Podcast: COVID-19: Non-binding Guidance: Regulation of Diagnostic Tests and “Return to Work”

July 21, 2020
23:49 minutes

The tenth installment of Ropes & Gray’s podcast series, Non-binding Guidance, focuses on FDA regulation of COVID-19 diagnostic tests, with a particular focus on employee testing. In this episode, chair of Ropes & Gray’s FDA regulatory practice, Greg Levine, interviews Lauren Silvis, a former FDA official and currently Senior Vice President of External Affairs for Tempus, a clinical molecular technology company, to hear her input on the role employee testing can play in maintaining a safe workplace. Topics covered include the role of antibody tests, testing of asymptomatic persons, pooled sample testing, and evolving technologies such as Next Generation Sequencing (NGS). Prior to joining Tempus, Lauren spent a number of years at the FDA, first as deputy center director for policy in the agency's Center for Devices and Radiological Health, and subsequently as the FDA chief of staff, where she was responsible for overseeing the daily management of the agency and leading the agency's activities on major initiatives. Additionally, Lauren co-authored several reports and articles on COVID-19 – most recently, “How Businesses Can Keep Employees Safe From Coronavirus.”


Greg LevineGreg Levine: Hi, I'm Greg Levine, chair of the life sciences regulatory & compliance practice group at Ropes & Gray, based in our Washington, D.C. office. Welcome to Non-binding Guidance, a podcast series from Ropes & Gray focused on current trends in FDA regulatory law, as well as other important develops affecting the life sciences industry. I'm here today with Lauren Silvis, Senior Vice President of External Affairs for Tempus, a clinical and molecular technology company. 

In her role at Tempus, Lauren oversees regulatory, public policy, and government affairs for the company. Prior to joining Tempus, Lauren spent a number of years at the FDA, first as Deputy Center Director for Policy in the agency's Center for Devices and Radiological Health, and subsequently as the FDA Chief of Staff, where she was responsible for overseeing the daily management of the agency and leading the agency's activities on major initiatives. 

Over the last few months, Lauren and former FDA commissioner Scott Gottlieb have co-authored several reports and articles on COVID-19. In March, they published a widely-read report through the American Enterprise Institute titled “National Coronavirus Response: A Road Map to Reopening,” which took an early look at the current COVID-19 pandemic in the United States and how to adapt our current public health approach to limit the epidemic spread of the virus. Later, in April, they released a report through the Duke Margolis Center for Health Policy that examined how our national surveillance system could be used to mitigate the pandemic wave and suppress future outbreaks. Most recently, they co-authored a Wall Street Journal article titled “How Businesses Can Keep Employees Safe from Coronavirus.” It was that article that prompted my interest in discussing today's topic with Lauren. 

Lauren, thank you for taking the time to join our podcast today. Before we get into today's specific topic on COVID-19 diagnostic testing by employers, what are your general impressions of how FDA has handled regulation of COVID diagnostic testing? There's been significant criticism: for example, that FDA initially went too far in loosening the regulatory reins, particularly with many of the antibody tests. We've now seen FDA have to withdraw the authorizations for some of those that they've found to be insufficiently accurate. I'm just curious about your impression of how the agency has been trying to achieve this balance between having sufficient access to testing and also ensuring the accuracy of the tests that are available. 

Lauren SilvisLauren Silvis: I think FDA is dealing with an unprecedented public health crisis where they've never had to authorize so many tests, and so many different types of tests under an EUA so quickly. I was at the agency when tests were coming in for Zika and Ebola, and it was a very different flow and those where different kinds of outbreaks. And now, they're having to deal with authorizing tests that are purely run by labs. Many different types of technologies are coming in. They’re also having to deal with sponsors who aren't as used to dealing with FDA and so they're trying to keep their approach flexible. On the antibody approach, I understand why initially they didn't review those because they weren't diagnostic and we were in a diagnostic crunch and those were the ones that needed the most oversight.  But I think once FDA realized that there were concerns about the tests' performance, they recalibrated and changed their approach. So they're being flexible and continuing to evolve, and we're seeing that even now by updating their templates for different types of testing and different types of technology because there's different things coming in the door for them all the time. 

