For many people, an infection with respiratory syncytial virus, or RSV, is little more than a troublesome cold. But the virus poses a serious danger to young infants, older adults and immunocompromised people. The disease is the leading cause of hospitalization in infants in the U.S. and was particularly bad in the 2022–2023 season. An estimated 58,000 children and 177,000 older adults in the U.S. are hospitalized with RSV every year. As many as 300 of these children die, along with approximately 14,000 older adults.

After a decades-long search, vaccines for RSV are finally here. Scientists have been working on the shots since soon after the virus was discovered in 1956. But some disastrous clinical trials in the 1960s and dozens of other failed attempts stymied progress for many years. Now not just one but two RSV vaccines for older adults have been approved by the U.S. Food and Drug Administration and the Centers for Disease Control and Prevention. As of this writing, a vaccine given to pregnant people, designed to protect infants after birth, was on track to be approved by the end of the summer. The breakthrough leading to these developments happened once researchers solved a 50-year-old mystery about the virus by examining the shape of its proteins. The discovery has ushered in a new era of vaccine development using designs based on the structure of proteins—the same approach that enabled the rapid development of a COVID vaccine.

Until recently, the main way to prevent RSV infection was the usual hygiene practices used to prevent common colds, such as wearing a face mask, washing one's hands and avoiding sick people. There was also one medication: palivizumab, a short-acting monoclonal antibody that provides passive immunity (protection with antibodies created outside a person's own body) to infants for up to one month at a time. But palivizumab, which was approved in 1998, requires multiple doses that cost more than $1,800 each. The drug is licensed for preterm infants born before 35 weeks who are younger than six months at the beginning of the RSV season, which starts in the fall. The American Academy of Pediatrics recommends restricting the antibody's use to the most vulnerable of these infants because of the cost.

A Grim History

Thirty years before that drug debuted, scientists were already working on an RSV vaccine intended to save lives. To their horror, it took lives away. In 1966 four clinical trials tested a vaccine made with an inactivated form of the virus in children who had never gotten RSV before. In one of the studies, 80 percent of the vaccinated children were hospitalized when they later contracted the virus itself, and two toddlers—a 14-month-old and a 16-month-old—died. Typical hospitalization rates for children with RSV are in the single digits, says Ruth Karron, a pediatrician and director of the Johns Hopkins Vaccine Initiative. Otherwise healthy children do sometimes die from RSV, but that is most likely to occur in the first six months of life, so the deaths of toddlers were especially telling.

“As you can imagine, this sort of stopped vaccine development for a very long time,” Karron says. “You took a pathogen that, even then, didn't kill that many children, and it killed children.”

The disaster was traced to a phenomenon called antibody-dependent enhancement, in which the body produces antibodies that don't adequately protect it and instead exacerbate the infection. Antibody-dependent enhancement had occurred with an early version of a measles vaccine in the 1960s that was later pulled from use, and it has since been reported with the dengue fever vaccine.

But the mechanism that causes this problem varies depending on the pathogen. With dengue, for example, there are four types of the virus, and antibodies to one do not protect completely against all the others. So when a person develops antibodies in response to one dengue serotype and then becomes infected with another, the body tries to fight the second infection with the antibodies from the first and fails, while the infection worsens. The problem with RSV was that scientists didn't know what caused its antibody-dependent enhancement.

For the next two decades RSV vaccine progress stagnated. Researchers developed multiple live attenuated vaccines using virus that was weakened instead of neutralized, or deactivated. These didn't cause enhanced disease, but they also didn't make it far in clinical trials. “The problem with live attenuated vaccines is that you've got a relatively small therapeutic window, meaning if they're not attenuated enough, they will cause disease. Too attenuated, and they won't be immunogenic enough to cause immunity,” says Barney Graham, a vaccinologist and senior adviser for global health equity at the Morehouse School of Medicine, who has spent his career studying RSV vaccine development for children but was not involved in those devastating early trials. “When you start putting it into lots of children, who have so much variability themselves, it's hard to get that therapeutic window to fit all of the children,” says Graham, who was also instrumental in developing a COVID vaccine.

