What We Actually Know About Waning Immunity
Reports of vaccines’ decline have been greatly overstated.
![microscope view of cells, superimposed on a red brick wall that is faded at the top](https://cdn.theatlantic.com/thumbor/-cCA2nbvZ8QT27kMsOoXIFx0A8I=/53x0:1820x994/960x540/media/img/mt/2021/09/Atlantic_immun_v1/original.jpg)
Vaccines don’t last forever. This is by design: Like many of the microbes they mimic, the contents of the shots stick around only as long as it takes the body to eliminate them, a tenure on the order of days, perhaps a few weeks.
What does have staying power, though, is the immunological impression that vaccines leave behind. Defensive cells study decoy pathogens even as they purge them; the recollections that they form can last for years or decades after an injection. The learned response becomes a reflex, ingrained and automatic, a “robust immune memory” that far outlives the shot itself, Ali Ellebedy, an immunologist at Washington University in St. Louis, told me. That’s what happens with the COVID-19 vaccines, and Ellebedy and others told me they expect the memory to remain with us for a while yet, staving off severe disease and death from the virus at extraordinary rates.
That prediction might sound incompatible with recent reports of the “declining” effectiveness of COVID-19 vaccines, and the “waning” of immunity. According to the White House, we’ll all need boosters very, very soon to fortify our crumbling defenses. The past few weeks of news have made it seem as though we’re doomed to chase SARS-CoV-2 with shot after shot after shot, as if vaccine protections were slipping through our fingers like so much sand.
The reality of the situation is much more complicated than that. Despite some shifting numbers, neither our vaccines nor our immune systems are failing us, or even coming close. Vaccine effectiveness isn’t a monolith, and neither is immunity. Staying safe from a virus depends on host and pathogen alike; a change in either can chip away at the barriers that separate the two without obliterating them, which is exactly what we’re seeing now.
As the hyper-contagious Delta variant continues to blaze across the country and much of the world, more vaccinated people are encountering the virus and occasionally getting infected enough to trip a coronavirus test. But our shots are still guarding against disease and death—the standard our shots were meant to meet, and the most crucial element of making the virus “a much more manageable threat,” Müge Çevik, a medical virologist at the University of St. Andrews, told me. “We need to have much more realistic expectations of these vaccines” and what they can teach our immune systems to do, Çevik said. The good news is, it’s quite a lot.
Immune responses don’t last forever. They’re supposed to wane, and the fact that they do works to our advantage.
The first time someone meets a virus or a vaccine, defensive cells must scramble. A wave of fast but imprecise fighters—members of the innate immune system—rushes in to wall off the assailant, buying time for the body’s more sophisticated sharpshooters to gather their wits. This latter group, which makes up the body’s adaptive arm, takes several days to really fire up. But the wait is worth it: After a couple of weeks, the blood is rife with antibodies—molecules, made by B cells, that can sequester viruses outside cells—and aptly named killer T cells, which can blow up cells that have already been infected.
Eventually, as the infectious threat passes, our immune response contracts; frontline B and T cells, no longer needed in their amped-up state, start to die off. Antibody levels—one of the easiest immune metrics to measure—slip downward over the course of several months, before roughly leveling off. That’s perfectly normal, Deepta Bhattacharya, an immunologist at the University of Arizona, told me. “You have a big increase at the beginning, then a decline.” Consider the alternative: If humans never quieted any of the immunological furor that follows infections and simply kept accumulating antibodies for every pathogen we came across, we’d all have burst a long time ago. Even attempting to maintain that kind of immune reservoir “would require so much energy—I don’t even know where you’d keep all those cells,” says Marion Pepper, an immunologist at the University of Washington.
A downtick in antibody levels can come with consequences. Antibodies are among the few immune actors capable of waylaying a virus before it infiltrates a cell; when present in high-enough amounts, they can quash a developing infection. But where a virus is abundant and speedy and antibodies are relatively scarce, the body’s defenses are much more liable to crack, which is why protection against infection will be the first to erode. This issue might be especially pronounced after receipt of a COVID-19 vaccine, which is delivered into an arm muscle. Injected vaccines are ace at prompting the production of IgG antibodies in the blood; they’re less good at coaxing out the IgA antibodies that patrol the moist mucosal linings of the nose and mouth and corral respiratory viruses at their natural point of entry. IgG antibodies are good travelers and can eventually flock to the site of a growing infection. That takes time, though, and when fewer of them are bopping about, their eventual arrival may not be enough to pen the pathogen in place.
