This transcript has been edited for clarity.
Abraham Verghese, MD: Welcome to another episode of Medicine and the Machine. It's a great pleasure to be here with my co-host, Eric Topol. Today we have as our guest someone I've looked forward to listening to every Wednesday morning without fail when he gives the COVID update to us at Stanford. He's been a linchpin in our abilities to respond to this epidemic locally, in Stanford, and also all over California.
Our guest is Benjamin Pinsky. He's an associate professor at Stanford, the medical director of the clinical virology lab for Stanford Health Care and Stanford Children's Health, and the medical co-director for Point of Care Testing. He got his MD/PhD at the University of Washington, and he was at the Fred Hutchinson Cancer Research Center in Seattle before coming to Stanford, where he trained in pathology and molecular pathology. His focus of interest is molecular infectious disease testing, and he has done a lot of great work on dengue and Zika, among all kinds of other infectious diseases. I can't think of a person better situated to address COVID testing. Welcome to the program, Ben.
When did you first hear about COVID? How did you get your head around it and anticipate what was coming?
Benjamin Pinsky, MD, PhD: First, thank you so much for having me on the program. This is a great honor to be able to share my story.
Initially, like most other folks in the world, I saw the outbreak occurring in China in early 2020 with the news reports on television. As you mentioned, I'd had experience with the Zika outbreak, which was not so concerning for the United States. And as a resident, I had been involved in our laboratory response to the 2009 H1N1 pandemic, which was almost as concerning as the COVID pandemic was initially but turned out to be not as severe. At that point, I saw that testing was very important in all those different outbreaks. And I realized quite early on with COVID that we were going to need testing and we were going to need it rapidly.
Verghese: I remember how cumbersome it was to get the Centers for Disease Control and Prevention (CDC) to do the tests because they were the only ones approved to test. But meanwhile, if I recall, you began to conduct testing here and then marshal resources from all over Stanford and do the tests for all of Northern California. Can you talk about those early pioneering days?
Pinsky: Those early days were quite challenging and exciting at the same time. I saw reports about this novel coronavirus on the news and was concerned that this was going to be coming to the United States sooner rather than later. The sequences from the virus were published very early on in January 2020, and then a number of groups published, on the World Health Organization (WHO) website, primer and probe sets for detection of the novel coronavirus. That, I believe, was in maybe the second week of January. Shortly thereafter, my laboratory — I have a small research laboratory that does translational infectious disease diagnostics development — ordered the primers and probes, got those into the lab, and started validating the test at that point with synthesized nucleic acids.
We couldn't get our hands on actual clinical specimens at that time. We had a test based on the test from Christian Drosten's group in Germany, which targets the envelope gene. We had that up and running by early February of 2020. At that point, we started retrospectively screening our samples because, as you mentioned at that time, testing at the CDC and local public health labs was limited. There were significant requirements that changed over time depending on whether someone was a returning traveler or had contact with someone who was known to be infected with SARS-CoV-2. So, we started looking back and seeing if we could find community transmission that was going unnoticed, because in addition, that was in the midst of our standard respiratory virus season and lots of folks were coming in with respiratory illnesses that were similar to SARS-CoV-2.
We started doing retrospective screening and then prospective screening after we caught up, and that was throughout the month of February, before widespread testing was available. We captured two cases of very early community transmission of SARS-CoV-2 by this method. Unfortunately, at that time, we weren't allowed, as a clinical laboratory and based on the regulatory requirements, to provide that testing for clinical purposes. That changed when the US Food and Drug Administration (FDA) and the CDC recognized that there was a deficiency in testing available throughout the United States. On February 29, the FDA came out and said, with their guidance, clinical laboratories able to perform high-complexity testing could now start to perform testing and then submit applications for emergency use authorization. We immediately did that and had a test up and running by early March; I believe ours was one of the earliest clinical laboratories able to offer testing in the United States and certainly in the Bay Area.
Eric J. Topol, MD: It's amazing how you were out in front of the virus. But some of the academic centers that tried to do what you did were shut down by CDC and FDA — for example, University of Washington and one of the Harvard sites — because not only did CDC and FDA not have a test, they didn't like the idea that academic centers like yours would go ahead with this. Did you ever get threats to desist during this stretch?
