can a vaccine do it? And here I’m getting into territory that I’m not an expert in. I’m not a vaccinologists or an immunologist at that level to be able to say whether any of the new vaccine strategies offers promise whether to this point specifically. And if for some vaccines you can actually generate a stronger immune response then might happen with natural infection. That’s out of my realm of expertise. But I think that the question is an important one. And so what is the extent of protection?

I’ll add another question to yours. How much does the variation in symptoms correlate with the variation in extent or duration of immune protection? How much do we need to worry that people who are asymptomaticly infected may develop a mild immune response so enough to clear perhaps their infection but again, not generating a response that is long lasting or particularly strong? I think that’s unanswered. But one that that that I worry about. So I think there are still many questions around immunity that remain to be addressed.

Q: OK. So it’s complicated.

YONATAN GRAD: In a word, yes.

Q: So you sort of answered this before. But I want to ask again for those of us who had a terrible flu experience and I think everybody else as well. When do you think there’ll be a kind of fairly simple like HIV, like fingerprint test that will tell us it doesn’t have to be right away? But, you know, because right now, the best one seems to require a blood draw. And there’s no way in the world I’m going away anywhere near a phlebotomist or hospital at this moment, but like an HIV type test, like an easy one, when would that be available?

YONATAN GRAD: Right. So I think there are some tests that have been demonstrated in principle and in pilot studies to be good from just using a finger prick. And many of these lateral flow assays that I was mentioning before are based on just a drop of blood. So really just using a lancet, one of these little – what you see people use for checking their blood sugar, for example. And so those tests exist and there are some that appear to be quite good and there are others that are not as good. I know people are working very hard on making these widely available. They aren’t yet, but the technology platforms exist and they’re starting to be used in serological studies of communities.

So, for example, the study that I mentioned earlier of Chelsea here in the Boston area, run by a group at Massachusetts General Hospital, used a lateral flow assay and they had characterized the test characteristics for that assay, basically showing that it works really well, well enough to they could believe the answers with a fair degree of confidence. So and that was just that is based on just essentially a drop of blood from a finger prick. So that so that technology platforms exist, but they aren’t yet available at the scale that we need.

Q: Alright. Thank you.

MODERATOR: And I’m going to ask a quick question myself here. So we’re talking about immunity to certain viruses and that you can have lifelong or very long lasting immunity against measles, but not so much against other viruses such as some coronaviruses. What is the difference between those viruses and what causes the different immune response life?

YONATAN GRAD: Everyone wishes they had a good answer to that. It’s a great question and one that is not clear. If we knew an answer to that, I think we’d be able to come up with better vaccines across the board. Right. You wouldn’t need to get a booster shot for your tetanus and diptheria, for pertussis, for example. You’re supposed to get t death updated every 10 years. Or maybe we won’t need to if we really knew an answer to that question and to generate with vaccines that are very long lasting. So, yeah, great question. Still a very active topic of research.

MODERATOR: OK. So it’s not that the viruses themselves or whatever the pathogen itself is mutating and that’s why the vaccines are no longer working. It’s something else about the immune system response.

YONATAN GRAD: OK. So that gets to another point. So, yes. So some viruses do change antigenicly. And so an immune response to one strain may not provide the same level of immunity or potentially even any immunity to another strain. So with influenza, for example, we know there is both antigenic terms or antigenic drift and antigenic shift. So where new flu strains are emerging regularly and those are evolving in response to population immunity, so basically the selective pressure from the immune system is actually driving their evolution to escape immune pressure. So it can also depend on the virus and its adaptation to immune pressure.

MODERATOR: Thank you. Next question.

Q: Oh, hi. So, I have a couple of retrospective questions I will ask go one after the other. So, they say that the immune responses are different depending on the strain of the coronavirus. I was wondering whether it would be possible in conducting intensive research to use the previous SARS outbreak, 2002, 2003, and basically studying the immune response of those who survived the outbreak to kind of create a serological test that could have some chance to work in the face of a wider variety of coronavirus strains? And would it be possible to develop also detection or viral test that could also potentially be used to detect and make the yardsticks of any given virus?