Antibody levels will taper in the months following vaccination or infection, but that doesn’t mean they plummet to zero, Bhattacharya told me. Although most of the B cells die off, some stick around in the bone marrow and keep churning out the virus-fighting molecules at more modest, but still detectable, levels. Though the life span of these long-lived B cells can vary, some studies have hinted that they’re capable of persisting as antibody factories for decades. Another population of immune cells, memory B cells, meanders around the body like sleeper agents, ready to resume making its antibodies whenever necessary. All of these B cells can continue to broaden and intensify their virus-vanquishing powers for months after a vaccine or pathogen leaves the body, in a sped-up form of antibody evolution. “The quality of antibodies in the body improves over time,” Bhattacharya said. “It takes way fewer of them to protect you.”
Populations of memory T cells, too, can hide out for many months or years in tissues, waiting to strike again. Although antibodies are very picky about what they attack, making them easy to stump with viral mutations, T cells are more flexible fighters that are great at recognizing variants. Shane Crotty, a virologist at the La Jolla Institute for Immunology, in San Diego, and his colleagues have documented durable T-cell responses to COVID-19 vaccines. They’re “exceeding expectations,” he told me. “Overall, it looks like there’s high-quality memory at six months.”
Memory responses take a few days to get going. That’s far faster than the response to a first inoculation, when B and T cells are naive to the threat. But if antibodies aren’t already lurking in and around the airway, the virus might get a chance to invade a few cells, maybe even cause some symptoms, before sufficient reinforcements arrive. That’s not necessarily a concern, said Crotty, who described SARS-CoV-2 infection as unfolding in two phases. “Initial replication is fast and tough to stop,” he said. Severe, hospitalization-worthy damage in the lung, however, tends to take at least a couple of weeks to manifest—plenty of time for “even a modest amount of antibodies and T cells” to interfere.
Checking someone’s SARS-CoV-2 antibody levels when there’s no virus around can be a bit deceptive, then. In the absence of a threat, immune cells are quiescent. But the capacity for protection remains intact: When new invaders arrive, they’ll reawaken our defenses. That’s why post-vaccine infections, when they do happen, tend to be milder, shorter, and less likely to spread to other people. When the new threat resolves, levels of antibodies and active immune cells decrease again. “You could call that ‘waning,’” Pepper, of the University of Washington, told me. “But that’s just how it works.”
Immune memories don’t last forever. Eventually, even the grizzled B and T cells in the body’s reserves might permanently retire. That’s when protection against disease and death could start to take a tumble, and when experts start to get worried. Some officials, including CDC Director Rochelle Walensky, have suggested that upticks in post-vaccine coronavirus infections are a sign of what’s to come, and that giving people extra shots could be a way to jog the immune system’s memory before it fades away.
The same rationale applies to many multi-dose vaccines: The first shot introduces the body to the notion of a threat; the ones that follow clinch the concept that the danger is real and worth taking seriously. A triple-jab regimen is already built into several well-established vaccines, including the ones that block HPV and hepatitis B; others require four or five inoculations before they take. But according to most of the experts I spoke with for this story, the immunological argument for a COVID-19 booster this early is shaky at best.
To start with, the recent numbers on vaccine effectiveness aren’t really that alarming. Vaccinated people are indeed getting infected with SARS-CoV-2 more frequently than they were a few months ago. But these breakthroughs remain fairly uncommon. Recent reports from the CDC show that the Moderna and Pfizer-BioNTech vaccines were blocking infection at rates of up to about 90 percent in the spring, when the vaccines had barely begun their rollout en masse; now those stats are hovering around the 60s and 70s, still a remarkable feat. (That doesn’t mean that 30 to 40 percent of vaccinated people are getting infected; rather, immunized people are 60 to 70 percent less likely than unimmunized people to be infected if they’re exposed.) Numbers from other studies look to be in a similar ballpark. And these stats might even undersell the vaccines’ benefits: Many “infections” are found simply through the detection of viral genetic material—with no guarantee that this material is active, infectious, or anything more than the carnage left behind from a victorious immune attack.
The outlook is even better when you consider symptomatic cases of post-vaccine COVID-19. Early reports, including Moderna’s and Pfizer’s original study estimates, put the vaccines’ efficacy against symptomatic illness in the range of 90 to 95 percent. More recent studies now document rates in the 80s, even when facing off against Delta—a variant for which the vaccines weren’t originally formulated.
Some reports from Israel appear to paint a more dire portrait: A few preliminary numbers released by the country’s Ministry of Health suggested that vaccine effectiveness against both infection and symptomatic disease had dipped to about 40 percent. But Çevik, of the University of St. Andrews, told me that these and other data reporting heftier declines are messy and might actually overestimate the problem. Across countries, early vaccine recipients tended to be older, in slightly worse health, and in higher-risk professions than those who got injected later on. That alone could make the protection that they got seem less impressive in comparison. Also, when initial effectiveness numbers were calculated, people were adhering more to physical distancing and masks. Measured these days, amid more lax behavior, risk of infection would rise. And as more of the unvaccinated have been infected, their collective immunity has grown, making them, too, less susceptible to the virus—which could make the effectiveness of vaccines look lower.
