What J&J Can Still Teach Us

A vaccine’s value isn’t just in its peak performance.

a vaccine syringe whose body is made up of chain links
Getty / The Atlantic

The Johnson & Johnson vaccine, perhaps more than any other COVID shot, knows what it is to be bullied by the American public. Since the spring, the shot’s been roasted, and roasted, and roasted again—first for its late arrival and its imperfect performance in trials, then for a rare but concerning side effect that temporarily halted its distribution in April. Tweets, memes, and listicles dragged it. SNL skewered it. CVS pharmacies stopped offering it. Then, in October, federal officials urged everyone on Team J&J to get another shot—any shot (but also, maybe try Moderna this time?)—rendering the vaccine’s one-and-done protection, its clearest advantage over its mRNA competitors, just about moot. The underdog dose, the “second class” shot, the nation’s vaccine-a non grata, seemed as good as dead.

This incessant ragging has been all too easy—and maybe shortsighted. According to some experts, the haters are overlooking a trait that could rescue J&J’s reputation, and possibly even keep it in scientific contention. “I think there is a silver lining to this vaccine that a lot of people don’t see,” David Martinez, an immunologist at the University of North Carolina at Chapel Hill who is studying immune responses to COVID-19 shots, told me. It’s a trait called durability—the ability of a vaccine’s protection to persist, despite the ravages of time. Several researchers, including representatives of the company that designed the J&J vaccine, say they’re seeing early hints of this with the shot. “It’s unequivocal,” Mathai Mammen, the global head of research and development for Janssen, the vaccine-manufacturing pharmaceutical company owned by Johnson & Johnson, told me. In tracking the vaccine’s effectiveness, “there is no change, month over month over month.” The shot’s initial magnitude of protection against sickness might not match Moderna’s or Pfizer’s. But after they’re built, J&J’s defenses seem to stick around in a way that their mRNA-driven counterparts might not, like a low-wattage bulb that keeps burning, long after all the other lights in the room have flickered and died.

Not everyone is ready to laud J&J’s staying power; we are, after all, still very early on in our relationship with these vaccines, and our understanding of their traits will keep evolving. But even the potential for tenacity, in a vaccine, has real appeal. A durable shot is low-maintenance, requiring only rare checkups or boosters; it can be delivered once or twice or thrice and, in the best-case scenario, never, ever again. As the pandemic heads into its third year, durability underpins some of the biggest open questions in COVID immunology—the long-term outlook for our current shots, the number we’ll ultimately need, and the possibility of engineering an even sturdier vaccine. A lack of durability might mean we’ll be getting COVID shots often, maybe even annually. Or, if we can figure out a clever way to give out shots now, we may not have to administer them again. A vaccine’s value isn’t just in its peak performance; also essential to know is when, and how quickly, protection might start to decline.

But the quest for durability has long been thorny. Several experts I spoke with described it as one of the most elusive concepts in vaccinology, an immunological white whale that researchers frequently chase but almost never catch. “We don’t have one right answer” for what makes a vaccine’s protection stick, Padmini Pillai, an immunologist at MIT, told me. “It’s always it depends.” As long as the virus continues to sprout new variants, and we as hosts continue to throw ourselves in its path, lasting protection may not really be in the cards. A vaccine that guards us doggedly, though, could cushion us against the pathogen’s constant pummeling. In a world without guarantees, we need a vaccine that doesn’t just punch back, but punches back reliably, again and again and again.


Establishing durability starts with first impressions. To offer truly long-lasting protection, a vaccine has to persuade the body to study its offering, then stably store that intel away. “The bottom line is, you have to convince the immune system that this is scary,” Mark Slifka, an immunologist and vaccine expert at Oregon Health & Science University, told me. When the process works well, it can work really well. Every time a microbe returns to trouble us, the defenses we mount against it get stronger, faster, more precise; the response becomes a reflex, built on the memories of immune cells that have thwarted the same threat before.

The major players in immune memory fall into two main camps, headlined by B cells and T cells. B cells are weapons manufacturers, tasked with pumping out microbe-trouncing antibodies; T cells are single-combat assassins that home in on infected cells and force them to self-destruct. Both Bs and Ts will show up to fight most infections of note, cloning themselves into complementary armies. As the danger passes, their numbers contract, leaving behind only a so-called memory contingent—dormant Bs and Ts that holster the capacity for protection, like sleeper agents waiting to hear a trigger phrase. And finding relatively sturdy levels of these cells and the molecules they make is a decent proxy for judging immunity’s longevity. High levels of antibodies and B cells that recognize the viruses responsible for smallpox and measles, for instance, have been found in people decades after they received those two very potent vaccines.

