COVID-19: Infection, spread, and testing

[0:00:00]

RICK WEISS: And welcome, everyone, to today’s media briefing. Before we get started, I want to do a quick one-minute introduction about SciLine for those of you who may not be familiar with us. We are a fully philanthropically funded free service for reporters, based at the American Association for the Advancement of Science in Washington, D.C. We are funded to do basically just one thing, which is to help you get more research-backed evidence into your news stories. And we do that through a variety of services that you can check out on our website, sciline.org, including a very – a valet service – matching service, through which you can get in touch with us and let us know what kind of an expert you need for help with your story, and we have a large database of articulate and deeply knowledgeable experts with whom we can connect you on deadline or as needed. We also, of course, sponsor media briefings like this one today.

In today’s, we have three experts who can talk to you briefly on the topic of COVID-19, aspects of COVID-19, to be followed by your Q&A. Their full bios are on the website, so I’m not going to take the time to fully introduce them now. I will just say that we will hear first from Dr. Vineet Menachery, assistant professor in the department of microbiology and immunology at the University of Texas Medical Branch at Galveston, and he will provide something of a profile of the virus so we understand what we’re talking about here – its behavior inside and outside the body, how it compares to some of its troublemaking viral cousins and what we might expect next from this family of pathogens. We’ll hear second from Dr. Jennifer Nuzzo, a senior scholar at the Johns Hopkins Center for Health Security and an associate professor in the department of environmental health and engineering and the department of epidemiology at Johns Hopkins Bloomberg School of Public Health. She will talk about the epidemiology of COVID-19, some of what we’re learning from the surveillance and models that scientists are using and some insights into what to expect next.

And last, we’ll hear from Dr. Jana Broadhurst, director of Nebraska Biocontainment Unit clinical laboratory and assistant professor in the department of pathology and microbiology at the University of Nebraska Medical Center. And she’s going to focus on what’s happening with testing – the kinds of tests available now, what – and how that may change in the future and some of the subtleties that reporters should pay attention to as you cover this increasingly important part of the COVID-19 story. So with that, let’s start with Dr. Vineet Menachery. Go ahead.

[0:02:43]

VINEET MENACHERY: So I will just give a quick background on coronaviruses in general. I’ve been working on coronavirus about 10 years, and there are three epidemic strains, the SARS coronavirus in 2002, the MERS coronavirus that emerged in 2012 and now the most recent and probably, arguably, the most successful, COVID-19, SARS-CoV-2. Coronaviruses are a family of coronaviruses that are – they’re a family of viruses that are high – are large RNA viruses. They contain a lot of tools to modulate the immune system. And they’re unique in that they are in animal reservoirs around the world. There are thousands and thousands of coronaviruses that exist in the world. Many of them are harbored in bats, specifically in their digestive system. And coronavirus emergence is on par with emergence of other viruses, and it’s something that we need to really consider moving forward – that this is not the last time a virus will emerge from a likely animal source and infect humans. These reservoirs have thousands of viruses and, for the most part, they haven’t been an issue prior to this because we haven’t really interacted with these species.

As the world continues to develop, we’re going to have these interactions more and more, and it’s a question of opportunity. Not every coronavirus that exists in a bat or an animal can infect a human and cause this kind of disease, but every once in a while, we’ll have that, and this is what we’re seeing with COVID-19. We saw it with MERS, and we saw it with SARS coronavirus. Now, in this transition from the bat into a human, there’s sometimes a reservoir species. For SARS 1, the original SARS coronavirus, it was civet cats that were able to amplify the virus in their respiratory tract and then infect people in these live markets in Southeast Asia. For MERS, the reservoir seems to be camel populations in the Middle East that have a very mild respiratory disease that can be passed – transferred to humans and cause significant disease in human patients. And for SARS-CoV-2, for COVID-19, we still don’t know the source, although there are highly related viruses in bats and there’s evidence that there are similar viruses – related viruses in pangolins in Southeast Asia.

And so in this transition, an important aspect of where – of coronavirus infection is where the virus is replicating. In many of the bat populations, the virus is replicated in the digestive tract and doesn’t seem to cause any significant disease in those animals. When it gets into a human, we’re finding that the virus is replicating in the respiratory tract. And where in the respiratory tract is really important. For viruses like SARS, MERS and SARS 2, all of these viruses can replicate deep in the lung, and this is where the vast majority of lethal disease is associated. When you think about the lung, the lung has a critical function in oxygenating your blood. And when you have infection deep in the lung, this can lead to major problems all throughout the body as oxygen is – deprivation to other organs is really problematic and puts a lot of stress on the body. The distinction between these viruses – these epidemic strains and the common cold coronavirus – is that ability to replicate deep in the lung. Most common cold coronaviruses replicate in the upper airway in your nose and throat but don’t really ever transition deep in the lung, causing significant disease. It’s relatively rare that that happens. These epidemic strains have caused the inverse of that, where there’s a great deal of infection deep in the lung.