Greg Levine: What about the fundamental question of whether EUAs are required for these tests if they're developed by laboratories? As you well know, there's been a whole debate about LDTs, laboratory developed tests, and how those ought to be regulated, and historically, FDA has not regulated LDTs. How is it and why is it that labs are having to go through the EUA process here? 

Lauren Silvis: FDA's approach has been laboratory developed tests generally are under enforcement discretion and don't need FDA review, but in the context of a declared public health emergency, when the declaration is made that there can be EUAs, then once that process is in invoked then even LDTs need to seek EUAs. That had been the policy before COVID. 

Greg Levine: Lauren, you mentioned this coronavirus situation currently as being very different from what the agency faced with respect to diagnostic testing for the Zika virus. Why is this such a different situation currently? 

Lauren Silvis: Because there the initial need for testing was more localized. There were labs running that testing, and CDC was running it mostly in Puerto Rico and the Southeast. And then you didn't need the nationwide testing stood up right away so quickly the same way we needed it here across both the public health laboratories and then hospital-based testing as well. So, the needs here were much more accelerated and nationwide from the beginning, whereas Zika could roll out more step-wise. 

Greg Levine: I wanted to turn now to some of the topics that are raised in your recent commentary article in the Wall Street Journal. In that article, you and former Commissioner Gottlieb discuss the notion that widespread testing will be an essential element in avoiding a second wave of coronavirus infections, and I'm wondering if you could just explain a little more about your thinking on that. 

Lauren Silvis: Sure.  Well, in order to deal with the increasing cases that we're having now, and certainly going into the fall, we need widespread access to testing and contact tracing. There has to be a larger public health backdrop to all of this. But there may be opportunities to expand testing, not just at health care sites and pop-up sites, but for employers to offer it and to have a number of different options and a number of different types of technologies that are used for testing so that we don't have the same waits and lines, and that we have different ways of testing people based on their need, based on local outbreak, based on their risk of exposure, based on what's happening in the community. And so the more options there are, I think the better place we'll be in as we head into the fall when we do think cases will rise. 

Greg Levine: And then on this topic of “return to work,” which is the main focus of your article, what is your view on the role of diagnostic testing and what that adds to the use of lower-tech options that, generally speaking, are going to be lower cost? 

Lauren Silvis: The low-tech solutions are incredibly important, and those are social distancing, wearing masks, and those kinds of tools are incredibly important in terms of preventing the spread. And then a step above that, the more we have an ability to figure out who needs a diagnostic test or how to incorporate it in a return-to-work plan so there's some understanding of how it can be useful. So, whether it's an app that identifies signs and symptoms but also potential exposures, or that has contact tracing in the office so that if there is someone who does have COVID and you know who they might have interacted with, those are all tools a step before the testing that can be extremely helpful in figuring out when you need diagnostic testing for members of your workforce. 

Greg Levine: To the extent there is going to be a role for diagnostic testing of employees, would this vary by the type of employer? What are the different considerations that an employer should bear in mind in thinking about whether to adopt some type of a diagnostic testing as part of its program? 

Lauren Silvis: It certainly varies by the work setting and the risks associated with your workforce. Is it an office setting where people are not coming and going and there is the ability to social distance? Or is it a retail setting where there's a lot of interaction and even though there may be masks and barriers, you are dealing with customers, so there's exposure? It's also depending on what’s going on in the local community, such as frequent use of mass transit. So there are a lot of different factors in the workplace that can lead to how to you want to develop a testing program and what your risks might be. There's also factories and people in health care settings who are serving patients, so even if it's not treating COVID patients, there may be higher risk or you may want to take more precautions than testing those workers. So there's not a one-size-fits-all solution for employers – there's just ways to think about the potential risk of exposure of your workforce, the risk of transmission in the work site, and what other tools you can build in, like de-densifying the offices, to help figure out when there should be testing and how much testing you might need. 

