By Sarah Zhang

One of the most perplexing and enduring mysteries of the pandemic is also one of the most fundamental questions about viruses. How can the same virus that kills so many go entirely unnoticed in others?

The mystery is hardly unique to COVID-19. SARS, MERS, influenza, Ebola, dengue, yellow fever, chikungunya, West Nile, Lassa, Japanese encephalitis, Epstein-Barr, and polio can all be deadly in one person but asymptomatic in the next.

But for most of human existence, we didn’t know that viruses could infect us asymptomatically. We didn’t know how to look for them, or even that we should. The tools of modern science have slowly made the invisible visible: Antibody surveys that detect past infection, tests that find viral DNA or RNA even in asymptomatic people, and mathematical models all show that viruses are up to much more than making us sick. Scientists now think that for viruses, a wide range of disease severity is the norm rather than the exception.

A virus, after all, does not necessarily wish its host ill. A dead host is a dead end. The viruses best adapted to humans have co-evolved over millions of years to infect but rarely sicken us. Human cytomegalovirus is a prime example, a virus so innocuous that it lives in obscurity despite infecting most of the world’s population. (Odds are that you have it.) Infections with human cytomegalovirus are almost always asymptomatic because it has evolved a suite of tricks to evade the human immune system, which nevertheless tries its best to hunt the virus down. By the time humans reach old age, up to a quarter of our killer T cells are devoted to fighting human cytomegalovirus. Pathogens and immune systems are in constant battle, with one just barely keeping the other in check. In the rare instances when human cytomegalovirus turns deadly—usually in an immunocompromised patient—it’s because this equilibrium did not hold.

The coronavirus that causes COVID-19 is much newer to humans, and severe cases have justifiably gotten the most attention during the pandemic. Scientists have made dramatic advances in understanding this virus and how to treat it. But unraveling why it makes some of us sick, and leaves others unscathed, requires an appreciation of the delicate dance between pathogen and immune system that begins each time the virus finds a new host.

Let’s begin where a COVID-19 infection begins, when virus meets cell. The initial infectious dose—the number of virus particles that enter the body—may influence the course of infection. The more particles that land in your nose, for example, the closer the virus is to overwhelming your immune system, leading in some cases to more severe illness.

Within hours of a typical viral infection, the first infected cells begin secreting interferons, a group of molecules that acts as “a fire alarm and sprinkler system in one,” says Angela Rasmussen, a virologist at Georgetown’s Center for Global Health Science and Security. The fire alarm alerts the two main branches of the human immune system: the fast but nonspecific innate immune system, which causes inflammation and fever, and the adaptive immune system, which over a series of days will muster antibodies and T cells that more precisely target the invading virus.

Interferons also “interfere” with the virus in a number of ways, such as degrading viral genes, preventing cells from taking up viral particles, suppressing the manufacturing of viral proteins, and causing infected cells to self-destruct. By slowing replication of the virus, interferons buy time for the rest of the immune system.

This is what happens when everything goes right. But every successful virus has to develop ways of evading the body’s defenses, and the coronavirus that causes COVID-19 is very good at a devilish trick: Several of its genes encode proteins that seem capable of blocking interferons. By quieting the body’s fire alarm and disarming its sprinkler system, the coronavirus can set fire after fire. In the race between virus and immune system, the immune system falls behind. The virus proliferates. Lung cells die.

Eventually, so many viral particles are infecting so many cells that the immune system knows something must be wrong. It begins to gear up—but too late. Without timely targeted strikes from the adaptive immune system’s antibodies and T cells, the powerful but blunt innate immune response ramps up and up, destroying healthy human cells in the process. This is one possible explanation for the immune overreaction observed in severe and fatal cases of COVID-19.

This delayed interferon response, Rasmussen told me, reminds her of Ebola, which she studied before our current pandemic. Ebola is a very different virus with a much higher fatality rate, but deadly cases of Ebola are also characterized by uncontrolled inflammation in the body following a delayed interferon response. And Ebola is asymptomatic in some people too—as many as a quarter of all those infected, according to one estimate. Surveys in outbreak areas have found many people with antibodies against Ebola but no recollection of illness.

