In April of 2022, almost two and a half years into the pandemic, the Centers for Disease Control and Prevention estimated that 40 percent of the US population had yet to contract SARS-CoV-2, the virus that causes COVID-19. Thanks to the highly transmissible BA.5 variant, that percentage is probably smaller now, but there are still some people who don’t seem to have ever had an infection—or at least, they’ve never had symptoms or tested positive. SARS-CoV-2 infections can be totally asymptomatic, and the variant in question as well as other factors can influence testing accuracy, raising the possibility that a person could have had COVID-19 and not been aware of it. In addition to leaving many wondering whether they’ve actually caught the virus, this kind of uncertainty also causes headaches for epidemiologists, especially those who want to track community spread. 

Generally, testing for prior infections involves looking for antibodies that bind to viral proteins, particularly those that target SARS-CoV-2’s nucleocapsid (N) protein (anti-N antibodies), as the available vaccines teach the immune system to spot the virus using its spike (S) protein instead (anti-S antibodies). However, a study published last month in Annals of Internal Medicine finds that antibody testing likely undercounts the number of vaccinated people who have had a previous COVID-19 infection, potentially by as much as 40 percent. 

So, is there any reliable way to test for a previous SARS-CoV-2 infection? The Scientist spoke to Lindsey Baden, an infectious disease specialist at Brigham and Women’s Hospital in Massachusetts, to find out. 

The Scientist: Why is it important to know if people have had a previous SARS-CoV-2 infection?

Lindsey Baden: Given that a fair amount of COVID is asymptomatic or minimally symptomatic, there are a lot of infections that people don’t realize they’ve had. So it becomes important to understand how much transmission there is in the community to help with epidemiologic studies and other types of studies that better define the burden of infection and transmission.

TS: What are the methods scientists and researchers use to determine if someone has had a previous SARS-CoV-2 infection?

LB: To measure the footprints of the virus after infection, we use molecular markers or antibodies, such as the nucleocapsid antibody. 

What’s important to note is that the widely used vaccines in the US and Europe . . . are largely based on the spike protein. In this case, the immune response to the spike protein could either be from natural infection or from vaccination. An immune response to the nucleoprotein can only be from exposure to the virus. That isn’t true everywhere in the world, because there are parts of the world that use inactivated COVID vaccines, which have the whole virus. There, some of the nucleoprotein may be present and therefore stimulate that immune response.

TS: How accurate are serological tests in determining if someone has had a SARS-CoV-2 infection?

LB: That’s a terrific question that’s more complicated than it seems, because one of the things that we have to do [when we] deal with a new virus is understand the immune response to that virus. We have to understand if it behaves like . . . other pathogens. So we are in the process of learning how well these serological tests behave, given that we’ve only been dealing with SARS-CoV-2 for two and a half years. There’s no way for us to know if the immune responses last for three years because we haven’t dealt with the virus that long. One of the other issues is understanding the biology of this immune response. Are there factors that can alter it, such as prior vaccination and then infection? Does that change the way in which the immune response [is] elicited? This is an area of active research.

TS: How long can we typically detect anti-N or anti-S antibodies?

LB: Typically, there’s a decay [in antibody responses] over a year or two—or that’s what we think from natural infection because that’s what would induce this type of antibody response.

TS: Is there any way that scientists can definitively determine if someone has had a previous SARS-CoV-2 infection?

LB: I think the nucleocapsid antibody response is very reliable. But what we’re learning is that in people that have been vaccinated, an immune response may not be elicited. There is a variety of other immune responses [s] that can be looked at, like antibody responses to other viral proteins and T cell responses. 

T cells . . . react to specific viral epitopes, often short protein sequences. And so one can look for T cell responses against other parts of the viral genome that are not part of the vaccine, basically looking at if T cells can detect those proteins. These methods could be useful. One would only have an immune response to those proteins if you’ve seen them before. One has to be careful about cross-reaction to other coronaviruses and you have to choose sequences that are unique to SARS-CoV-2, but that’s easily done in the laboratory. But these methods aren’t validated yet and we don’t know how accurate they are.

[This avenue of research] helps us define with detail . . . who’s infected or not, but it’s not scalable [to] a population level. That’s why we use antibody responses—[they’re] very fast, easily deployed laboratory tests that can measure nucleocapsid antibod[ies] and therefore screen thousands and thousands of samples in a lab in a day.

TS: Is there any method that would be the most accurate?

LB: I’m not sure we have that information. What came from the Annals study was showing that by vaccinating, we alter some of the immune kinetics. Part of what we have to realize with COVID is we’re learning as we go. As the virus changes and as our response to the virus changes, that may alter the biology, and we have to study it to understand it. It may well be that if you have such a mild infection, you don’t even stimulate an immune response. At some level, that may not be a bad thing, as long as it’s not being spread to vulnerable people. So we have to think carefully about what we mean by protection from COVID. The immune system can abort infections at a very early stage if it’s primed. This has gone on with many other viruses over decades, or millennia, actually, in terms of other viruses that we’ve all learned to live with.

Editor’s note: This interview has been edited for brevity.