Jason McLellan and Barney Graham knew they had it. At the end of November 2012, Graham recounts, they could see the crystallized respiratory syncytial virus (RSV) F protein at atomic resolution using X-ray diffraction. Earlier research had suggested that if they could stabilize one particular conformation—the so-called pre-fusion form—of the F protein, which RSV uses to enter human cells, they might be able to use it as an effective vaccine, says Graham, who has now recently retired after more than 20 years at the Vaccine Research Center at the National Institute of Allergy and Infectious Diseases (NIAID) and has joined Morehouse School of Medicine. Visualizing that protein conformation was a leap forward on the path.
An effective immunization for RSV, a virus that had thwarted vaccine development attempts for 60 years, could save lives and protect lung health for millions around the world. RSV infection resembles a cold in older kids and adults, but can be far more serious in people who are much younger or older. More than 60,000 hospitalizations for RSV occur each year among US adults 65 and older, with upwards of 6,000 deaths. Meanwhile, one in 50 infants in the US—most of them full-term, healthy kids—are hospitalized for RSV each year.
“In the winter season, if you go to the intensive care unit, you would probably see that more than half of all the children are mechanically ventilated for RSV infection,” says Louis Bont, a pediatric infectious disease specialist at University Medical Center Utrecht in The Netherlands, who specializes in RSV and who advises several companies working on RSV vaccines. In lower-income and middle-income countries that lack access to oxygen and ventilators, RSV is the second leading cause of infant mortality after the neonatal period, following malaria. Despite this, RSV is a virus that has historically received little public attention. “I’ve been a pediatrician for 25 years. Everybody knows meningitis, everybody knows pneumonia, everybody knows urinary tract infections, but RSV is really a big unknown” to the public, says Bont.
Until now. Just 14 years after McLellan (then a fellow in Peter Kwong’s NIAID lab) and Graham first began working together, a deluge of clinical trials based on their findings is yielding promising results for vaccines against RSV. On February 28th and March 1st, the FDA will convene an Advisory Committee meeting to discuss the safety and effectiveness of two of these vaccines in older adults. The news is particularly relevant after this past winter’s severe RSV season, as the virus surged in older adults, and children’s hospital wards have filled up with RSV-afflicted kids. Pediatricians’ offices have started to test for RSV along with the flu and COVID-19. James Campbell, a pediatric infectious disease specialist at the University of Maryland School of Medicine who has run trials on RSV vaccines with industry partners, says: “It’s such a bad disease for young children that if we have a way of protecting them, all pediatricians and parents are going to want to have that option.”
From failures to a structure-based vaccine
Scientists first isolated RSV from chimpanzees back in 1956, but later studies showed that most humans already had antibodies against the virus, suggesting it had been around for a long time. In the 1960s, encouraged by recent successes in the development of other vaccines, researchers tried creating a vaccine for RSV, giving formalin-inactivated RSV to infants who had never had RSV before. The inoculation did not reduce kids’ chances of getting RSV, and worse still, made them even sicker the first time they encountered the virus in the real world. Of 20 babies who got RSV during the trial, 16 were hospitalized, two of whom died.
As a result, “for a long time people didn’t dare to develop RSV vaccines,” says Bont. Other sorts of preventive medicine against the virus fared slightly better: In 1998, MedImmune (later acquired by AstraZeneca) successfully developed a monoclonal antibody called palivizumab (Synagis) that targets RSV’s F protein and reduces risk of hospitalization for high-risk infants born prematurely or with preexisting heart or lung disease. Given as a monthly shot during RSV season, palivizumab is only approved for the most high-risk kids, leaving other children and adults without protection—although more-recent antibody treatments may help address that gap (See Sidebar).
By the 1990s and 2000s, some researchers had returned to working on RSV vaccines, typically using the F protein in its so-called post-fusion conformation. Formulations developed around this time were safer than before—researchers better understood the immunology behind the 1960s vaccine failure, tested vaccines in adults who had already been exposed to RSV, and avoided using vaccines with whole inactivated RSV in RSV-naive kids. The shots were still largely ineffective at protecting people from RSV.
It’s such a bad disease for young children that if we have a way of protecting them, all pediatricians and parents are going to want to have that option.—James Campbell, University of Maryland School of Medicine
In 2012, however, a group in Spain showed that the majority of neutralizing activity in rabbits inoculated with a recombinant vaccinia expressing the F protein targeted the flighty pre-fusion form of F, not the stable post-fusion form. Not much was known about pre-fusion F because it was so transient, but soon afterwards, Graham and McLellan and their teams at NIAID’s Vaccine Research Center figured out the structure of the pre-fusion F protein as it was bound to a powerful antibody called D25. In particular, Graham and colleagues saw that D25 and other strong neutralizing antibodies attach to the apex of the pre-fusion F protein, each at a different angle, interfering with the protein rearrangement required for the virus to fuse with and enter cells. They named this apical region antigenic site 0, and obtained high-resolution X-ray diffraction data on the complex’s crystals, solving the structure with molecular replacement.