Pinsky: I did not get threats to desist from any of the public health authorities — local, state, or CDC. I had some interesting conversations when trying to get some samples tested. That was challenging because the samples did not meet the requirements, as I mentioned before, for their current rules about testing. We never used the CDC assay, which I think was fortunate. It turns out this envelope gene assay works quite well and has maintained its performance throughout the pandemic, despite all the different variants. I didn't get any cease-and-desist comments from CDC. I kind of flew under the radar a bit and just got it up and running and asked for forgiveness later.
Topol: So, on February 29, you got the go-ahead from FDA. On March 6, at a historic meeting of the press, the CDC, and the triumvirate of Donald Trump, Robert Redfield, and Alex Azar, Trump said anyone can get a test. Anytime. They're beautiful. But there were no tests and the virus was spreading like wildfire throughout the entire country because of the lack of testing. It's just incredible, the lies that we're being told to the public throughout this time. It's how we got so far behind. I don't know if we had further thoughts about that, because you were watching what you were doing and what the government, including the top leadership, was telling us, which were blatant lies. Were you thinking about that?
Pinsky: At the time, I knew that we needed tests, from my past experiences dealing with how tests get rolled out during epidemics or pandemics. I knew that it was on the clinical laboratories to pick up some of the slack and that the public health authorities could only do so much, particularly in those circumstances. So we were focused on getting our testing up, making sure we could first provide testing for the Stanford patients and the Stanford community, but then also provide testing for what turned out to be the entirety of the Bay Area for approximately 6-8 weeks, in those very early days. We were getting samples from UCSF, from Kaiser Health, and from the various counties and other hospital and clinic facilities. We were also trying to make sure we had enough reagents. That was a huge problem throughout the first year of the pandemic, and it has popped up occasionally since then, during the last surge and the Delta surge last summer.
Verghese: I'm struck by your curriculum vitae showing how active you've been, with papers just rolling out on SARS-CoV-2 — incredible stuff, both on the antibody side and on surveillance testing, and the nasopharyngeal tests. Step back for a moment as a virologist speaking to our audience, which comprises largely nonvirologist, non–infectious disease types. What are your thoughts on this virus compared with some of the other viruses you've focused on, from influenza to dengue to Zika? What makes this virus special? Some thoughts about the enemy.
Pinsky: Thanks for mentioning some of that other work. A lot of that has been because of great collaborations with lots of folks across the country and particularly at Stanford. Much of the antibody work was done with Scott Boyd, a talented B-cell immunologist, and folks in his lab, Katharina Roeltgen in particular. That's been a great collaboration. He developed our antibody assay very early on. I think we had that up in April or May of 2020. We've been trying to be at the forefront of testing, and that's most of what those publications have been about.
SARS-CoV-2 has been an impressive foe. Perhaps we had hints of this with SARS-1 and MERS, with the significant disease those could cause. Combine that with a virus that is so easily transmissible between individuals and continues to vary, based on different selective pressures that are changing constantly. SARS-CoV-2 is a challenging infection to deal with.
Topol: Thinking about its evolution, it went almost the whole first year going from the ancestral virus to the D614G variant. It took almost that whole first year to see a significant variant with heightened transmissibility. And now, of course, we're seeing what appears to be much more substantial and perhaps accelerated evolution. Can you trace that? Even some people in this field thought, Well, maybe this virus won't go through rapid evolution and we'll nail it, and the vaccines will be great. But it turned out to be a different story, as it's still unfolding.
Pinsky: I don't pretend to be an immunologist, but this has been fascinating, monitoring the emergence of the subsequent waves of variants and all the mutations and sublineages. It's a field day for the taxonomists.
I'm not exactly sure what this underlying mechanism is for. You're describing this year of a pretty stable virus, followed by all of these different subsequent waves of variants — whether it's differences in immune selection, I'm not exactly sure. In the lab, we've been focused on making sure that we're able to detect these variants as they arise, because important clinical decisions need to be made, particularly around the use of monoclonal antibody therapies. Also, various public health decisions need to be made in a timely fashion, if we have those data quickly.
One thing we've done is targeted mutation analysis; this allows us to get the results quite a bit more rapidly than if we were to sequence every single positive that came through the lab. We've been quite on top of those transitions between the emergence of Delta, for example, and then Delta to Omicron over this past winter. Now we're seeing that transition to BA.2.
The idea of these variants arising in immunocompromised individuals is another interesting area. We and a number of other groups have seen this. We saw an individual who had uncontrolled HIV and developed spike mutations in about 2 weeks, which was impressive. One wonders if that's a reservoir for some of this massive replication and development of mutations when there is suboptimal immune surveillance.