Before RSV vaccine research could lead anywhere fruitful, researchers needed to know what had gone so wrong in the 1960s trials to cause antibody-enhanced disease. The mystery wasn't solved until 2008, when Fernando P. Polack, founder of the Infant Foundation in Argentina, and his team at Johns Hopkins University published a study in Nature Medicine describing experiments with mice that demonstrated how the antibodies produced by the vaccinated children's immune systems bound to RSV but did not neutralize it.

With those antibodies failing to neutralize the virus, it proliferated, resulting in a lot of ineffective antibodies and a lot of viral antigens clumping together with those antibodies, Graham says. These clumps built up in the tissue, attracting immune system proteins that caused a cascade leading to inflammation. That inflammation damaged lung tissue and created mucus that constricted the airways and made the children sicker than they would have been with no preexisting antibodies. But a big question remained: Why didn't those antibodies adequately neutralize the virus? Later that same year a serendipitous meeting would answer that question and lead to the final steps necessary to make RSV vaccines a reality.

A Tale of Two Protein Shapes

In 2008 Jason McLellan, now a molecular biologist at the University of Texas at Austin, had just begun a postdoctoral fellowship at the National Institutes of Health Vaccine Research Center, where he met Graham. Graham, who was leading RSV vaccine efforts there, learned that McLellan, who specialized in mapping the atomic structure of proteins, wanted to work on something “a little off the radar,” Graham says. “Well, we have no structural information on RSV yet,” he told McLellan. Graham was particularly interested in the “F protein,” the antigen that was the main target for RSV vaccine development.

The idea piqued McLellan's interest. “It became clear that RSV was one of the major childhood pathogens for which we didn't have a vaccine, so working on a vaccine that can help save the lives of babies and young children was very motivating,” he says. The pair's goal—discovering the F protein's structure—would become the key to creating a successful vaccine. But the F protein isn't stable: when it fuses with a cell, allowing the virus to enter and hijack the cell to reproduce, it changes shape. Antibodies against the so-called postfusion shape—the ones produced by the immune systems of the children in the 1960s trials—don't fully neutralize the circulating form of the virus before it binds to cells, known as the prefusion form. But if a vaccine could induce antibodies against this form, they should bind properly with the virus's active form. The trick was to figure out what that prefusion protein looked like and how to lock it into that shape.

To do that, the two researchers first took a harder look at the other form, the postfusion protein. “That's the one that's easy to make; it's stable, and so it's relatively easy to work with,” Graham says. Knowing the structure of both the prefusion and postfusion proteins would enable Graham and McLellan to understand how the protein morphs between the two shapes. By 2010 McLellan had determined the structure of the postfusion protein using x-ray crystallography. Next, to decipher the prefusion protein, his team needed to find an antibody that neutralized the virus without binding to the postfusion protein.

In collaboration with researchers at Xiamen University in China, McLellan and Graham screened over 13,000 mouse antibodies until they found one that effectively neutralized RSV's prefusion F protein without binding to the postfusion one. The winning antibody was about 50 times more potent at neutralizing the virus than palivizumab, the FDA-approved antibody against RSV. The finding suggested previous vaccine candidates had failed because none produced antibodies potent enough to neutralize the virus.

The researchers then used a similar human antibody to determine the F protein's structure. “After we had that structure, everything really fell in place,” Graham says. “All of a sudden, we had a new, very vulnerable target on the virus for making a vaccine.”

Now, finally able to see exactly where the antibodies attached to the protein, McLellan and the team substituted two amino acids in the sequence that encodes the F protein to create a covalent bond that effectively “stapled” the protein together, preventing it from pulling apart into its postfusion shape. They published their method in late 2013, then spent the next few years growing human cells that would produce the prefusion protein and learning how to purify it for use in a vaccine.