When investing resources, though, our immune systems must be stingy. Not every potential threat they encounter gets locked in the body’s defensive memory. Generally speaking, they’ll devote more storage space to bugs they deem dangerous repeat offenders. Many durable vaccines, then, are really annoying ones, pestering cells so much that they have little choice but to remember what’s up.

One decent approach to making a vexing vaccine involves a weakened version of the bona fide pathogen. The measles vaccine, for example, contains a neutered virus—one that won’t cause true measles, but is otherwise a dead ringer that bops through the body, infiltrating cells and copying itself in much the same way its scarier cousin would. This approach isn’t foolproof: Manufacturing vaccines like these can be slow and difficult, and the payoff doesn’t always come through. Shot-induced defenses against mumps, for instance, seem to slowly ebb over time. And toying around with vaccines that still replicate can be dangerous. The tamed virus in the oral polio vaccine can, in very rare cases, acquire mutations that allow it to cause full-blown disease.

To ratchet down risk, researchers will sometimes opt instead for killed microbes—completely incapable of causing harm, but still very much recognizable as the real thing. This strategy is akin to giving the immune system target practice with a corpse, and can be very, well, hit or miss. Our most used flu vaccine, for instance, is of this ilk, but offers only modest protection that appears to atrophy just months after people get their shot. Still other tactics simplify things further, and use only select pieces (often proteins) of a pathogen’s anatomy. The idea here is to teach immune cells about the bug’s most salient or dangerous features, in hopes of coaxing out a hyper-precise, hyper-potent attack. These vaccines are especially safe, and easy to mass-produce. But there’s always the risk that they fixate on the wrong microbial feature, especially if it’s one easily modified through mutation. Many of these limited-focus shots have also, over the years, delivered lackluster results because they poorly titillate T cells. By themselves, “proteins just aren’t irritating cells that much,” and sometimes, vaccine makers have to add other ingredients just to rouse immune cells into reacting, Sallie Permar, a pediatrician and vaccinologist at Cornell, told me. Protection offered by the protein-based pertussis vaccine, for instance, is infamously brittle.

Common vaccine lore holds that some in this last class of vaccines might be too unlike the bugs they’re modeled on; the immune system has a hard time appreciating what they represent. After all, loose jumbles of free-floating protein are, architecturally, nothing like the intricately patterned pathogens they’re meant to teach immune cells about. For someone who’s never seen a flower, a pile of petals is not enough to form a picture; first, the petals have to be arranged.

In this way, the very unusual shape of the HPV vaccine makes it an outlier among its colleagues. This shot contains viruslike particles—hollow shells, ornamented with a tight array of HPV proteins. The particles aren’t themselves infectious, but they seem to do a heck of a job eliciting a persistent antibody response. After just two doses of the vaccine, levels lift, then dip down to a startlingly stable plateau, thanks to a small armada of B cells in the bone marrow that continues to churn out antibodies. The HPV shot “has become the poster child of durability,” John Schiller, whose discoveries helped develop the vaccine, told me. Its recipients seem to retain a shield against the virus that is, as far as researchers can tell, as close to impermeable as one can get. If Schiller were to take a stab at designing a SARS-CoV-2 vaccine, he said, “I would do virus-like particles.”


The future of COVID-19 vaccines could include some of these flashier, virus-mimicking models. For now, though, our best shots are the ones we already have. The Moderna, Pfizer, and Johnson & Johnson varieties seem to sit somewhere in the middle of the spectrum of immunological irk. Like protein vaccines, they serve up only chunks of the virus, but part of their innovation is that they don’t offer those tidbits directly. Instead, the vaccines instruct our cells to manufacture SARS-CoV-2’s spike, a protein that normally decorates the virus’s surface, and parade those spikes in front of immune cells, partially simulating an infection. In J&J’s case, the comparison to infection is particularly close. The vaccine contains a different virus, called an adenovirus, that pushes into cells and delivers its protective payload. It’s modified to be benign, but it’s still, to the immune system, a virus. That might be why some studies have found that the J&J shot is especially good at tickling certain types of T cells, which prefer to take their lessons from vaccines that will pantomime infected cells.