Topol: You touched on a few things I'd like to pick up on. This is the first time I can remember where sequencing or whatever method or assay, if it could be done quickly, could tell you which monoclonal antibody to use to treat a person. Unfortunately, we don't have that at the individual level. We have an approach like, Oh, well, we're kind of in a BA.2 wave, so you better not use this antibody. This could be truly individualized medicine, but we don't have the capability at the bedside or rapid point of care. Any comments about whether we could get there? Because this is a major switch from how we're used to treating an infectious disease.
Pinsky: Absolutely. Most of the sequencing of viruses has been for surveillance and perhaps, as with influenza, to make decisions about what vaccine to use in the next year. There is the possibility of using this testing more specifically to impact patient care immediately. Certainly, one can take an approach such as we've done, which is standard, allele-specific PCR to look for mutations that are associated with particular variants. One can make fairly accurate variant calls based on a small number of mutations, at least up to this point. I know a number of companies are working on more comprehensive panels that could potentially be run in a high-complexity clinical lab. I'm not aware of folks trying to bring those into point of care, and maybe that's not necessary, but if you could turn this around in a day, for example, that should be enough time to inform that selection and get someone the right monoclonal antibody.
It's particularly important at those transition points; we were getting tons of questions in December and January of this past winter when we went from Delta, where folks had the Regeneron antibody, to Omicron, where we needed to use sotrovimab. Of course, that will change with every subsequent wave and subvariant. In those cases, we were on top of it and reporting frequently to grand rounds and to the chairs.
They'd say, we can give this person the Regeneron antibody but we need to know if they have Delta. At that point, what we were seeing was 50/50. So, you can really make a difference in those decisions. I believe when we get to 100% BA.2, then it's less important that it's reported on an individual level.
Topol: The other question I have is about the immunocompromised hosts. They are central to the leading theory for how Omicron developed in southern Africa and that, not only was there accelerated evolution of the virus in a person who was immunocompromised, but then that person infected others. It seems that we're in a vulnerable position because we have tens of millions of immunocompromised hosts around the world. We have a virus that isn't contained. We have animal reservoirs, with spillover to humans from everything from pet hamsters to white-tailed deer, and all these other species. In addition to all that, and what I wanted to ask you about, we have an emergence of lots of co-infections with different strains, and we have new hybrid fusion recombinants. We have potential for a recombinant between animal and human, as was the case with bird flu.
So, the calculus here doesn't look good going forward. What are your thoughts?
Pinsky: I think there's a lot of potential for more transmissible and potentially more pathogenic versions of this virus to arise, given all those possible reservoirs out there. Fortunately, in the HIV case I mentioned, there was no evidence that the individual transmitted to anyone else. And then once they were on antiretrovirals, they actually cleared the virus very rapidly. So that was good news. But, as you said, there are millions of immunocompromised folks throughout the world and there are all these animal reservoirs, so I always worry about things like this, which is why I want to have the test out there. I don't want to be alarmist, but at the same time, I think we must have very good surveillance for this and understand that these could be problems in the future.
Topol: Can you talk a bit about bird flu? That illustrates the human-animal and human-human co-infections and these recombinants we've seen with Omicron — this animal-human combination.
Pinsky: One of the other hypotheses for the emergence of Omicron is that it may have not only passed through an immunocompromised individual, but from an immunocompromised individual perhaps to an animal reservoir and back to humans. And in humans, we've seen, as you mentioned, Delta and Omicron recombinants because, I think, of a switch-over during discontinuous replication. We also see various recombinants between different Omicron lineages like BA.1 and BA.2. I know at least one is named, but I believe there are a number of those. Then certainly for influenza — the 2009 swine influenza was a recombinant between avian flu, human flu, and swine flu, with all the different segments.
Of course, coronavirus is not a segmented virus, so it wouldn't have that same sort of reassortment. But certainly, this discontinuous replication can occur, and you could imagine some potentially devastating consequences were it to combine with some other bat coronaviruses or something like that.
Verghese: You've done a lot of studies on antibodies, antibody responses, and types of antibodies in nature, both in individuals and community surveys. Can you summarize the state of the art in terms of understanding the antibody response to COVID and testing for it? Is it meaningful to test?
Pinsky: A number of professional societies have come out with statements saying that serologic testing for SARS-CoV-2 is only useful in very specific circumstances. Some — the Infectious Diseases Society of America, and others — don't recommend it in most cases. As I mentioned, we use serologic testing a lot here, because we have such a large immunocompromised population, mostly transplant, and we're trying to understand whether and how those individuals are responding to vaccination, making a sort of risk assessment for how they're going to behave as society opens up again. So, a lot of our testing is on those immunocompromised individuals, to learn how they're responding to vaccination.
