Hold on Antibody Testing: 'The FDA Has Done Us a Disservice'

John Whyte, MD, MPH; Michael Saag, MD


May 06, 2020

Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

  • Whether the presence of antibodies to COVID-19 grants immunity remains unknown.

  • People are probably not getting reinfected, and any positive results on retesting are more likely a result of residual viral RNA.

  • The FDA's relaxed standards and emergency use authorization for antibody testing was "a disservice to us" because of potential cross-reactivity with other coronavirus antibodies.

  • For antibody testing to work in a disease that affects 2%-3% of the population, we need a test with at least 99.7% specificity.

John Whyte, MD, MPH: You're watching Coronavirus in Context. I'm Dr John Whyte, chief medical officer at WebMD.

Coronavirus can be scary. There is a lot we don't know. Should you get tested? What's the role of antibodies? What treatments actually work? When will a vaccine come? And how long do we need to stay in place?

Despite all of this uncertainty, there are some things that we do know, and we need to rely on sense. That's what's missing from much of the discussions: a common-sense approach to dealing with coronavirus. That's why I think you'll want to listen to my next guest, Dr Michael Saag from the University of Alabama at Birmingham. He offers the common-sense insight that we need to make good decisions.

Dr Saag, thanks for joining me.

Michael Saag, MD: Thanks for having me.

Whyte: Let's start off with who should get antibody testing.

Saag: Ideally, the prime people are those who feel that they've been exposed to COVID-19 and did not have symptoms, and they're suspicious that they may have been infected but are asymptomatic. If the test is working well—and that's a big if—then that would tell them that they had the infection and may be protected by immunity.

Whyte: Do you think the presence of antibodies to COVID-19 does grant immunity? There has been a lot of debate about whether or not that's the case. What do you think?

Saag: Well, I hope it does, for a lot of reasons.

Whyte: We all hope it does.

Saag: Right. I've had COVID, and my antibodies are through the roof. In fact, tomorrow I'm donating plasmapheresis so that it can be used in research. But the take-home point is that we don't know for sure. And I can argue with equipoise that it would, from other viral infections such as measles and mumps. But then we look at the other extreme, dengue fever. You get antibodies, but not only do you get reinfected, your disease can be worse on the reinfection. So, somewhere on that spectrum, this will fall.

My personal stance so far is that we're not seeing any cases to speak of where somebody had it and then got sick again. I think that as more time goes on, we'll answer the question, but right now, we don't know for sure.

Whyte: Do you think there's more evidence than not that it does give some degree of protection?

Saag: Leaning that way.

Whyte: Okay.

Saag: But we can't say more right now.

Whyte: What about these reinfections that have been reported? Most people are saying that those are issues of testing and not of reinfection. It's a simple RNA virus; it's not like zoster. What are your thoughts on that?

Saag: Well, if we dig into the weeds a little bit on those stories, they were mostly people who tested positive for virus, then tested negative, and then tested positive again.

Whyte: With symptoms, correct?

Saag: Without symptoms.

Whyte: On the retest?

Saag: On the retest. They had been sick, they had virus, they got better, they tested negative, and then they tested positive again later. That's my understanding.

Whyte: Okay. Why were they retested, then?

Saag: I think they were looking to see if there was persistence. And so the issue is, was it truly a reinfection or was it just persistent RNA that they picked up a second time later on? I don't know the answer, but my feeling is that it probably [wasn't reinfection]. The way to really know for sure is to do molecular virology, where you do sequencing and see if it's related.

But right now, I think common sense is telling me that people are probably not getting reinfected, and that any residual testing that's positive—and some I've seen going out to almost 5 weeks—is due to residual RNA.

The question is, is that an infectious virion that's being picked up or just a remnant? So far, the culture tests show that by day 14 after onset of symptoms, you can't culture virus anymore. So I think it's probably remnant.

Whyte: And some of it may relate to the timing of testing, correct? We don't always hear when exactly they were tested.

Saag: That's exactly right.

Whyte: I appreciate your common-sense approach. You say we need some common sense. So I want you to make sense, if you can, of what we're hearing about the number of antibody tests that are out there. I've worked at the FDA, so I know that in authorization, there's a term "in the weeds of regulation." It's a lower level of standard. They don't even have to get all of their data completely done before coming on the market. We're in a public health emergency. We want to get things out there. But that has caused challenges with the accuracy of tests. How concerned are you about the accuracy? And how would I, as a patient, if I wanted to go get one, know that I'm getting a good one?

Saag: Start at the end and say you aren't going to know, and that's a huge problem. And I get it. I understand why the FDA relaxed and went to early authorization use. But I think in retrospect, it was probably a disservice to us, from a common-sense perspective. Here's why.

We all have been infected with a coronavirus in our lifetime, with common cold. Not every common cold, but a lot of common colds. That means that we have antibodies to coronavirus in our bloodstreams that are just hanging around. If the test for SARS-CoV-2—COVID-19 virus—antibodies is not really specific for that particular virus, there will be cross-reaction from these other coronavirus antibodies onto that test.

