How Long Are COVID-19 Antibodies Protective?

John Whyte, MD, MPH; Alexander Greninger, MD, PhD


September 01, 2020

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

  • In a new study, 85% of a Seattle fishing boat crew were infected onboard with COVID-19 in May. The entire crew had tested negative for infection and had blood drawn before departure. Upon return, the three people who had neutralizing antibodies before departure were not infected on the ship.

  • The big question is, how long are COVID-19 antibodies protective? It depends on the titers, or the concentration of antibodies, in an individual's blood.

  • People with more severe COVID-19 infection probably have more antibodies, which could possibly protect them from reinfection for a year or more. Milder infections lead to fewer antibodies and could possibly provide protection for up to 6 months.

  • Reinfection of COVID-19 is possible. Studies indicate that about 10% of people with mild infections show little immune response, increasing their risk for reinfection.

  • Reinfections are often milder and may be asymptomatic.

This transcript has been edited for clarity.

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

What's the role of antibodies against coronavirus infection? It's one of the biggest questions that we've had over the past 6 months.

To provide some insight, I've asked Dr Alexander Greninger, a virologist from the University of Washington, to join me. Dr Greninger, thanks for taking time to speak today.

Alexander Greninger, MD, PhD: Thanks for having me. It's great to be here.

Whyte: There was a very interesting study from a fishing vessel that leads us to believe that perhaps antibodies can provide protection against COVID infection. Can you tell us about this study?

Greninger: Sure. I work at the Clinical Virology Lab at the University of Washington Medical Center. We've done about 600,000-620,000 PCR tests for COVID as well as about 50,000 serology tests. Serology doesn't have much of a penetration compared to the PCR tests. People are much more interested in testing for active infection.

We do a lot of testing for different groups, and we were doing testing for the fisheries, trying to prevent outbreaks on boats. On a couple of the fishery boats there were outbreaks. In May, one particular boat had an outbreak that affected 85% of the individuals on board. With that high attack rate, it allows you to get statistical significance with not a lot of people if you're looking at protection.

Since we had blood from individuals before they got on the boat, and we had follow-up over 30 days after they got off the boat, we were able to look at correlates of protection before they got on the boat.

We saw that three individuals had neutralizing antibodies. Now, we don't necessarily know if they had prior infection, but presumably they did. They had neutralizing antibody titers that weren't super high, they did not get infected on the boat, and they are the only three people who had those antibodies, based on our algorithm.

Whyte: What does this tell us about the presence of antibodies and infection? This has been one of the biggest debates for the past 6 months.

Greninger: Well, it says that they're a correlate of protection, which means that they are a correlate, so you're not actually proving that the antibodies are doing the work. We use different levels of evidence and causality here based on some analogy as well as our prior experience that we think that antibodies are going to have a role here. Here we show that they were statistically significant and associated with protection from the virus. Based on the overall evidence, I'd be pretty comfortable saying the antibodies are definitely associated and probably protective. Then the next question becomes, for how long?

Whyte: We don't know that yet, do we?

Greninger: That takes time.

Whyte: What's your best guess as a virologist?

Greninger: That's a tough one. The hedge is, it depends. The reason it depends is that we see somewhere between a 50- and 100-fold difference in the titers of antibodies that people generate after infection. You see that big diversity of response. If you're going to see some declining titers over time, you would expect that it really depends on how high you started, and it's going to depend on different individuals. We've seen some reports and cases out of Europe and Hong Kong, potentially of reinfection. Those are individuals probably infected the least.

Whyte: I want to ask you about reinfection. Those were people with moderate disease, right? In your study, these were people with low to moderate titers — that measurement for people in terms of how many antibodies — yet they still were protected. What are you thinking about reinfection? Do you think it's possible? And I'm still going to push back on you in terms of how long you think antibodies may provide protection for us.

Greninger: If I had to throw a number on it, I'd say for probably about 6 months you should be able to get protection. For significant infections where you drive those titers quite high, you would be looking at a year, a couple of years.

Whyte: More antibodies probably means more protection, for those folks who [aren't experts in] virology.

Greninger: What we've seen from other viruses is that titer matters. The higher titer you start with, then the longer you're going to be protected.

Whyte: Do you think reinfection occurs?

Greninger: Yeah, I bet it does. This is not our work but other people's work, showing that 10% of people don't really show a very strong immune response at all in mild infection. Those people would be at a much higher risk for reinfection. Typically, those reinfections are milder and the people are usually asymptomatic. There's some memory response that gets jolted, and there's also some antibody and some T cells. So, that's all good news.