Of course, that will be very difficult to find a viral test and serological test, but at least having something already made to avoid basically making some efforts onto a pandemic and have something prepared to be done. Or is it a science fiction question? That’s the first question.

YONATAN GRAD: OK, so just to tackle that one quickly. So a couple of points about the prior, so SARS-CoV-1, and some things that we knew about it. So there are a couple of interesting things.

So first, the immune response, at least in terms of the antibody titers for SARS-CoV-1, people were monitored after their exposure and recovery. And you know, I’d mentioned that for the Alpha coronavirus, you know, it seemed like immunity by challenge to be at least  could last you know under a year. And by looking at the antibody levels, we saw that those waned for SARS-CoV-1 after around three years. So it wasn’t – the antibody levels themselves were not necessarily extremely long lasting on average. So there’s some information there.

Another interesting thing is about cross reactivity. So it seems like people who had some of the common cold associated coronaviruses, their antibodies wouldn’t really protect against SARS CoV-1 infection but the other way around seems to be the case, where people who had SARS-CoV-1, their antibodies would actually protect against infection with the common cold coronaviruses. So the relationship of immunity was not  not an equal one. So and we don’t know what the situation will be with SARS-CoV-2. But it’s a super interesting question.

For serologic tests, you know, as we’re saying, you’re trying to imagine, could we have developed a test for all coronaviruses or that would capture all coronaviruses? I think even before the emergence of SARS-CoV-2, I think that’s quite difficult because we don’t know what’s that immune response would be. We don’t know what it would look like. And the point – one of the challenges with these serological tests is actually to distinguish among the responses to the different coronaviruses by having one test for all coronaviruses and isn’t particularly useful because we expect most people will have been exposed to the common cold coronaviruses which circulate every year.

So what you really want is not a common test, but one that is specific to the particular pathogen of interest. And this this the cross reactivity of the various diagnostics is in fact exactly one of the issues that we worry. That’s the diagnostics like these lateral flow assays that I was mentioning before, that their specificity may be limited because perhaps they’re finding antibodies that are to some of the common cold coronaviruses instead of antibodies that are specific for SARS-CoV-2. So yeah, so that is my answer to your science fiction question.

Q: The second question is a follow up to what you said about the fact that some family members are infecting others because they’re exposed to contagion from their neighbors or people living in the same building. So that’s a very I mean, it’s a question that it’s got to be asked about. Maybe you can you can you can give it a try. So I wondering to which extent the lockdown measures, so basically confining many people when we don’t know who was infected and who was not, contributed to reduce the contagion or somehow boost it?

YONATAN GRAD: I’m sorry, I didn’t quite catch the question. Could you say it again?

Q: Yeah, I was wondering, considering the findings of what you said before, I was wondering to which extent confining so many people in the same place, in the same space, the same building without knowing which which were contaminated and which were not contributed actually reduce contagion or some bias to increase it?

YONATAN GRAD: Yeah. I think it’s you know, that may have been – so when people only interact at home, right, the only way for there to be spread is at home. But if people are also still interacting in the community and then also going home, then there’s still opportunities for transmission in the community and then brought into the home.

So, you know, a lockdown early on where people really adhere to the lockdown may prevent the spread in the community. And it would just be very limited spread of people who are infected and where you could see it in households. And I think in Wuhan we saw this as well, that after the lockdown, primarily that whatever spread there was was household spread for exactly that reason. The only people who are infected, they’re not going out. The only people they could infect are people who are in their immediate vicinity. Right. The people with whom they live. So I think that the idea is that the overall extent of spread would be dramatically reduced because you’re limiting community exposures and really just restricting the exposures as much as possible.