A distinction between this new COVID-19 and the two epidemic strains is that, very clearly, SARS-CoV-2 infects the upper airway and that nasal and oral swabs are consistently being found to be positive in the upper airway, and this was rarely seen during the SARS and MERS outbreaks that patients had it in the upper airway. Another clear distinction is that there seems to be a number of patients that are transmitting the virus before they have strong symptoms. For both SARS and MERS, infection has been associated with severe disease, often a fever, and then the vast majority of SARS infections were associated with hospitalized infections. We may have missed the people that had milder infection, but it was very clear that when these people had a fever, they were transmissible. For SARS-CoV-2, it’s not as clear when they become transmissible. It may be before they have strong symptomology, which would explain why the virus is spreading like it is – that you may be infectious before you know you’re even sick. The other thing is where these viruses replicate really dictates how they spread and how severe the disease is. Infection in the upper airway leads to viruses that are highly transmissible, so the common cold is the common cold because it transmits relatively easy and it replicates in the upper airway. For these epidemic strains, for example MERS, they don’t replicate as well in the upper airway, and so they are not as easily transmitted, but they are far more lethal.

So the MERS lethality rate is much higher than it is for SARS – around 1 in 3. For SARS – the original SARS, it was about 1 in 10. And the case-fatality rate is lower in COVID-19 overall, although it’s infecting a lot more people. As the virus infects, there are a number of things that can happen. So in the upper airway, your immune response is robust, can eventually deal with the virus. As you go deeper into the lung, you don’t really want the same type of immune response, and the damage that’s associated with the infection can be driven by three things. One, the virus itself can drive damage in the lungs, the alveoli, and that disrupts air exchange and oxygenation. The second thing that can happen is that your immune response can actually overrespond and cause a significant amount of damage, and this damage can be just as bad or even worse than the virus itself. And this is actually exacerbated in people with health disparities, as well as in older people. The third thing is that you can have opportunistic infection, so bacteria that normally don’t cause problems can then traffic deep in the lung, and this can cause a pneumonia that’s often seen with other respiratory viruses like flu.

But each of these three things can cause this damage and infection to make serious disease happen and really disrupt oxygenation and your lung function. The big thing about transmissibility that we don’t know yet is how infectious the people are. There seems to be various levels of infectivity and when you become infectious. Classically, coughing and sneezing has been the main driver of spread. It is less clear that that’s what’s driving spread here, and it may just be regular breath, and that’s why in contact – limiting your contact with other people, that 6 feet distance really protects you from the spread of this virus. There is some thought that it can live on fomites, so this is pieces of equipment or things that you come in contact with. And then if you come in contact with something like a doorknob that’s contaminated and then you touch a mucosal surface – your eyes, your nose, your throat – you can infect yourself that way through your hands, not necessarily through breathing. The expectation in my mind is that the vast majority of the spread is happening person-to-person, and it may be through a person that’s not necessarily sick yet.

Overall, the big thing to consider is that this is probably not the last emergent virus. Over the last decade, we’ve had the emergence of new COVID-19, before that, Ebola, H1N1 in 2009, Zika virus. As the world continues to interact with these new animal populations, preparation is going to be key, and not only just in terms of vaccines, but therapeutics and being ready to respond. I think COVID-19 has really been a wake-up call in terms of how we need to be ready to respond. And I think, hopefully, this will lead to better preparation as we go forward.

RICK WEISS: Thank you so much. Dr. Nuzzo, over to you.

[0:10:44]

JENNIFER NUZZO: Great. I just have a few graphics to share with my remarks and just trying to look at where we are now and anticipate a little bit what we may be in store – what may be in store in the coming months. So where we are is a fairly grim picture. Globally, there’s about 2 – close to 2 million cases have occurred – 1.6 as of today. The vast majority of the cases to date have been reported by the United States. We, several weeks ago, outpaced China in terms of the total number of cases, and many other countries have similarly done the same. But in total, about 185 countries are now reporting cases, which means this is truly a global scenario. And those countries that have not yet reported cases – I don’t think we should take that as a sign that there are not cases there. There are clearly surveillance challenges. It’s just important to recognize that what we’re talking about here is a new virus. And the fact that we’ve been able to stand up surveillance in 185 countries to look for a novel virus I think is an important feat that should – maybe one point of light in what is otherwise a very challenging situation.

Looking at the situation here in the United States, you know, as of a few weeks ago, I think life here changed quite a bit, in part, I think, because we were looking at what was happening in Italy and seeing a very strong, sharp rise in cases coupled with reports of health systems, you know, at the brink, really – health systems seeing a large influx of critically ill patients and projected out that they would exceed demand if drastic actions weren’t taken. At that point, many started to worry if we in the United States were, you know, just weeks away from that reality. And so you saw across the United States many cities and state governments decided to put in very – you know, a range but increasing amount of restriction in terms of movement and restriction on mass gatherings – you know, these things that are all sort of lumped in under the phrase of social distancing – in part because we were worried about that sharp rise that you see here.

