Post-vaccination Infections Come in 2 Different Flavors

Lumping all breakthroughs together, regardless of symptoms, miscasts what our COVID-19 vaccines can do.

The word "breakthrough" haphazardly strewn about
The Atlantic

Updated at 1:21 p.m. ET on July 13, 2021.

The first thing to know about the COVID-19 vaccines is that they’re doing exactly what they were designed and authorized to do. Since the shots first started their rollout late last year, rates of COVID-19 disease have taken an unprecedented plunge among the immunized. We are, as a nation, awash in a glut of spectacularly effective vaccines that can, across populations, geographies, and even SARS-CoV-2 variants, stamp out the most serious symptoms of disease.

The second thing to know about the COVID-19 vaccines is that they’re flame retardants, not impenetrable firewalls, when it comes to the coronavirus. Some vaccinated people are still getting infected, and a small subset of these individuals is still getting sick—and this is completely expected.

We’re really, really bad at communicating that second point, which is all about breakthroughs, a concept that has, not entirely accurately, become synonymous with vaccine failure. It’s a problem that goes far beyond semantics: Bungling the messaging around our shots’ astounding success has made it hard to convey the truly minimal risk that the vaccinated face, and the enormous gamble taken by those who eschew the jabs.

The main problem is this. As the CDC defines it, the word breakthrough can refer to any presumed infection by SARS-CoV-2 (that is, any positive coronavirus test) if it’s detected more than two weeks after someone receives the final dose of a COVID-19 vaccine. But infections can come with or without symptoms, making the term imprecise. That means breakthroughs writ large aren’t the most relevant metric to use when we’re evaluating vaccines meant primarily to curb symptoms, serious illness, hospitalizations, and death. “Breakthrough disease is what the average person needs to be paying attention to,” Céline Gounder, an infectious-disease physician at Bellevue Hospital Center in New York, told me. Silent, asymptomatic breakthroughs—those that are effectively invisible in the absence of a virus-hunting diagnostic—are simply not in the same league.

To put this in perspective, consider the original criteria laid out by the FDA about this time last year, back when the United States was still solidly in its second infectious surge. An effective inoculation, the agency said, should be able to “prevent disease or decrease its severity in at least 50 percent of people who are vaccinated.” It’s an easy benchmark to forget. By the close of 2020, two vaccines absolutely obliterated those expectations; two months later, a third followed, and now there’s buzz of a fourth.

If disease is our yardstick, then breakthrough COVID-19 cases—a very small subset of all known breakthroughs—might meet our criteria for concern. These are actual illnesses, events where the shots’ protection has apparently crumbled; these cases are the same ones that vaccine makers searched so diligently for in clinical trials, to ensure that their products were working. By the same logic, asymptomatic coronavirus infections fall outside our shots’ protective purview as we defined it so many months ago. And although they’re important to track and glean data from, conflating them with the rest, experts told me, risks misrepresenting what our vaccines can do. (The CDC responded to an inquiry about its designation by saying that while a “SARS-CoV-2 infection” indicates any positive tests for the virus and a “COVID case” refers to a person with a positive test who meets other case definitions, “throughout COVID the terms infection and case have often been used interchangeably.”)

The term breakthrough has long been a staple of the infectious-disease community, where it’s used to describe the detection of vaccine-preventable pathogens in immunized individuals. “This is definitely not a new idea,” says Kevin Escandón, a physician and infectious-disease researcher at the University of Valle, in Colombia. But as a popular notion, it was always doomed to cause some confusion. Breakthrough is still used as an adjective of praise; the pandemic has now warped the word into a foreboding noun that tends to eclipse all clarifying qualifiers. “It’s confusing, it’s fuzzy, it’s already loaded,” Alison Buttenheim, who studies human behavior around vaccines at the University of Pennsylvania, told me. And when news appears in a headline or push alert, or on social media, “people pay attention to the word breakthrough” and not much else, Ryan McNamara, a virologist at the University of North Carolina at Chapel Hill, told me. That’s unfortunate, when the simple addition of asymptomatic or symptomatic can make all the difference. As they stand, blanket breakthroughs sound far scarier than they should.

Joseph Allen, a public-health researcher at Harvard, recently pointed out on Twitter one such ambiguity, in a study documenting a very small number of breakthrough infections at a prison. All were asymptomatic—though you wouldn’t know it from the paper’s title.