So, you know, ideally what you could do is be able to identify the people who are either infected or at risk for being infected and have ways to isolate them at home or have community resources and be able to help remove them from the home, of course, voluntarily and with full support, both, you know, keep them in some place in a hotel room or someplace where they would or some treatment facility if needed, where they would have support and also support the people and their families. If, for example, it’s a single parent family and the parent is or it’s a two parent family and the person infected is one of the person who is working. Right. I mean, you have to figure out how to support a community through something like that. But yes, I think that the notion is that you would have to identify the cases and then isolate them and quarantine the people at risk, even within within households to try to really diminish spread.

Q: So what think is it that you talked about in that ideal early lockdown situation in the Western countries, so we experienced a late lockdown situation. So that’s why I was wondering whether having the lockdown late could have actually contributed to having a higher rate of infection? And now we have many active people that got infected just as we were getting locked down which are now going back to the street to work and it might infect other people.

YONATAN GRAD: Right. I think that there there will be some degree of that. I don’t know the the numbers, but yes, there you know, whenever you have an infectious person and you start lifting that mitigation efforts and they start interacting more broadly with the community, this is why we’ll start to see cases go up. This is by we should expect to see a resurgence of virus, at least when there are enough susceptible people in a population.

Q: OK. Thank you.

MODERATOR: Next question.

Q: My questions are a little less technical, I think. Given all the symptoms and the possible symptoms and the emerging presumed maybe symptoms and so forth. Curious from your perspective, how unusual is this COVID-19 compared to other viral diseases?

YONATAN GRAD: For many infections, we see a range of presentations from asymptomatic infection to severe disease. We see it even with influenza. There are people with flu who have very mild symptoms. And then we see people who die from influenza. So it is not uncommon at all to see a wide range of presentations, meaning from asymptomatic to severely symptomatic.

Q: OK. That I understand the range. What about the variety from respiratory to digestive to potentially sounds like neurological and even skin conditions?

YONATAN GRAD: Oh, yeah. Viruses are so weird and wonderful that as someone who studies them and admires their versatility, let me give you an example.

Polio is in and is an enterovirus. It actually know it’s something you may have been vaccinated with that, you know, an oral polio vaccine. That means you take it by mouth. But this – and it comes from a virus family that infects the gut, right, the entero for our gut. But what do we worry about with polio? Polio can cause paralysis. Right. So somehow this gut virus gets in and can actually infect neurons and it can then cause paralysis. And, you know, it’s just a crazy range for it for that virus.

Another virus, herpes simplex virus. So herpes can cause cold sores, but it can also cause encephalitis, right. It can cause infection of the brain. Right. So, I mean, add chickenpox when you get chickenpox first as a kid or at least used to before the introduction of a vaccine, you get this itchy rash and all these bumps. And then later on in life, you can end up with shingles, which is just, you know, because that virus has been hanging out in your neurons, just waiting for a time when your immune system or your immune response to the virus is limited enough that it can reactivate. And then you end up with a stretch of the virus across just a patch in a neuron where you had reactivated. So there are all sorts of unusual manifestations for viruses out there. So this is this is one of, one in class of all sorts of unusual – viruses as a group can do all sorts of unusual things. And this is just kind of within that group of unusual.

Q: Fascinating. A quick last one. Do you think we’re anywhere near done seeing the symptoms that this virus is likely to cause?

YONATAN GRAD: I think that the clinical understanding or the understanding of the clinical manifestations continues to grow. But I think we have a pretty good grasp now of the variation in what’s common and then perhaps where we’re now really filling out what’s uncommon in in potential presentations. But I think we have now pretty good grasp of the manifestations of viral infection.

Q: That’s good to hear. Thank you.

This concludes the May 7 press conference.

Roger Shapiro, associate professor of immunology and infectious diseases (May 6, 2020)

Phyllis Kanki, the Mary Woodward Lasker Professor of Health Sciences in the Department of Immunology and Infectious Diseases (May 5, 2020)

Michael Mina, assistant professor of epidemiology (May 4, 2020)