This graph is actually a graph from The New York Times, which actually pulls data from the Johns Hopkins COVID tracking website, but I like the way that it’s depicted, so I’m showing it here. And what you see is, you know, there’s that increase – that rapid increase in cases, but hopefully – and it’s really hard to interpret recent data, so just as a plea to anyone looking at recent numbers, always view those with a lens of suspicion because there’s often reporting delays in those recent numbers. But in many parts of the U.S., we are hearing stories about the rate of growth slowing, which is good news for sure. But we can’t, you know, sit back and get too complacent hearing that because where you are seeing rates of growth slowing is often in places that are possibly further along in their epidemics, places that have already experienced that rapid growth and have also put into place very serious social distancing measures in an attempt to stem that growth. There are other parts of the country, as shown here, that are experiencing the much more fast increase in cases. And just, I wanted to show you this map because it shows that this is truly a national problem.

Sometimes in – if you sort of rely on the news reports, it’s overpopulated by things that are called hot spots, and those tend to be stories from cities – cities, obviously, because there’s a lot of people there and, possibly because of the density – it’s not quite clear – just the case numbers are large. But case numbers alone I think give us an imperfect picture of what’s going on because the reason why we are implementing these social distancing measures is to slow the rate of growth of cases, but in particular so that we can keep the rate of growth and the total number of cases occurring in any one period – time period – we want to keep that number below the kind of upper limits of our health care capacity. And so when we think about where we may need to take measures and where there are possible risk areas, we also have to think about what the health care capacity is in those areas. And so there may be some parts of the country – possibly some of the areas shown here – where the case numbers themselves aren’t high compared to, say, New York City, which is one of the more populous areas of the country, but may have fewer health care resources to deal with the cases that they have.

And so it is possible that less-urban areas are particularly at risk given the lack of health care resources or limited health care resources. Often in cities, there’s multiple hospitals, and they can share resources. They can reallocate patients. But in other parts of the country, that may not be a possibility. And so I think sometimes, some of the story about hot spots in the United States neglect the fact that there are probably other parts of the country that have fewer cases but that are particularly at risk given the availability of health care resources. So when we look and we hear encouraging stories about rate of growth slowing, we also need to consider that not every place may be experiencing that – some places may be on the upswing – and that we also have to compare these numbers to what the availability of health care resources is. There’s been a lot of news about trying to understand what the ultimate impact of the pandemic is going to be, lots of discussion of models and projections about deaths and hospitalizations. This model from IHME is one that’s often in the news because it’s the one that the task force is pointing to – the president’s Coronavirus Task Force has been pointing to. And they have made projections about total numbers of deaths and anticipated hospitalizations.

What I want to say about this and other models, first of all, I think is a overarching caution in that they are only as good as the underlying data. And the underlying data is flawed in a number of ways. That said, we can’t throw out the models. We still need to make decisions, and having tools to guide the decision-making is helpful. IHME is a very well-respected group. They are taking a different modeling approach than some of the other groups that do modeling on this. Namely, they are trying to project based on accumulated cases and sort of fitting a curve to that. Other modeling teams take a different approach where they do more sort of probabilistic – they take more probabilistic approaches where they kind of assume that a certain fraction of the population will be infected, and they look at that over time and layer on top of that potential interventions. But what amounts – what I think is an important takeaway is that there is a considerable range in the projections from these models. And so I think IHME is on the lower end of the range in the sense that they have been – recently sort of ratcheted down their estimations of the numbers of deaths that could be seen in the United States to about 60,000, whereas another modeling group in the U.K. had anticipated, you know, over 2 million.

And different assumptions go into those models. One assumes that social distancing will be maintained for months. The other one – that estimation comes from assuming we do nothing, which is clearly not something that’s happening. And then they look at different scenarios. But the numbers themselves, I think, are probably not as useful as understanding how the numbers can change should we employ different measures. Another question that’s often asked – I feel like I get asked a lot – is, how long is this all going to last? Here’s the Imperial model that I was mentioning, which was one of the kind of upper-tier models in terms of potential deaths estimated. I’m showing the graphic from the Imperial College model because I think what it illustrates is this timing issue. And the takeaway here is that social distancing measures, all of the various measures listed here are not a cure. They’re a pause button. And they are employed in an attempt to slow the growth in cases with the goal of keeping the growth below the upper limits of health care capacity. But they do not stop the transmission, and they are likely only going to – the effect is likely going to last as long as the measures themselves last.

So what the group looked at was essentially a possibility of having to maintain measures for 18 months. And they chose 18 months likely because that’s one of the estimates of when a vaccine could become available. But the overall takeaway here is that were we to dial back on social distancing, we should very much anticipate that there will be a rise in cases. And therefore, if we want to change what we’re doing now, we have to have a plan to deal with those rise in cases. Otherwise, we’ll potentially find ourselves back to where we started, which would very much not be ideal. But we need a new path because these measures that we’re doing right now, while having an impact, clearly, I think, from looking at the data that I showed earlier, they’re disruptive. I mean, they’re – I don’t think it’s something that society potentially can maintain for 18 months. I think that would be very, very difficult to do. So we need to have a new path. And so there’s been a number of groups thinking about what that next path is. I have looked at this. One of the concerns that I have had is that in order to make decisions about when and how to lift social distancing restrictions, we need to have better surveillance both for the disease and also for the impact on the health system.