To be clear, breakthroughs of any severity are an entirely expected part of the vaccination process. No vaccines are 100 percent effective at preventing infection or disease. But our current crop of COVID-19 shots comes pretty damn close with regards to stymieing symptoms, especially the severe ones that can signal a deadly case. The Moderna and Pfizer shots have consistently demonstrated very high COVID-prevention rates, often in the 90s; Johnson & Johnson’s, for the most part, isn’t far behind. Symptomatic breakthroughs are the cases that wedge themselves in the gap between excellent effectiveness and perfect effectiveness; in other words, we saw them coming.

Even out in the messiness of the real world, symptomatic breakthrough cases are proving themselves quite rare. The overwhelming majority of the COVID-19 cases we’re seeing are among the unvaccinated. And when the virus does affect the immunized, it seems to accumulate to lower levels, and spread less enthusiastically to new hosts; it’s causing, on average, milder and more transient symptoms.

All of this is a reminder of how vaccines work—by ratcheting up our immunity against the version of SARS-CoV-2 that the shots were formulated to mimic. If humans are wood that fuels a flame, and coronaviruses are the sparks that ignite it, vaccines are the fire suppressants that protect best against the worst of the viral burn: severe disease, hospitalization, and death. Stopping milder cases requires more immune investment, and blocking asymptomatic infections—ones that barely singe the bark—is most difficult of all. It’s part of why the vaccines’ goalposts were at first set so conservatively. “This is not a magic shield that just bounces coronavirus right off you,” McNamara told me.

Considering that we first took aim at stopping disease, it’s great news that the majority of known breakthroughs have actually been asymptomatic infections, not COVID-19 cases. The proportions of silent breakthroughs reported by various studies and federal agencies are certainly undercounts, because vaccinated people aren’t regularly screened for the coronavirus. (On May 1, the CDC controversially switched its reporting strategy to documenting only breakthrough cases involving some form of hospitalization or death, skewing national counts further.) Since the vaccines first deployed, the news has only improved: Researchers didn’t bank on it, but in many people, the shots seem to stop the coronavirus from establishing itself at all. “The vaccines are better than anything we ever dreamed of,” Gounder told me, exceeding our first expectations in more ways than one.

The shots are even holding their own against SARS-CoV-2 variants. A few versions of the virus have picked up mutations that help them dodge certain anti-coronavirus antibodies. But these genetic alterations chip away only incrementally at immune protection, rather than obliterating it. Against Delta, for instance, vaccines like Pfizer’s are still curbing severe disease, hospitalization, and death to an extraordinary degree. And while the shot’s strength has slightly slackened when it comes to milder illnesses and silent infections, those are simply lower hurdles for a virus to clear. Pfizer’s protection is still hitting its mark where it matters the most. (One asterisk on this is long COVID, a condition whose relationship to vaccination is still being actively researched.)

None of this means, of course, that asymptomatic breakthrough infections should be ignored. To fully understand what the virus is doing and where it might be headed, experts need as comprehensive a picture as they can get of whom it’s afflicting, and what form those infections take, across the entire spectrum of disease. They also need to know how and when it’s most likely to spread. Asymptomatic infections are a part of that. Researchers around the world are still diligently sequencing any and all test-positive coronavirus samples they can, regardless of symptoms, in part to check whether any particular variants are disproportionately infiltrating the inoculated. They’re also tabulating who’s experiencing breakthroughs, and testing whether select populations might benefit from an early vaccine boost.

And when vaccines start to consistently falter against more severe tiers of disease—because of either a new variant, waning immune memory of the virus, or both—the diligent monitoring of breakthroughs will pick it up. Tracking milder breakthroughs is also crucial to figuring out how well the virus can be transmitted from vaccinated people, something that’s much more difficult to determine than whether inoculations merely block disease. From a surveillance standpoint, casting a broad net for breakthroughs—one that accounts for infections of all types—is essential, Buttenheim said. “That’s how you catch everything.”

The question of which breakthroughs matter ultimately depends on another: What’s the goal of vaccination? Gounder thinks that, for now, the focus should stay on using immunizations to control COVID-19, especially while so much of the world remains unvaccinated; understanding whether we’re accomplishing that goal, then, hinges on symptomatic breakthroughs. Eventually, we’ll have the bandwidth to turn our attention to halting transmission and infection more comprehensively. Then, we’ll pull asymptomatic breakthroughs back into the conversation, with more data to guide our next move.

The Atlantic’s COVID-19 coverage is supported by grants from the Chan Zuckerberg Initiative and the Robert Wood Johnson Foundation.

Katherine J. Wu is a staff writer at The Atlantic.