Viral evolution makes all of this even more confusing. Delta isn’t a perfect match for the version of SARS-CoV-2 the vaccines were designed to fight. Defensive memory in the body could persist indefinitely—the opposite of waning—and yet still be stumped by a virus that develops a good-enough disguise. Delta is also a very fast version of the coronavirus, capable of cresting in the body and spilling back out within a few days, potentially before a memory response can activate. Studies do suggest that Delta’s a bit more likely than its predecessor Alpha to infect people and cause some symptoms among the vaccinated; that gets easier during a massive surge in cases, when even the immunized are getting clobbered with high doses of the virus on the regular. We have yet to disentangle how much changes in vaccine effectiveness are due to Delta, versus our immune system.
When it comes to severe disease and death, though, vaccine effectiveness hasn’t really budged at all: Immunized people seem to be thwarting the worst cases of COVID-19 just as well as they did when the shots debuted, often at rates well into the 90s. That’s fantastic, considering that the FDA’s original benchmark for vaccine success, announced in June 2020, was reducing the risk of disease or serious disease by 50 percent among people who get the shot. So far, there is simply no “evidence of a substantial decline” against the worst outcomes, Saad Omer, an epidemiologist at Yale, told me. (One important note: We still don’t have enough data to know how well the vaccines prevent long COVID, which can occur even after relatively mild infections.)
All of this underscores the importance of considering effectiveness against infection and disease as “absolutely separate,” Jennifer Gommerman, an immunologist at the University of Toronto, told me. The two can wane in lockstep, but they don’t have to. “Different immunological mechanisms come into play,” Gommerman said. And while a drop in antibody levels might foretell more vulnerability to infection, that doesn’t always translate to a loss of immune memory and more susceptibility to serious sickness. At this point, researchers don’t know how many antibodies people need to ward off infection or disease, and whether levels of these molecules are even the best proxy for vaccine protection.
Pandemics don’t last forever. Eventually, the viral burn ratchets down to a smolder; the disease it causes becomes, on average, more survivable. Vaccines help us tame the flames safely by putting up shields where they didn’t exist before. In a more vaccinated world, fewer trees get scorched; fewer flames hop from branch to branch. The more immunizations go around, the less kindling there is for the virus to burn.
Our collective defenses are sure to wax and wane. In the years and decades that a pathogen stays with us, more vulnerable people will be born, as immunized adults eventually die. Vaccination doesn’t make people impervious; it just gives them more immunity than they had before. When that protection fades, whether through immunological amnesia or because the virus has donned an unrecognizable costume, even inoculated people will slip toward the susceptible state they occupied before.
Waning is not disappearance, though. Even if vaccinated people sometimes do get infected and sick, it will happen less often, and less severely. That, in turn, makes it much harder for the virus to stick around and spread. The goal of vaccination, Çevik told me, is to tame the virus, gradually, into something less formidable, more weatherable. Infection will no longer have to be a crisis. “The point isn’t to protect you from getting even a tiny amount of virus in your body,” she said. We’re not out to eradicate positive test results: “That’s not what vaccines do.”
As for boosters, the pros and cons will vary by context. For people who never responded well to their first vaccines, including people who are moderately or severely immunocompromised, additional shots will be very important, Omer said. Their third jabs don’t provide an extraneous “boost” so much as they help complete the original inoculation schedule.
For the rest of us, though, the perks are harder to visualize. In someone with a fully functional immune system whose defenses were already substantially shored up by their first shots, more doses would probably increase antibody production. That, in turn, could further cut down on infection and transmission, Gommerman told me. Very early data hint that this may be happening in Israel, which is already boosting widely. But it’s not clear how long that preventive bump would last. Ellebedy, of Washington University in St. Louis, said boosters would have “real gain” only if they expanded on the body’s capacity to manufacture antibodies long term, instead of just fueling a temporary boom-and-bust. It’s especially unclear whether that would happen with yet another injection of the original vaccine recipe, delivered to the arm—as opposed to, say, a nasal spray with Delta-specific ingredients.
Right now, some forms of vaccine effectiveness are slipping, but the most important ones aren’t. Unless that changes, widespread boosters in already vaccinated countries are likely to provide diminishing returns, like topping off a drink that’s already on the verge of spilling over. In the meantime, billions around the globe have yet to take a sip at all.
The Atlantic’s COVID-19 coverage is supported by grants from the Chan Zuckerberg Initiative and the Robert Wood Johnson Foundation.