And right now, we have very little surveillance on the health system and quite imperfect surveillance for the disease. There are some insights that I think we can be – that can be gained from looking at other countries’ experience, and one country that gets a fair amount of credit is Singapore. Singapore did not implement most of the social distancing measures that the United States has implemented. Nonetheless, they were still able to bring down their case incidents. So they initially had a rise in cases, and then they responded in such a way that they were able to reduce the number of cases that were occurring. And they did this through very aggressive case-based interventions. So if social distancing is applied to the whole population, it’s considered a population-based measure. What Singapore did was they went after the cases themselves. They very aggressively tried to identify cases, isolate those cases so they couldn’t transmit their disease to others, identify contacts of the cases and monitor those contacts so that they would not be able to transmit to others if they became a case themselves.

And through that – those case-based measures – they had great success. Similar approach in Singapore – South Korea, rather. South Korea is also in the news a lot for not only its case-based measures. Like Singapore, they did fairly aggressive contact tracing, so identifying the people that cases had been in contact with 14 days prior to becoming ill. They used all sorts of technological advances – like cellphone location apps, closed-circuit TV, credit card bills – a variety of different methods to identify who cases may have come in contact with while they were potentially able to transmit their infection. But they also had a very aggressive approach in testing in the sense that they rapidly scaled up testing such that they could more completely identify cases within affected areas so that they can more early identify those cases and isolate them before they have the potential to transmit to others. And what they have been able to do is to bring down their case numbers. Now, both South Korea and Singapore and other countries who have taken these case-based measures are not insulated from the world.

And so as the rest of the world experiences a rise in cases, these areas are subject to increases in cases just because they’re coming into the country in a variety of ways. Even if they’ve implemented travel measures, they still have citizens returning from overseas. But they have an infrastructure in place to handle these cases so that they don’t allow the case numbers to explode to the fact – to the place where they have to implement wide-scale lockdowns such as we’re seeing in a number of other countries. So just before I move on, these case-based measures are going to be essential for the United States in its next phase because if we release our social distancing measures, we will have to have a plan to deal with the rise in cases. Otherwise, it will just escalate again, potentially to the point where we were before we implemented these population-based measures. And so one of the hindrances for the United States in being able to enter this next phase, in addition to – I already mentioned that we need to improve our surveillance. Part of improving our surveillance is expanding our testing capacity. This is also a controversial topic. It’s in the news. But if you talk to almost any public health practitioner in any state, testing is severely limited. And test – and having limits to testing really limits planning and response for dealing with the next phase in the COVID-19 response.

One of the – I – there’s been increasing call for developing a national testing strategy, one that looks at how should we be using testing capabilities, how can we expand testing capabilities. Different types of testing approaches are coming online, so not just looking for the virus but also looking at antibodies to see who may have been infected in the past. We need the strategies to optimize these approaches so that we can get the most information out of it to best inform our response. We also need to deal with the fact that different states are taking different approaches now. And while that may be necessary for operational flexibility, it also limits our understanding of what’s happening across the nation. So at the very least, we need to understand the state-by-state approaches. We need a much more coordinated approach to thinking about how to expand testing and, in particular, eliminate some of the bottlenecks, which are largely a national issue, like shortages of the supply chain, not being able to access reagents – the chemicals that are used for testing – or the swabs that are used to take specimens from infected people or the personal protective equipment that is needed to protect the people who are taking the specimens.

So all these are things that need to be addressed. And ideally, that happens at the national level so you don’t have 50 states competing against each other. But that will be an important requirement for going forward and hopefully being able to get to the point where we can release some of the restrictions that have been put into place. I’ll end there.

RICK WEISS: Thank you, Dr. Nuzzo, a fantastic arc of story to help us understand what’s going on there and a wonderful segue to Dr. Broadhurst to talk about testing.

[0:27:31]

JANA BROADHURST: Yeah, thank you. That was quite a nice lead-in to this topic. So I’m coming from the laboratory perspective for diagnostic testing, and I’m just going to touch on three themes in the topic of diagnostic testing for COVID-19. So I’ll touch on the types of testing, and those were introduced. I’ll go into a bit more detail on the test methodologies that are currently available. I’ll also talk about in a bit more detail the challenges in implementing diagnostic testing and why it is that we have the limitations that we do and what we can project going forward. And finally, I’ll say a few notes about the role of testing currently and moving forward in our response to the COVID-19 crisis. So beginning with the types of testing methods that are available.