It's really going to be how high can we drive the titers in these vaccine studies, and how long-lived they are. I feel pretty confident and optimistic in the vaccines, especially because we're not having to fight against much genetic diversity of the virus.

Whyte: Is the virus mutating? That's not necessarily a bad thing, correct?

Greninger: It's always mutating. It throws down two mutations a month, but the key here is the month. COVID-19 has only been around for 10 months in people. You look at influenza, RSV, other respiratory viruses, and they've been circulating in people for hundreds of years or in different animal hosts that can infect people. There's an incredible amount of diversity; some of them evolve faster. Other coronaviruses have been circulating for tens to hundreds of years in people, and so there's a lot of diversity. You're having to fight against a lot of different strains.

Whyte: Help break this down because this can be confusing to a lot of people. What if the vaccine is only 50% vs 70% effective?

Greninger: The reason people anchor on the 50% number is because that's what the FDA has set as the minimal threshold ─ the bar that has to be crossed for them to authorize or approve a vaccine ─ and I think that's a reasonable floor.

I think with six or seven shots on goal, we'll attain that. It allows these companies to move forward and invest, and it's also reasonable. If something is not 50% effective, we probably can find another one that's going to reach that threshold.

Let's be honest. We've never really done this for the other coronaviruses. We barely made some early candidate vaccine for SARS and MERS. They weren't high on the list. Now we're putting all of our attention to it, and we're using a lot of new modalities.

The other silver lining to this cloud is that we're going to be able to test these different platforms, and hopefully one or two of them really shine through. We're testing some new hypotheses around prefusion protein, so new ways to make these proteins and new ways to deliver them. New adjuvants are on board. Hopefully, one of these will work well and we'll be able to continue the momentum into the rest of the respiratory viruses.

Whyte: When do you think we're going to see a vaccine that's approved as safe and effective?

Greninger: That's a good question. The guy who knows the most about it is Dr Tony Fauci. Why would you go against that? He's practically designing the trials. I think that's a reasonable answer. I do want to say that there have been recent vaccine initiatives for emerging viruses compared with other viruses (Ebola or Zika), where they sort of tailed away. We don't talk about Zika virus right now, right? There aren't a ton of cases, and that actually complicated the vaccine trials. By the time you rolled out the vaccine trial, the cases were diminishing. It makes it harder to show that you can prevent cases.

If we still have 30,000, 40,000, or 50,000 cases a day in the United States, Brazil, and South Africa (adjusted for population), that means that there are a lot of cases to prevent. It makes the trials a little easier to open, enroll people, and achieve significance that first time you look at the data. It could be a lot quicker than the average vaccine trial.

Whyte: In the meantime, we'll continue to learn about the role of antibodies and whether that provides protection.

One other quick question: The current labs that measure antibodies right now, they're not measuring neutralizing antibodies?

Greninger: Nope.

Whyte: And that's what you looked at as the real measure to know about the effectiveness of antibodies.

Greninger: For full disclosure here, we have a clinical lab and also Jesse Bloom's lab at the Fred Hutchinson Cancer Research Center. Clinical labs don't often run neutralizing antibody tests. Most places don't, actually.

Whyte: Is that what we really need, though?

Greninger: If you look across the whole antibody and clinical lab spectrum, you have two different antigens. You have the nucleocapsid, which wraps the genome, and you have the spike protein, which is what binds the cells.

Most of the tests that are done in the United States are done against the nucleocapsid, because it's the most sensitive assay and it's better at telling you whether you were infected.

What we want are antibody tests that show that spike, the outside glycoprotein from the virus that's involved in attachment and entry. If you can get antibodies against that, you can say more about what can prevent infection.

Whyte: But that's not what current labs typically are doing, right?

Greninger: Yeah, there's a mix of them. The market is probably 75%-80% nucleocapsid, 20% spike.

Whyte: But I really want the spike. Is that what I should be looking for?

Greninger: Yes. I think that for the receptor-binding domain, you want to look at the outside of the spike, the part that binds the receptor. That's what you'd like. That, I think, is what you're most likely going to see in a clinical lab that could be achievable and will be the best potential correlative protection.

Whyte: You've given us a whole virology lesson today. I want to thank you for taking the time today in helping us understand what is the role of antibodies in terms of preventing infection, what's the risk for reinfection, are where we are on the vaccines. We've covered a lot, and I appreciate you taking your time.

Greninger: Thanks for taking the time to talk to me. I appreciate it. Thanks for doing all of this. The media is so important when it comes to this pandemic.

Whyte: And I want to thank you for watching Coronavirus in Context.

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.