Greg Levine: And then how frequently might an employer test their employees? With molecular-based testing, the result is going to tell you only whether you're currently infected, but there ordinarily is going to be some time lag between when you take the test and when you get the result. And then of course, after you have that test and that result, you could get infected later, so should employers be periodically re-testing employees, or how would this work in practice? 

Lauren Silvis: It also depends on what kind of test you're using. A rapid test can give you results right away, but it's not going to be as accurate, whereas a PCR test that might be more accurate, you would have to wait for the results. And whether you're collecting the samples on site and sending it to a laboratory off site, or whether you're having employees use a home test, a home sample collection option, where they're getting a kit shipped to them at home and they can collect a sample and send it back to the lab. So if you're talking about frequency and scope, you're also building who am I testing and when – you're not just testing every single person every single day with a test outside that's rapid right away. You have to think about, “Can I test cohorts of my employees so I'm getting a sample of different cohorts that interact? Or can I test my higher-risk employees maybe once a week or every other week if I want to offer regular testing to them because they might be interacting more with others?” So the frequency depends on the factors I was naming before, that if you are in a work environment that allows for a lot of social distancing, then you may not need to test as frequently or offer testing as frequently. And then, if you're somewhere in a factory setting where there is fewer ways to social distance at work, you may want to test more frequently because if there's an outbreak, it's also more likely to spread. It's hard to prevent a case of COVID from entering your workforce, but you can do a lot to prevent an outbreak in your place of employment. 

Greg Levine: And what about antibody tests – is there a role for antibody testing for employers? 

Lauren Silvis: Antibody test may be useful for individuals in places where there were known large outbreaks and a lot of people were not able to get diagnostic tests right away. New York is a good example that if you got an antibody test there and it came back positive, the general recommendation with antibody tests is that you should get a second test by a different type of test because they can generate false positive results. If you got two positives in a place where there was a lot of spread by two different tests, then you can consider yourself positive. But we're not at the point yet where there's enough widespread antibody testing and consistency of results that there's some sort of green path that you can go back to work – we're just not at the point yet. So I think it may be helpful to people in determining their individual health status, but for employers, we don't have enough understanding of all the different tests out there and their performance to know how to incorporate that necessarily into a return-to-work plan. But it certainly can help people figure out their risk and the activities they may engage in if they've taken these tests and gotten a positive result.  

Greg Levine: Turning to a couple of areas that the FDA has been getting a lot of questions on and just recently has issued some technical guidance on. One involves the testing of asymptomatic populations, where someone doesn't have signs or symptoms of COVID, or isn’t known to have been exposed to someone who has the disease. To this point, there are no tests that are currently authorized specifically for that use, but the CDC, as we know, has guidance recommending use of authorized tests to test asymptomatic populations in some circumstances. I'm just wondering about that disconnect between the FDA authorizations that have been issued to date, what the CDC has been recommending, and whether you are aware of any concern about this by employers who are considering this kind of testing, and just more generally, what the state of play is on this issue? 

Lauren Silvis: I think the concern is that we don't have good data yet on the performance of tests in asymptomatic populations, and I think FDA's been pretty clear about that. What they've said is, look, "Tests can be ordered by health care professionals based on their medical judgment for whoever they think should get a COVID test, and that could include asymptomatic people." But the issue is we don't yet have a large data set on the performance on a test solely in an asymptomatic population. And so they issued this template – they'd like to see that data come in. So you certainly can test an asymptomatic person, but overall, this if is going to be done widespread, we don't know what the potential false negative rate is, or other concerns about the performance, or what samples or specimen collection we should be using that's better. We just haven't generated that data yet, and so they’re pointing that out as a concern as opposed to access issues. Earlier on in the outbreak, states didn't have enough testing so they were restricting by their recommendations who should get a test. Now, there's enough capacity so that more and more people can get a test, including asymptomatics, but the issue is more the performance in those populations. 