So as noted, there are two basic methods that are employed. One is to detect the virus directly. And the way that we do that currently is to detect the genomic material of the virus, the RNA. And in order to do that, we need to sample compartments where the virus is actively replicating, so we tend to sample the respiratory tract – into the upper or lower respiratory tract. And our first speaker also introduced nicely how that correlates with the pathophysiology of the infection. And something I want to say about, really, the advantages and disadvantages of that testing approach – an advantage is that that testing approach is very specific. So if you get a positive test result, looking for the RNA of the virus with the current methods that we have, it’s very likely to be a true positive test. Now, if you get a negative test result, however, the possibility for a false negative test is there. And that’s something that maybe we can touch on a bit more if there’s questions about it.

I know that I frequently get questions around what our sources and how to understand the potential for false negative tests. And in order to understand that, you really need to look at the entire testing process and who is tested, how they’re tested and what tests are being used in the laboratory. Now, we currently have more than 30 tests that are – have an emergency use authorization by the FDA to detect virus RNA. So we have, really, an incredible explosion of tools available to perform these tests. But again, despite this, you know, daily increase in the availability of diagnostic tools, that’s not translating into a similar rate of increase in our ability to actually deploy that testing and to get test results to physicians and public health authorities as well as we need to do. So the second basic methodology is to detect antibodies that form against the virus when an individual is exposed. And this is what we call serology testing. So in order to detect antibodies, it’s a very different approach.

You would sample the blood, as opposed to the respiratory tract where the virus is replicating. So it’s a very different strategy for how you would go about collecting these specimens from large groups of people. Now, I mentioned that for detecting the RNA of the virus, the specificity of the test is very good, but the sensitivity can vary. Now, with antibody detection methods, we worry, actually, a bit more about the specificity of the method. So in this case, particularly with the variability and the inequality of serology tests that are out there, a positive test may not be so reliable to indicate that there’s been an exposure specifically to SARS-CoV-2.

Now, the sensitivity of the test is something that we have to give thought to, as well. So antibodies form once you are exposed to the virus. And there is a window period after you’ve been exposed and developed symptoms until those antibodies develop. And that can be variable from person to person and depending on the severity of their exposure. So the sensitivity of the test is going to depend on who you’re testing and when you’re testing them. But there is a durability to that antibody response. So it does allow you to determine that an individual has been exposed to the virus for much longer than you would be able to detect the virus itself. Now, our capability to offer serology testing or detection of antibodies is much more limited right now than our virus RNA detection methods. So there is currently only a single test that’s approved by the FDA to be distributed for serology testing.

That’s going to increase very rapidly over the coming weeks. There are many, many of these tests that are in development. And again, they vary widely both in their format – whether or not they’re a small point-of-care detection device that may be able to use just a finger prick of blood to very large, high-throughput platforms that would be placed in central laboratories for referral testing. And so all of these tools are in development but not currently widely available or implemented in clinical laboratories. So next, I want to move on to some of the ongoing challenges in diagnostic testing. And this – you know, as I said, we’re not actually limited in the tools that we have to perform this testing, particularly for virus detection. So what are the challenges, and why do we continue to have these delays and these limitations in how widely we can offer testing? Well, one of the challenges begins with just who to test. This is a very active conversation, and it has to be balanced with the resources that are available and the allocation of those resources.

And again, these strategies have varied state by state and even community by community as these priorities are balanced, and resource allocation is quite variable in different parts of the country and within communities within each state. So, you know, it’s a very major strategic decision to how to prioritize maybe the limited test kits that are available and employing those, you know, either for those who are sick and presenting to the hospital versus reaching out into the community to get a better idea of the extent of spread within the community. So another major challenge, again – who to test, how to allocate your resources. And another is how to ensure that your testing is accurate and reliable and interpret those test results appropriately. This comes back to the question around a concern that is present nationally for the occurrence of false negative tests.

So I want to just say a couple of words about that, what it means and what may be the source of this concern and potential reality. So what we mean when we say a false negative test is that the test used to determine that an individual has COVID-19 infection does not detect the presence of that infection. And there’s a lot of sources of – that can feed into that type of discordant result. And one place that starts is with the biology of the virus. So individuals may have variable levels of the virus and shedding of the virus in the anatomic sites that are sampled to perform the test. So that’s quite important to remember. Some individuals may be negative on an initial test and positive on a subsequent test, particularly later in their disease, of course, when the levels of those virus may be lower and more fluctuating. So in some cases, it’s really just the underlying biology of the infection that’s leading to variability in that test result.

Another big source of potential variability is the specimen collected and the quality of that specimen. So we have a mantra in the lab that’s garbage in, garbage out. So our tests are optimized for a certain type of specimen and quality of that specimen. If that’s not what’s received in the laboratory, then that test may not perform as well. I think this is a really important point to consider as we see really, you know, I think, creative and resourceful strategies to increase the breadth of our testing, including strategies like home-based testing kits and drive-through testing sites. But it’s just absolutely critical that there’s a way to ensure the quality of the specimen that’s collected and understand that that may not be comparable to a specimen that’s collected by a health care worker in a clinical setting.