That’s not to say the mRNA vaccines are T-cell slouches. The post-shot T-cell counts for all three vaccines are respectable, and their levels all look quite stable, several months out. There’s also now strong evidence that great numbers of B cells will post up in the blood and the bone marrow after COVID vaccination, some of them retaining the ability to make antibodies long-term, even sharpening their sniperlike skills against SARS-CoV-2. Those forces are a big part of why “the level of protection against severe disease is still very good” with all three brands, Kizzmekia Corbett, an immunologist at Harvard who helped develop Moderna’s vaccine, told me. And sustained protection against severe disease certainly counts as a type of durability.

When defenses drop, though, they tend to do so stepwise: The strongholds against infection fall first, then transmission, then serious disease, and finally death. Pfizer’s effectiveness against milder COVID cases, and probably transmission, gradually but notably ebbs in the months after people are inoculated. Some of that dip is probably attributable to fast-spreading, slightly immune-evasive Delta, and the world’s growing ennui with distancing and masks; if a new variant like Omicron rises, we could be due for yet another trough in protection. But decreasing effectiveness could also reflect our bodies’ reaction to the shots. Antibodies seem to be tied tightly to protection thresholds, and “we’re quickly seeing antibody levels fall” in the months after people get their Pfizer and Moderna shots, Ai-ris Yonekura Collier, a physician and immunologist at Beth Israel Deaconess Medical Center, in Boston, who’s been studying immune responses to COVID-19 shots, told me.

That in and of itself isn’t catastrophic—antibodies always contract after the first flush of infection or vaccination—but the slope is steeper than some researchers would like. In a small, recent study, Collier and her colleagues showed that about eight months post-vaccination, virus-blocking antibodies are down roughly 40-fold from their peak, and it’s not clear when or where the downslope will flatten into a plateau. Perhaps the molecules have already settled at a stable level, with safeguards against severe disease strong, and defenses against milder outcomes middling. Or maybe they’ve still got a ways to fall. “It’s normal to see rapid decay,” Slifka told me. “The question is, how high above the protective threshold do you land?”

It’s still early, but Collier thinks the dynamics might look a bit different for people who got a single dose of J&J. Her recent work shows that their antibody levels start significantly lower than mRNA recipients’, but that eight months out from vaccination, the numbers have stayed stable, and have, perhaps, even gone up, shrinking some of the gap between brands. “I liken them to a fine wine,” Collier told me. “They get better over time.” Still, antibodies aren’t everything. J&J’s success might come down to how those antibody levels behave further out, and how the rest of the immune system behaves in concert.

With only months of data to back the J&J shots, “the jury’s still out” on how long their strongest effects will last us, Slifka said. For now, though, he isn’t betting that any of our current COVID vaccines will be immortalized in the durability hall of fame.


A strong shot design might jump-start durability, but even less-than-triggering recipes aren’t doomed to fail durability tests. The how, when, how often, and how much of administration can also cement protection. Those gentle hacks are what we’re all obsessing over now, with COVID-19: how many shots we’ll need, and how far apart.

True staying power almost always requires more than one dose; most shots need to usher in multiple immunological reminders to really get the message to stick. The HPV vaccine and the measles/mumps/rubella (MMR) shot are two-dosers; the vaccines that block hepatitis B use three shots; the diphtheria/tetanus/acellular pertussis (DTaP) vaccine uses five, and that’s all before boosts. “It can make a huge difference,” Slifka said. “The immune system thinks, I must not have dedicated enough of an immune response to it the first time.” Each additional injection tends to have a nice effect on the strength of the response, pushing B and T cells to be far feistier than they were before. No clear-cut logic dictates how many refreshers a vaccine regimen will need. Some shots have added doses over the years, while others have stripped them away. Researchers around the world are still debating how often we need boosters for the shots we use to block mumps (more often?), tetanus (less often?), and yellow fever (depends whom you ask). “And every time we start to look at different pathogens, the rules can change a little bit,” UNC’s David Martinez told me.