Greg Levine: The other issue that was subject to very recent FDA guidance is the use of pooled samples. There’s been a lot of interest in testing in this fashion, where you pool samples from multiple patients and then test individuals afterwards if there’s a positive result in that pool. But to date, as with the test for asymptomatic populations, there are no tests that have been authorized specifically to be used in this manner. I’m just curious about whether this is something to your knowledge that’s already happening? Is this something that people are waiting to implement until there's actually a FDA authorized test to use in a pooled sample? Just curious what you've heard on that subject? 

Lauren Silvis: Scientifically, pooled sample testing has been around for awhile – it's done by blood banks, there have been other recommendations for how to do it, and it's been done in other countries for COVID. So I think FDA did a good job by issuing the template while they're getting a lot of questions. Previously, FDA had been issuing templates for submissions after there had already been some authorizations, and so now they're doing it up front so sponsors know what sort of studies to do and what data they need to collect. I think there's a lot of interest in it from an efficiency standpoint. FDA has said in areas where there's suspected low prevalence, it's a good way to be more efficient, test multiple samples at once and clear people as not having had it. If there's a positive, you'd need to test the individual samples. And their concern is the low positive sample being missed, and so they want to see really good test performance if you're doing this because presumably you're testing a lot more samples at once in multiple runs than you would be. Therefore, if your test isn't performing as well, you'd miss a lot more positives. So I think that they've put out some suggestions on how to develop these and are looking for sponsors to come in with that data, and I think there's a lot of interest in doing this because it would be more efficient as we think about continuing to ramp up testing. 

Greg Levine: Great.  In your article, your commentary piece in the Wall Street Journal, you talked about next generation sequencing and that your company, Tempus, as well as Illumina, and there may be others entering in this area as well, are using next generation sequencing technologies for COVID-19 testing. We hear a lot about NGS testing and I’m wondering if you could just explain what that means – how that technology is different from non-NGS testing?  And then what is the added value that that might bring in the context of COVID-19 testing? 

Lauren Silvis: There's a lot of different ways that NGS can be deployed. You can sequence the virus, so you can over time, for research, identify different mutations. You can also use NGS panels to distinguish between COVID-19 and other respiratory viruses because you can use NGS for other purposes. These are really powerful machines that can be really efficient in terms of doing more pooled sample testing and generating more and more results more efficiently. So there's a bunch of different ways that it can be deployed by labs who have the NGS platforms. 

Greg Levine: Maybe we can also talk a little bit about the use of software apps. What is the role of apps, do you think, in diagnostic testing and results reporting, for example? 

Lauren Silvis: I think apps can be used a number of different ways. They can certainly help people figure out if they should get tested, based on how they're feeling or potential exposures or what's going on in their community, and results can be reported back to a patient on an app. If you register, there's different labs and companies that are reporting results back so they can provide information – not just your results, but other information about how to manage your condition, who else you might want to be in contact with. And then there are the contact tracing apps that are available. I think folks are going to have to become comfortable with contact tracing and figuring out who they might have been in contact with, both to understand if they were exposed and also if they have COVID, understanding that for public health reasons, we need to think about where else it might have spread. And that's what we do in these kinds of epidemic situations. 

Greg Levine: And to my knowledge, the FDA has not been, itself, regulating these apps actively. Are you aware of any FDA act of regulation of these kinds of apps? 

Lauren Silvis: I can't think of any that would meet the device definition and need to be regulated under an EUA. I think there's a number of helpful tools, but I haven't seen any that would be regulated by FDA. I don't know – there could be sponsors working with FDA on it– but I'm not sure. 

Greg Levine: Thanks, Lauren. I know in your past, I didn't talk about it in your bio at the beginning, but you’re a lawyer in your background, and you formerly were a partner at a law firm. I know you're not in a legal function per se right now, but in your work at your current company, are there other legal considerations that you deal with, issues like privacy, or any of the reporting obligations? What kinds of issue do you deal with at your company or do you find that your customers or potential customers are struggling with? 