And so the same tests may be performed. But if it’s on a very different quality of specimen collected, then the results may be more variable. So that’s really an important consideration in policy decisions around strategies to increase the breadth of our testing. I’ll move to just a – two more points around the challenges in test performance and interpretation. So another is the inherent characteristics of the test performed. So I mentioned there’s now more than 30 tests that are available for detecting the RNA of the virus in a clinical specimen. And those tests, while they operate with a similar underlying technology for RNA detection, namely nucleic acid amplification, the platforms do vary considerably. And some of those are designed to be employed at or near the patient. So we call those, you know, rapid point-of-care testing. It’s very attractive to have that, you know, quick turnaround time and results sort of in the same sitting as the specimen being collected.

But we know that there’s really quite a lot of variability in the performance of those tests, the sensitivity of those tests. So that’s, you know, certainly one source of variability in test results – are the platform and specific characteristics of the test. And those vary very widely, as was noted by a prior speaker. There is no cohesive strategy in how these different testing platforms are being deployed across states and within communities. In some ways, that’s been a product of what platforms are already in place and where kits have been developed that can be used on existing platforms that are present in clinical laboratories in different regions. And in some ways, it’s really been, you know, just a game of who gets on the priority list for the very limited number of test kits or other testing components required to run a test. So it’s been very difficult. I can, you know, speak for my state that it’s been quite difficult to project a strategy more than a few weeks out with the uncertainties around the availability of test kits and components.

And there’s really been a need to stay very, you know, flexible and adaptable as new testing – as new tests themselves become available but also the resources needed to run those tests at the volumes required to have – to sustain our health care practices and public health priorities. And finally, the last thing I’ll say about, you know, some of the concerns and potential sources of variability in test results is really the quality of those materials that are being manufactured and sent out to labs. So this is, you know, a crisis that has not just strained our health care system, but it has strained the manufacturing and supply chain that services the health care system. And it’s been, I think, a reality we’ve had to face that, when those manufacturing supply chain sources are strained, that can lead to variability in the components that are provided.

And that has really highlighted what we know in the laboratory as our everyday job, which is to ensure the quality of the testing we’re doing and the quality of the components used to perform those tests. And it’s never been more important than it is today that those quality assurances be very stringent. So we’ll move on now to the role of testing in the response today and moving forward, and I’ll break that down I think just by, again, focusing on the different types of tests that are available.

RICK WEISS: And Dr. Broadhurst, I’m going to ask you to wrap up if you can in just two to three minutes on that because I do want to get to Q&A.

[0:41:57]

JANA BROADHURST: Absolutely. So just to note that, you know, direct detection in virus remains very important both for the management of individual patients, as well as for understanding prevalence in communities and monitoring our health care workers. But there’s been a lot more discussion recently and a desire to understand the role of serology testing, and that’s going to certainly play a role – and maybe we’ll talk about this more in the Q&A session – around understanding whether or not an individual is – does have protective immunity. That’s not something that can be assumed from a positive serology test, but it is also going to help us, you know, understand the overall prevalence of the infection in our communities and ultimately, hopefully, help us understand the efficacy of vaccines that are – that come to bear. I’ll stop there. Thank you.

[0:42:53]

RICK WEISS: Fantastic. Very rich presentation. Thank you so much to get us – for getting us started there. I’m going to read the first question here. It’s from Andrew Joseph at STAT News. And, actually, it touches on something you just brought up, Jana. But Dr. – it’s directed to Dr. Menachery. I’m curious – says Andrew Joseph at Stat – I’m curious about where things stand with what’s assumed and what’s known about immunity after an initial coronavirus infection. Specifically, could a lower viral load, in addition to potentially leading to milder infection, also produce weaker and less sustained protection? What do we know about that?

[0:43:33]

VINEET MENACHERY: So that’s still not abundantly clear if severity of disease is driving lower immunity long term. Most of the work that’s been done in this area has been focused on serum neutralization – so antibody in your blood. And we know with coronaviruses that your antibody in your blood doesn’t last a lifetime. Often, for the common cold coronaviruses, which are often less severe, it can last two to three years. Similarly, we’ve seen the same things with MERS coronavirus – that antibodies in your blood are only lasting a couple of years. But there’s reports that, for SARS – the original SARS – most of those patients were pretty severe that got diagnosed, and they had greater than 10 years of serum neutralizing antibody.

The key here is that antibody in your blood can protect you from immediate infection, but these people that lose that serum neutralization – it doesn’t mean necessarily that they’re not going to have some level of immunity. Your immune system hasn’t forgotten. It just may take them a couple of days to generate that immune response and be able to clear a virus quickly. So my expectation is that the people that get infected and clear the virus will have some level of immunity. Whether they cannot be infected ever again is not as clear, but the expectation would be that the virus, on a second round of infection, would cause a much more mild disease. And those people may not even know they’re infected.