With COVID, the math remains extra muddy, but might fit the pattern of more is more, as my colleague Rachel Gutman has written. Moderna and Pfizer are both double-dose vaccines, a one-two punch that seems to land better than J&J’s single jab. The second injections, in particular, send antibody levels soaring. The higher that early-post-vax peak crests, the more those antibodies can accomplish: They’ll be able to rally against variants they weren’t initially roused to spot. And their level will have more room and time to fall before it dips past the point of protecting against infection, disease, or death, “even if it’s decaying at the same rate,” Diane Griffin, an immunologist and measles-vaccine expert at Johns Hopkins, told me. Post-Moderna antibodies in particular surged up so high that, despite some declines, effectiveness still looks substantial, many months out; immunity-dodging variants, too, keep falling prey. The Moderna moxie might be due to how much viruslike stuff is in each dose as well. Each shot pumps in more than three times the mRNA that Pfizer’s does, perhaps scandalizing B cells into squeezing more antibodies out.

The antibodies lured out by the single J&J jab are, in comparison, not much to brag about. Adding a second J&J dose two months after the first, though, offers a substantial bump, and a concomitant skyrocketing in effectiveness: 94 percent against moderate to severe COVID-19, compared with just 74 percent from the solo shot, at least among Americans. That puts J&J roughly on par with the mRNA vaccines in their early days. If those extra-elevated antibody levels hold, J&J could end up being a vaccination dark horse.

Timing, too, can help a vaccine’s protection cling. Moderna’s vaccine, whose two doses are delivered four weeks apart—a week longer than the Pfizer interval—appears to be the more obstinate of the mRNA duo. Other multidose vaccines are doled out over months or years, which might make for a better learning experience, allowing immune cells to mull a vaccine’s contents, rather than frantically skim through them. Early data on COVID-19 vaccines back this up: Waiting to give the second dose seems to drive an even heftier antibody response for all three vaccines.

And then there’s the question of location—where vaccines are administered, and where their ingredients end up. A shot in the arm can be a big mismatch for a microbe that enters through the gut or airway, where specialized immune cells might better respond to a vaccine that’s swallowed or sprayed into the nose. Vaccines that vanish from the body too quickly can also be forgettable. “The mRNA vaccines are a quick show” and might not give the immune system much time to wise up, MIT’s Padmini Pillai told me. The HPV vaccine, by contrast, might owe some of its success to the prolonged study session it delivers to cells, according to Schiller. Some experts suspect that the J&J shot might also smolder slowly, giving cells more time to sharpen their skills.

There’s likely still room to keep pushing our current COVID vaccines’ protective potential. Doses could be spaced further apart, brands mixed and matched. Perhaps the key is to boost at the appropriate time, to shore up immune responses that have started to crack and crumble. Pfizer’s boost, for one, seems to rejuvenate defenses against infections of all severities—especially in older adults—probably in part by reawakening legions of antibody-producing B cells that the body has stowed away. “Boosters are kind of a way to patch durability,” Martinez told me. What we’re now calling “boosts” could someday even become part of the primary-shot series: Pfizer and Moderna would be three-dosers; J&J, a double whammy or more.

When testing the Moderna vaccine on nonhuman primates, Harvard’s Kizzmekia Corbett has seen something encouraging. When she and her colleagues boost rhesus macaques six months after their first two doses, antibody levels rocket up, then mosey back down—but the peak is higher, and the slope of decline in the weeks following “is not as steep as previously.” That could be great news for us humans: Antibodies may take much longer, after a third shot, to reach their plateau. Better yet, that set point could be higher than before, a sign that the body’s been goaded into further investing in its SARS-CoV-2 defense. In a best-case scenario, that additional shot could be the last one we ever need. But we won’t know either way for a good while yet.

Durability isn’t airtight. The length of time our vaccines protect us also hinges on the microbes they guard against, which could shapeshift out of our shots’ grasp. Human behavior, too, dictates the dynamics of protection. Pathogens spread through us and because of us; the more blasé we are about exposure, the more often our defenses get battered. When a virus runs rampant, “we can’t just ask our vaccines to pick up all the slack,” Corbett told me.

But when deployed in the right context, vaccines that are durable—really and truly durable—can completely change our relationship with a pathogen. Viruses and bacteria, starved of proper hosts, don’t circulate as much. The post-vaccination cases that do occur become, on average, less severe, more ephemeral, and less likely to spread, untethering infection from serious disease. We stop checking if vaccine protection is still in place, because we don’t have that much use for it anymore; the shields can fall because the attacks have stopped. For SARS-CoV-2, a pathogen that has so deeply intertwined itself with us, that future is a long way off. But it’s a realization of the dream of a near-perfect vaccine: one so excellent, so steadfast, that it prompts our bodies to remember, long enough that even our minds can eventually forget.