Some of the differences among patients’ interferon responses might be genetic. When Dutch doctors investigated severe COVID-19 cases in two pairs of brothers, they found that all four had a genetic mutation that impaired interferon production. The brothers, who were from two different families, were all healthy and young, from 21 to 32 years old. One of them died of COVID-19, and all four needed to be put on ventilators in an intensive-care unit. But their specific mutation is not common, and genetics are unlikely to completely explain the variation in COVID-19 cases. Every scientist I spoke with emphasized how little we know. “It would be an understatement to say we do not fully understand,” Alessandro Sette, an immunologist at the La Jolla Institute, told me.

The asymptomatic end of the severity spectrum is the most difficult to study. The first challenge is finding the cases: Asymptomatic people do not come to the hospital and are unlikely to get tested. If they are tested, their early immune response is typically long over by the time results come back. Finding asymptomatic patients usually means following a large group of healthy people for a long time, waiting for some of them to catch the virus of interest.

In the summer of 2020, Antonio Bertoletti, a virologist at Duke-NUS Medical School in Singapore, had one such opportunity to study asymptomatic COVID-19 patients. Although Singapore had so far largely controlled the spread of COVID-19, an outbreak was raging among its migrant workers, many of whom were from India and Bangladesh. To contain the outbreak, the government paid the workers to isolate at home and track their symptoms with thermometers and oximeters. During the isolation period, Bertoletti and his colleagues recruited 478 workers who were willing to have their immune responses tracked through periodic blood samples. Over a six-week period, about a third of the study participants caught and recovered from COVID-19. A large majority of cases were asymptomatic, and the rest were mostly mild.

Bertoletti and his colleagues were interested in virus-specific T cells that are essential to the adaptive-immune response. When they isolated these cells from blood samples, they found that asymptomatic patients had more specific and coordinated T-cell responses with high levels of an antiviral molecule and another that regulates other T cells. Their adaptive immunity looked more “fit,” Bertoletti told me. The sicker patients’ cells released a broader range of inflammatory molecules, suggesting that their immune response was less targeted.

Although COVID-19 antibodies got a lot of attention early in the pandemic, T cells are now emerging as key to fighting COVID-19. Patients can recover from COVID-19 without antibodies at all—as long as they have T cells to fight the virus. T cells may play an additional role in milder infections: Depending on where in the world you look, some 28 to 50 percent of people have T cells that predate the pandemic but nevertheless react to the new virus. These T cells may be remnants of infections with related coronaviruses, a theory supported by one study, which found that people who were more recently infected by other coronaviruses tended to have milder COVID-19 infections. In Singapore, 93 percent of the cases Bertoletti found were asymptomatic—a much higher percentage than in other closed settings, such as cruise ships—a result he attributes to the migrant workers’ relative youth and possible cross-reactive T cells, which seem to be more common in some parts of the world than in others. To further understand the role of cross-reactive T cells, Sette, at La Jolla, is studying whether patients who possess them also mount a stronger immune response after receiving a COVID-19 vaccine.

T cell responses also weaken with age, which may help explain why COVID-19 is dramatically more deadly for the elderly. Humans have a huge diversity of T cells, some of which are activated each time we encounter a pathogen. But as we age, our supply of unactivated T cells dwindles. Immunosenescence, or the gradual weakening of the immune system over time, is influenced by both age and the system’s previous battles. Human cytomegalovirus—that otherwise innocuous virus that infects much of the world’s population—seems to play a particular role in immunosenescence. So many of our T cells are devoted to suppressing this virus that we may become more vulnerable to new ones.

Unlike human cytomegalovirus, the coronavirus doesn’t seem capable of hiding inside our bodies in the same way for decades. Once it sneaks in, its goal is to replicate as quickly as possible—so that it can find another body before it kills its host, or its host eliminates it.

Now that this coronavirus has found humans, it will have a chance to hone its strategy, probing for more weaknesses in the human immune system. That doesn’t necessarily mean it will become more deadly; the four coronaviruses already circulating among humans cause only common colds, and the virus that causes COVID-19 could one day behave similarly. Variants of the virus are already exhibiting mutations that make them more transmissible and better able to evade existing antibodies. As the virus continues to infect humans over the coming years, decades, and maybe even millenia, it will keep changing—and our immune systems will keep learning new ways to fight back. We’re at the very beginventually, so many viral particles are infectin