We recently switched to a new assay, a quantitative serologic assay that looks at multiple different targets. It has a full-length spike receptor–binding domain, which will identify people with natural infection as well as those individuals who were vaccinated with a spike-based vaccine. It also has nucleocapsid so we can distinguish between natural infection and vaccination. That's been quite interesting and helpful, in addition to having the quantitative aspect. This test was used in a number of the clinical trials for vaccine approval. So, we have at least some correlative protection. Of course, the 95% CI starts at 1.0 and goes to almost the end of the race. But at least we have the median values to give some idea of whether someone's responding as we would expect for vaccination.
Verghese: Are these standardized across labs?
Pinsky: These are standardized assays, using the WHO International Standard unit — binding antibody units per mL, so one can calibrate this assay to other assays that are reporting quantitatively. This will enable us to start to gather more data about the quantitative response and how this correlates to protection.
Of course, the antibody response isn't the only response. There's also the T-cell response. Niaz Banaei, at our laboratory, developed an interferon gamma release assay that monitors the T-cell response, so we are fitting that in and figuring out how that plays into protection as well.
Verghese: Lots of exciting stuff. I forgot to mention that you are the editor-in-chief of the Journal of Clinical Virology (JCV). I imagine the whole journal is now all about SARS-CoV-2, given how rapidly changing this is.
Pinsky: Yes, I'm the co–editor-in-chief with Bert Niesters, who's in the Netherlands. We get papers from all over the world, and it has been primarily SARS-CoV-2. Every lab is a COVID lab now, right?
Topol: I'm sure it added to your workload.
Pinsky: The first year of the pandemic was pretty crazy for the journal. I think for every journal, it's been a challenge to get reviewers because everyone is so busy dealing with their clinical work and their own research. There's so much to learn about this virus. It's calming down on the JCV front, but we're always looking for good papers. So, listeners, please send them along.
Topol: I'm interested in your perspective on the long COVID story, because this appears to be leading to some substantial effects that weren't anticipated with, respect to the brain, type 2 diabetes, the cardiovascular system, and many other systems. Has any other virus behaved like this in history?
Pinsky: That's a really good question. The spectrum of post-COVID disease is substantial, as you mentioned. Certainly, a lot of viruses can trigger inappropriate immune responses — for example, chronic fatigue syndrome. In some individuals, I believe that's analogous to long COVID, though there's still a lot of work to be done on that. But certainly, a variety of viruses have been associated with or pre-date the onset of a chronic fatigue–like syndrome. But with COVID, basically every organ system can potentially be involved, and I don't know that we've seen this with other viruses. This is quite unique, the combination of the various sequela and just how many individuals have been infected.
Verghese: Recently, your lab announced your one millionth COVID test. It's a staggering number compared with the early days, when we were counting patients in the Bay Area on the fingers in one hand. Now we're in this incredible situation where I don't think we know the number of infected. You're not getting to test all the people who have COVID because they're all at home and doing the rapid tests or not testing at all. We're in this no man's land right now. We have no idea what's going on. Am I incorrect?
Pinsky: That's definitely a concern. We hit our millionth test, and I know a lot of clinical laboratories around the country have hit that mark as well. Our testing volume has been pretty consistent during the past 2 months after the Omicron surge, which was massive. But then, with much wider availability of rapid antigen testing, it's not clear how many individuals are using that and not coming in to be tested through our laboratory. So we certainly have biased information based on the individuals being tested. In our case, these are a lot of transplant recipients because we're still doing pre-procedure screening of folks coming in for surgery, and perhaps less community testing. Our numbers are increasing a little, so we're keeping an eye on that. We don't need to be too worried yet, but we have been at about a 3% positivity rate for about 5-6 weeks.
Verghese: You've been a huge asset to the whole community, not just at Stanford but also the whole Bay Area. Sometimes in all the discussions of COVID among politicians and policymakers, we forget the critical importance of good diagnostic testing and understanding that testing. So, you have done a huge service to all of us, and I hope you can continue with the exciting work and keep moving the bar forward. It's a scary time, but it helps to know that someone like you is at the helm with our testing. Thank you so much for being with us.
Follow Medscape on Facebook, Twitter, Instagram, and YouTube