[0:44:57]

RICK WEISS: Great. A question here that might be best for Dr. Nuzzo is from Sabrina Wilson, WVUE-TV in New Orleans. In Louisiana, the number of positive cases is still alarming, even as there appears to be some flattening of the curve. What will most affect the growth in cases in your opinion? And then, of course, the question everyone wants to know – realistically, when should we expect the virus to disappear?

JENNIFER NUZZO: So the question – I’m sorry, so the question about the social…

RICK WEISS: The question is really what will most affect the growth in cases in New Orleans – or really appropriate for any city, I guess.

[0:45:36]

JENNIFER NUZZO: So it’s really hard to disaggregate various social distancing measures, in part because we’re still learning about the transmission of the virus. But also I think it – a lot about the population structure plays into that – you know, whether they’re multigenerational households or what kind of exposures people have. So let’s just assume that social distancing is a very blunt tool, and it’s very hard to kind of disentangle all of the measures that are being implemented all at once. That said, fundamentally important – no replacement for this – is case isolation. So being able to find cases and rapidly isolate them so that they don’t transmit their infection to others – this is important above all. And this is why, you know, if you look to a place like Singapore that didn’t implement social distancing measures largely, they were still able to see a decline. So I think we can safely say that that should be our top priority.

Secondly, we have to protect those who are most vulnerable from developing severe illness and death. This is less about overall case numbers but about trying to offload or reduce burden on the health system. I’m particularly worried about places like long-term care facilities, where you have large numbers of particularly vulnerable individuals who, if they become infected, first of all, it can spread very quickly in these large congregate settings in these facilities. But secondly, when these patients become ill, they are most likely to be the ones who will require intensive care and potentially could die from their infections. So going forward, there may be some level of social distancing that remains. And, perhaps, cocooning, protecting particularly vulnerable individuals, that that will have to continue to occur until we have some other tool like a vaccine or possibly, you know, if there’s a therapeutic that can be used to prevent people from becoming severely ill if they are to be infected. Otherwise, we have to continue to protect the most vulnerable.

[0:47:42]

RICK WEISS: And just – I’ll stick with you for one more moment to follow one question from Zack Sampson at Tampa Bay Times. Could we in the United States have avoided social distancing altogether with different preparation? And how much of the move towards reopening the country is in our control versus to the virus at this point?

JENNIFER NUZZO: We have avoided in the U.S.?

RICK WEISS: Could we have avoided the social distancing strategy with a…

JENNIFER NUZZO: Yeah.

RICK WEISS: …With different preparation, had we been better prepared?

JENNIFER NUZZO: I can’t say for sure, but I have a feeling that it could’ve been possible. And again, I look at Singapore. Now, could we have ever had the kind of resources that they have? It’s a smaller – you know, it’s a small island, and they are remarkably well-resourced. And I think we have more challenges here. I think we also have some limits on the approaches. Some approaches that they’re taking I think may not be translatable here. But I do often think about the lost time and whether we could have avoided at least the level and extent of the social distancing that we are now having to do to play kind of catch-up. We’re suddenly, you know, seeing this rapid expansion in cases, in part because we weren’t testing, in part because we weren’t doing these case-based interventions from the beginning.

[0:48:58]

RICK WEISS: Dr. Menachery, a question from Erin Garcia de Jesus at Science News regarding CDC’s new guidance that essential workers can continue going to work as long as they don’t have symptoms and they take additional precautions, such as wearing a mask. I’m curious how this recommendation fits in with the current knowledge. Given what we’re learning about presymptomatic spread, is this a good idea?

VINEET MENACHERY: So, I mean, the key here is that the essential workers that – the mask doesn’t protect you from necessarily getting infected. If you’re wearing a hospital mask, that is not necessarily going to stop a coronavirus particle from getting to you. What it does is it limits your ability to spread it to other people. So if you are presymptomatic but you are infected, wearing a mask will diminish the spread of the virus from you, and that’s the key factor here. This is a protection not necessarily for you, but for you to infect other people. With the limited numbers of masks, specifically N95 masks, we really need to prioritize those for the essential workers that are in the line of fire, specifically health care workers who are getting the most exposure, and that is the biggest threat. From there down, obviously, people who are interacting with lots of people are another source. But until masks are readily available, you know, the cloth masks are limited in their protection for getting the virus, but there’s good evidence that it can limit spread.

[0:50:29]

RICK WEISS: Question here for Dr. Broadhurst. I think that’s from Jenifer Joy Madden, a freelance reporter. Is there a clearinghouse for information on tests being developed, what stage they are in and who is developing them?

JANA BROADHURST: Well, certainly, the best source of information for the tests that are currently available is on the FDA website. So there is information there both on – really clear information on the tests that have been authorized for use and the data that is available on the performance of those tests. But there’s also links from that source to, for example, manufacturers who have indicated their intent to pursue authorization. So if that’s – that’s really, I think, the most comprehensive resource.