So, somebody could get a positive read for an antibody to coronavirus and assume that it's because of SARS-CoV-2 and think, Maybe I'm protected. I can relax. And that's a disservice. So a tip of the hat to the FDA for what they do every day that we take for granted. They insist on rigor before they approve things. And by relaxing those standards for this, though I understand why they did it, I think they've done us a little bit of a disservice from a common-sense perspective.

Whyte: Does it matter if I get a pin prick or I go to a lab and have my blood drawn? Is the blood draw a better test?

Saag: Absolutely.

Whyte: Why is it better?

Saag: Because with a blood draw, you're getting plasma. You're getting a large volume that can be run on a test. With the pin prick, you're getting just a drop or two of blood, and it's less likely to be accurate. So I think that for right now, what we need are standardized tests with published false-positive rates.

I'm going to go into a little statistics here. Let's say we have a test that's got a 3% false-positive rate, which is 97% specificity. If we apply that to a population that has less than 5% prevalence—so the majority of people haven't had it—that test, in terms of what we call positive predictive value, could be wrong 50% of the time. That's not helpful.

Whyte: So it depends on the number of people who have the disease in the population.

Saag: Exactly. For a test to really work well in a disease that is only affecting 2%-3% of the population at most, we need a test that is at least 99.7% specific. And I'm not sure we have that proven yet. So, the take-home point is that I would hold tight on getting antibody tests until we have a lot of validation.

Whyte: You were infected with COVID-19. You graciously agreed to share your story, and we're going to hear what you did about testing. Tell us how you got infected with COVID-19 and what has been your experience.

Saag: It was an accident of sorts. My son, who lives in New York, was coming back to Birmingham. He was going to drive. I was in Boston. I took the train and met him. En route, he started not feeling well. By the day we got to Birmingham, he was having fever. We looked at each other. We knew that he almost certainly had COVID.

We quarantined ourselves in the house. I got sick the next day. Not terribly bad, but by day 5 and 6, it was pretty awful. And as a physician, as a provider, the worst thing was knowing what would happen if my breathing deteriorated. I would be in the hospital, possibly on a ventilator. And that was a nightmare scenario for me. So, every night for 8 nights in a row, I suffered through that anxiety of worrying about it. Fortunately, by day 14, it went away.

Whyte: Did you and your son get a test?

Saag: Yes. We were tested virologically on day 3 of my illness, when we both tested positive for the virus. So we definitely had it.

Whyte: That's a success story, as most are success stories.

Saag: Yes, they are. We volunteered for a research study looking at immune response on the day we got tested for the virus. That study has continued, and we sent the blood off to colleagues of mine at the University of Pennsylvania, who have made a very specific test for research purposes. I have an [antibody] titer of about 30,000, which is really high. So tomorrow I'm being plasmapheresed so they can use it for a lot of test validation.

Whyte: What about the role of convalescent plasma? Is there value in that in the treatment of critically ill patients? Is that something that you're considering?

Saag: Yes. It seems, [based on] early reports, that it is helpful in some patients. My son made a plasma donation, and I don't know how many patients received it in New York. I elected—because of the [high] titer—to first donate for research, and then, if needed, I'll donate for patient plasma as well.

Whyte: You've mentioned common sense throughout this, and I love that description. What common-sense advice do you have for viewers who are concerned about coronavirus? We're several weeks into the epidemic. Most people think we've reached our peak and are on the downward trend, though it's still a serious condition. What common-sense advice do you have for folks?

Saag: Common sense is that all this virus cares about is survival and replication. And it's still here. We have the vast majority of citizens in our country who are susceptible. We've been successful at putting a wall between [the virus and these susceptible people] with these stay-at-home orders. If you take the wall away, what does common sense tell you? These things will come back together, and we're going to get another surge in cases.

We have to be smart and be able to keep [susceptible people] away from the virus. And that can involve a series of measures. But the fact is that this thing is here, and it's not going away any time soon. We all need to start getting that into our heads—not just awareness, but to the point of comprehension. What does it mean to be living with this virus in our midst for the next several months? It's going to take sacrifice.

Whyte: Is a vaccine going to be here in 6-12 months?

Saag: No. Even if we had one that we knew worked, we're going to have to mass-produce it. If there are 300 million people who will need the vaccine in the US, and if you assume two doses, that's 600 million doses. And this is a worldwide pandemic. Even if we had a vaccine, we aren't going to be able to scale up to that degree of production for at least 18-20 months.

What's the optimism? Hopefully, some of the drugs that are in development will be shown to work. That would be a game changer because, even if someone got infected, we can catch them early, like we do with oseltamivir (Tamiflu) for influenza. If you catch it early, intervene, and abort the symptoms, and then the person still develops immunity but they won't get as sick, that's the hope.

Whyte: Well, Dr Saag, I want to thank you for sharing your insights.

Saag: Thank you for having me.

Whyte: And I want to thank you for watching Coronavirus in Context. I'm Dr John Whyte.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.