Lauren Silvis: I know that there are a lot of employment law issues related to this and that varies by state. From my perspective, given my HHS experience, in addition to the FDA regulatory issues, we also focus on our laboratory compliance. And so if you're an employer, you are going to have the tests done by some laboratory or some health care professional, even if it's being done with the on-site rapid tests. The results do need to be reported to state and local public health departments. Now, it's generally the laboratory that takes that on – so for example, we can test employees and then we do the state reporting as needed. From a public health perspective, CDC and HHS have issued more guidance about what sort of information they want to see reported so they can better understand the demographics of who's getting COVID, how it's spreading, and where the cases are. That's a lot of reporting obligations, but that falls on the lab, not something that the employers have to do, but they have to make sure is done. But what an employer does with the actual results, we've seen all different plans for that – how the results are returned, whether employees opt-in to having it returned to the employer, and then what sort of tracing or follow up is done for workplace safety from the employer's perspective really varies. And one of those times it makes me happy – I don't have to actually practice law in that area. 

Greg Levine: I'm sure that's comforting- that's great. So one closing question for you: I’ll ask you to put your FDA hat back on, or your former FDA hat, I guess. But just in this area of COVID-19 diagnostic testing generally, what do you think your former FDA colleagues are going to be most focused on in the upcoming weeks and months? 

Lauren Silvis: The real world evidence of how the tests are performing and understanding the different platforms that are out there and if there are different considerations for the different platforms with different samples, with different modifications. I know they're very concerned about it because when you authorize something, you put it on the market based on the benefit-risk calculation at the time that there was a widespread need for diagnostics, but understanding the overall performance of these tests I know is something on their mind, and I think it's the right place for them to be looking. I know folks in the community are looking at generating the kind of data they need to understand the overall benefit-risk because COVID testing will be with us for some time. And so usually, you'd see the tests improving over time as we do more and more testing and understand their performance in different populations more. 

Greg Levine: One of the things that's been striking to me is the frequency and speed with which the FDA has been updating its guidance, whether it’s guidance documents on different topics, or here, of course, with the diagnostic tests, we've seen them pull certain tests off the market, or very quickly crack down on certain players that were promoting their tests for home use and things like that when they weren't authorized. I've never seen anything quite like it in my career. I'm just wondering when you were working on Zika and other subjects, did you see FDA have to spend so much time reevaluating its decisions and course correcting as it was going along? 

Lauren Silvis: I have seen FDA over time in EUA situations work to get these tests permanently authorized and have them come in for permanent approval and meet the longer-term safety and efficacy expectations, so that would be something that I expect sponsors to do. But the pivoting in approach and issuing new templates so rapidly, no, I can't remember a time like that. I can't remember a time where there's this 24/7 response mechanism that they set up at CDRH, so I think my former colleagues are working incredibly hard. But in a public health crisis, that's what they're there for and that's what they do. 

Greg Levine: Yes, absolutely. I think as we said earlier on, this situation is unprecedented, in our lifetimes anyway. Well, thanks so much, Lauren, for your time. It's been great talking to you, and I really appreciate you taking the time out to participate in our podcast today. 

Lauren Silvis: Thanks, it's been great to talk to you. I think these are important topics and a lot to think about, and it's always good to reconnect. 

Greg Levine: Thank you to our listeners today. That’s all the time we have. Thanks for tuning into our podcast, Non-binding Guidance. For more information about our practice or other topics of interest to life sciences companies, please visit our FDA regulatory and life science practice group pages at www.ropesgray.com. You can also listen to Non-binding Guidance and other RopesTalk podcasts in Ropes & Gray's podcast newsroom on our website, or subscribe to this series wherever you listen to podcasts, including on Apple, Google and Spotify. Thanks again for listening.