[0:51:24]

RICK WEISS: Great. Question here. Do we know whether this – sorry. This is from Christine Herman from Illinois Public Media. Do we know whether this coronavirus strain will evolve, similar to the influenza virus, so that people who are immune now may not be immune next year or later? That may be for you, Dr. Menachery.

VINEET MENACHERY: Yeah. So coronaviruses are large RNA viruses, but they’re much more stable than your average RNA virus. We’ve worked with common cold coronaviruses that are over 30 years old. And they’re not appreciably different than the strains that circulate today. So I would not expect that in a year’s time you would have so many changes that the immunity you generate now would not protect you a year from now. The question there will be how long your immunity lasts and how effective it is long-term.

[0:52:15]

RICK WEISS: Great. Follow-up here for Dr. Broadhurst from Julie Mack from MLive Media Group in Michigan. To get more specific on the issue of accuracy, what is the general accuracy of the COVID-19 tests? What’s the rough percentage of false negatives? And how does that compare with, say, flu tests?

JANA BROADHURST: Yeah, it’s a great question. And that requires a gold standard. And that’s – I think one reason I framed it around, you know, really – is – we’re talking about not so much that the piece of the virus that we’re looking for is there, but is that individual exposed and infected with the virus is really the standard that we’re trying to compare against. So, you know, there’s different numbers out there. The short answer, I think, in summary, is that PCR – detection of virus RNA by PCR, if you just summarize across the breadth of folks who may be tested – those who are acutely ill, those who are, you know, infected but asymptomatic, those who are late in their disease and have variable shedding – it’s probably something around 70%. But that’s a very difficult number. There’s variability in the tests themselves. But again, who you test and when they test and when you test them and the specimen that you use to collect all really impact the performance of the test. So, you know, the summary number, I think, loses some meaning.

RICK WEISS: Yeah. I just want to make sure I heard that number right because of the broadcast quality – imperfect here. Did you say 7% false negative or…

JANA BROADHURST: I said a 70% sensitivity.

RICK WEISS: Sensitivity, meaning 30% false negative. Is that saying the same thing?

JANA BROADHURST: That would be saying the same thing.

[0:54:11]

RICK WEISS: Yeah. OK. All right. I know we’ve got some hard stops here at 3. I’m so sorry. We’re not going to get to every question. I’m going to give one minute or a half-minute to a minute for each of you to maybe hammer home a take-home point. I’m going to start with you, Dr. Nuzzo, ’cause I know you got another call you need to get onto. Is there something you’d like to close with as we start to wrap?

JENNIFER NUZZO: Yeah. I guess one of the kind of pleas is that how we are able to move forward very much is dependent on having access to the right information. And clearly, testing is essential for all of this. I mean, all of the game plans for the next phase really come up against a hard limit, which is not being able right now to expand testing given limitations in supplies and other things like that. But we very much need to be able to do that in order to move forward. The other thing is that just beware of projections and case numbers. They don’t tell the full story. It requires a little bit more of a nuanced look. And, you know, these are very high-consequence decisions that have to be made. And so just using very gross estimates is a risky endeavor when you’re trying to make very high-consequence decisions and requires data from multiple sources and multiple looks at those data.

RICK WEISS: OK. Dr. Broadhurst?

[0:55:42]

JANA BROADHURST: Yeah. I think just a summary take-home point from my perspective – and this has been really well-reflected in the conversation overall. From a testing perspective, just that the, you know, the test results that we can provide have an important impact on public health policy both on a local and regional and national scale, but also that public health policy impacts the test results that we’re able to provide. And that has to do, again, with who we test, how we test and how we allocate resources. So I think there’s a very important and intricate interplay there going in both directions.

RICK WEISS: Right. And Dr. Menachery?

[0:56:26]

VINEET MENACHERY: Yeah. I think the take-home I want you to consider is that this is maybe not the last coronavirus we’ll deal with in our lifetime. This one seems to have a, really, truly awful blend of being highly transmissible and highly virulent, so we may not get unlucky again. But the preparation for this will have to be very different going forward as we are dealing with a now – forward approaches to develop therapeutics, vaccine platforms, be able to get capacity high enough to deal with these kind of infections because I don’t think they’re going away. And it may not be a coronavirus next time. It may be influenza. It may be an Ebola-like virus. But preparation in the public health and in the scientific endeavors is going to be really important going forward.

[0:57:17]

RICK WEISS: OK, a fine warning that maybe we can learn some better lessons from this time going forward. I want to thank our three guests very much for fantastic amount of information you’ve conveyed over this hour. I want to remind our reporter attendees that you can check out SciLine’s services at sciline.org. You can follow us on Twitter at @RealSciLine. Please, when you sign off, you will get a prompt to look at a survey. There are just three very quick questions there. Everyone hates surveys – it’s really helpful to us so that we can keep providing you briefings and other opportunities like this that you most want and that can help you the most. If you would take the literally one minute it will take to fill out this three-question survey, we’d be so grateful. And with that, we thank you all for attending. Thanks again to our guests. And we’ll see you at the next SciLine event.