The first small clinical trials to evaluate the vaccine's safety began in 2017 and produced encouraging results two years later. By then, “RSV vaccines had a life of their own,” Graham says, as the pharmaceutical industry took over their development.

McLellan, meanwhile, turned his focus to coronaviruses. The RSV work would ultimately pave the way for determining the spike protein structure of SARS-CoV-2, the virus that causes COVID, and enable Moderna, Pfizer and other companies to develop a COVID vaccine in record time. The era of protein-structure-based vaccine design—starting with figuring out a pathogen's protein structure and building a vaccine around it—had begun.

Vaccines Arrive

The two pharmaceutical companies that took the lead in developing vaccines based on this science were GSK and Pfizer. The FDA approved GSK's Arexvy, the first RSV vaccine for adults 60 and older, on May 3 and then approved Pfizer's Abrysvo for the same age group on May 31. GSK said its vaccine is 94 percent effective against severe disease and 83 percent effective against symptomatic disease in adults 60 and older. Pfizer said its vaccine is 86 percent effective against severe disease with at least three symptoms and 67 percent effective against symptomatic disease with at least two symptoms in adults age 60 and older. A CDC advisory panel voted in favor of recommending that people in that age group may get an RSV vaccine in consultation with their health-care provider but stopped short of recommending them for all older adults. Both vaccines should be available this fall as the next RSV season begins.

The more challenging need was a vaccine to protect newborns, especially because immune systems less than four months old are too immature to respond to most vaccines and develop the immune memory needed to fight a disease. Researchers used the same approach that protects newborns from flu and pertussis—administering a vaccine during pregnancy so the parent's antibodies will cross the placenta to the fetus. An RSV vaccine would protect infants for the first six months after birth, when babies are most at risk for serious complications from the disease.

The FDA's Vaccines and Related Biological Products Advisory Committee voted on May 18 to recommend FDA approval for Pfizer's parental RSV vaccine candidate. Pfizer has said its RSV vaccine for pregnant people is 82 percent effective against severe RSV in newborns for up to three months and 69 percent effective through six months. The committee was impressed with the vaccine's effectiveness, but some members had reservations about its safety data.

The FDA advisory panel voted unanimously in favor of the Pfizer vaccine's effectiveness and 10 to 4 in favor of the vaccine's safety. The primary reason for the four nay votes for safety was that the premature birth rate was slightly higher in the vaccinated group but not statistically significantly so. “I think the four votes were really just an abundance of caution,” Graham says. “They weren't saying it wasn't safe. They said they wanted more information before they said yes,” although it's not easy to get that additional information until studies are conducted in the general population.

GSK stopped its own trial of a vaccine for pregnant people because infants born in the vaccine group were 38 percent more likely to be born premature.

Beyond vaccines, AstraZeneca and Sanofi announced in March 2022 that nirsevimab, a prophylactic monoclonal antibody drug similar to palivizumab, is 75 percent effective against cases of RSV that require medical care in infants younger than one year with no history of RSV—and the protection lasts five months, which is about the length of a typical RSV season. Europe approved nirsevimab in November 2022, and the FDA approved it in July 2023. A similar long-acting monoclonal antibody made by Merck, clesrovimab, is in late-stage trials.

One challenge will be assuring the protection of children in low-income families, who are already more vulnerable to worse outcomes from RSV. The U.S. Vaccines for Children program ensures all eligible children can access vaccines recommended by the CDC, even if they lack insurance. But that program doesn't currently include vaccines for pregnant adults, and as of this writing, it has yet to be determined whether the program will cover nirsevimab.

It's still not clear how insurance companies might decide whether and when to cover nirsevimab. Either way, by the end of 2023, it's very likely that infants, like older adults, will have at least one highly effective option to reduce their risk of RSV for the first time in the half a century since scientists began the effort.