Please Don't Freak Out About the Latest Coronavirus Antibody Study

People voting at the Brooklyn Armory in New York City during early voting on October 28, 2020.
People voting at the Brooklyn Armory in New York City during early voting on October 28, 2020.
Photo: Spencer Platt (Getty Images)

Early data from an ongoing study in the UK is setting off a wave of alarm over the pandemic this week. It found evidence that levels of antibodies to SARS-CoV-2 took a sharp plunge in England’s general population during the summer, prompting fears that immunity to covid-19 may start to significantly wane in as little as three months.


Though the study’s findings are relevant, they’re not as scary as they appear. These kinds of studies alone can’t tell us if immunity fades away so quickly, and, importantly, antibodies are not the only factor that determines immunity to infectious diseases like covid-19. Other research, including a study just published today, suggests that our most important antibodies to covid-19 aren’t in as much peril as the UK study indicates.

Researchers at the Imperial College London, with the help of the government, have been conducting the Real Time Assessment of Community Transmission, or REACT, study in England since the start of the summer. One part of the project focuses on antibody levels to the coronavirus in the community, by asking people to use take-home tests that collect a finger prick blood sample. This week, on the preprint website medRxiv, the researchers released preliminary data from the study, involving more than 350,000 residents.

In late June, they found, about 6% of volunteers had detectable levels of these antibodies. But in September, that number dropped to 4.4%.

“On the balance of evidence, I would say, with what we know for other coronaviruses, it would look as if immunity declines away at the same rate as antibodies decline away, and that this is an indication of waning immunity at the population level,” Wendy Barclay, head of the Department of Infectious Disease at Imperial College London, told reporters at a conference announcing the results this week.

Barclay is absolutely right to point out that our immunity to other coronaviruses does fade away over time. Experts have long said the same will probably happen with the coronavirus that causes covid-19, at some point. But beyond that, immunity to any one specific disease can be a tricky knot to untangle. And though these results may be solid, given the large sample size, the implications for our immunity to covid-19 aren’t so clear.


For one, it’s perfectly normal to see antibody levels to any recent germ drop over time. So seeing this decline over the past three months isn’t, by itself, proof that immunity will go away in that time period. The other consideration is that not all antibodies are the same. The most important are called neutralizing antibodies, because they can directly stop something like the coronavirus from infecting new cells. Even if our levels of antibodies to the coronavirus drop in the short term, most of us may still have enough neutralizing antibodies to stop reinfection for a while longer.

Just this afternoon, a new study seems to show that very thing happening. The study, published in Science, looked at antibody levels among 30,000 New Yorkers diagnosed with mild to moderate covid-19. After three months post-infection, the researchers continued to find a stable level of neutralizing antibodies in over 90% of their volunteers; in a subset of volunteers seen five months after their illness, the levels had only taken a small dip.


“Although this cannot provide conclusive evidence that these antibody responses protect from reinfection, we believe it is very likely that they will decrease the odds ratio of reinfection,” the authors wrote.

As experts have told Gizmodo previously, a person’s current antibody levels aren’t the only relevant factor to figuring out immunity. Our immune system also has memory B cells, which wake up the rest of the immune system when they detect a germ making a reappearance, including the ramping up of antibody production. This usually results in a much faster immune response the second time around. On top of that, there are certain T cells that “remember” a past germ and step into action when it tries to reinfect us. These other components may not prevent reinfection from the coronavirus entirely, but they should usually make the second infection much easier for the body to handle.


All of this is very complicated, and there’s a lot about our immune response to the coronavirus that we’re just now starting to understand (not to mention how these responses will affect a vaccine’s effectiveness). Unfortunately, there have been several reports of confirmed reinfection in multiple countries, including the U.S. In some but not most of these cases, the person got far sicker the second go-around, and at least one person has died from their reinfection. But many experts do not think that these reinfections are happening en masse right now, though the data needed to confirm how often it’s happening is still sorely missing.


The main takeaway here is that we should exercise caution in jumping to conclusions about any scary-sounding antibody studies. It’s still true that aiming for “herd immunity” is a ludicrous and dangerous idea, but moreso because it will lead to lots of preventable deaths and illness, not because we’ll get all reinfected in a few months.


Indeed, as I noted in June, the most present danger from covid-19 remains the first bout of infection. Despite the U.S. now being in the middle of its third large peak of the pandemic, the vast majority of Americans—and indeed people around the world—still haven’t caught covid-19 yet. Hopefully, most of us never will.

Science writer at Gizmodo and pug aficionado elsewhere



Assume nothing with this virus. That is, assume nothing in your/our favor.

Be smart. Stay safe. Wear your mask. Keep your distance. Be patient.