Because of its location on the pre-fusion F’s apex, antigenic site 0 is accessible to antibodies even on the crowded surface of a virus, helping to explain why the strongest natural antibodies to RSV target pre-fusion F. “That’s when we really got serious about trying to do the protein engineering steps to stabilize the molecule in the pre-fusion form,” in order to use it as a vaccine antigen, Graham says.
Within a year, they’d cracked it. In a 2013 Science paper, the researchers reported that if they added cysteine residues to certain sites and filled some cavities in the protein structure, the protein remained in the pre-fusion state. Injecting this stabilized pre-fusion F protein, which they called DS-Cav1, into mice and macaques generated an RSV-specific neutralizing antibody response many times higher than what is needed to thwart RSV infection. “It was much more immunogenic in terms of inducing neutralizing activity than anything we had [ever seen] before,” says Graham, who is named with McLellan, Kwong and others as a coinventor on patents for pre-fusion F protein antigen design, and consults for RSV vaccine developers.
RSV vaccines’ journey to the clinic
In a Phase 1 study with results published in 2019 and 2021, Graham’s group showed that healthy adults who received a shot of a DS-Cav1 vaccine had more than a 10-fold increase above baseline in neutralizing activity from antibodies, most of which targeted pre-fusion F (although some targeted both pre- and post-fusion F). In addition, healthy adults who received one of several different doses of DS-Cav1 experienced no serious vaccine-related side effects, and all doses of the vaccine boosted antibody levels for at least 10 months post-immunization.
The pharmaceutical industry caught on to the approach. Of six ongoing vaccine trials currently in Phase 3, five exclusively use pre-fusion F, and one uses pre-fusion F with post-fusion F and other proteins. Although most of the Phase 3 trials are in older adults, one ongoing trial is in pregnant people—antibodies transferred across the placenta to the fetus could provide protection for several months after birth. There are at least 17 more Phase 1 and 2 vaccine trials targeting a variety of RSV proteins, according to a recent review of the topic and a compilation of trials by the nonprofit health organization PATH.
One of the most advanced candidates is a vaccine made by GSK (formerly GlaxoSmithKline) for older adults. The formulation contains stabilized pre-fusion F protein from a strain within the RSV-A subgroup, one of the two subtypes of the virus (the other is RSV-B), together with an adjuvant to boost recipients’ CD4+ T cells, part of the cellular immune response. The vaccine also induces neutralizing antibody responses against RSV-B. Results of the company’s Phase 3 trial of the vaccine were published today (February 16) in the New England Journal of Medicine: the vaccine was more than 82 percent effective at preventing RSV-related lower respiratory tract disease (LRTD), and GSK spokesperson Alison Hunt writes to The Scientist in an email that the vaccine also has “over 94% efficacy observed against LRTD in adults with at least one comorbidity of interest, the population that drives the majority of RSV-hospitalizations, and in severe disease.” In addition, she writes that it’s approximately 94 percent effective against LRTD in adults aged 70-79 and against severe LRTD. She writes, “The vaccine candidate has the potential to be the first available to help protect adults aged 60 years and older from lower respiratory tract disease caused by respiratory syncytial virus.” The US Food and Drug Administration (FDA) has granted GSK’s vaccine priority review and an advisory committee will meet on March 1st to discuss the vaccine, the European Medicines Agency has granted it accelerated assessment, and regulatory submissions were accepted in Japan.
Pfizer, meanwhile, is using a bivalent vaccine, with pre-fusion F from both the RSV-A and RSV-B subtypes. The formulations don’t include an adjuvant, as Phase 1 and 2 studies in both pregnant people and older adults suggested adjuvants “did not substantially improve the immune response beyond what was accomplished with the pre-fusion F proteins by themselves,” writes Kena Swanson, vice president of viral vaccines for Pfizer, in an email. The company has recently announced promising findings from two ongoing Phase 3 trials using this approach.
One, which included more than 30,000 adults 60 years and older, found that the vaccine protected people against lower-respiratory tract illness (such as pneumonia and bronchitis) from both strains of RSV and was safe and well-tolerated. The other included approximately 7,400 pregnant people, half of whom were vaccinated in the late second to third trimester of pregnancy. Pfizer reported in November that the vaccine was more than 81 percent effective at protecting infants from severe lower respiratory tract illness due to RSV through the first 90 days after birth, and 69 percent effective through the first 6 months; there were no safety concerns for the mothers or infants. Last December, the FDA accepted for priority review Pfizer’s Biologics License Application for the vaccine. An FDA Advisory Committee will meet to discuss Pfizer’s RSV vaccine on February 28th.
Some companies have been experimenting with other types of vaccines. Janssen’s formulation, which is currently being evaluated in a Phase 3 trial of older adults, contains an adenovirus vector expressing the pre-fusion F protein. Denmark-based Bavarian Nordic also has a viral vector vaccine, in this case based on a so-called Modified Vaccinia virus Ankara that does not replicate in the body. Its vaccine contains genes for five RSV antigens, including the F protein expressed in both its pre- and post-fusion forms. “Instead of just generating antibodies against the pre-[fusion] F protein, you generate antibodies against a variety of [RSV] proteins,” says Victoria Jenkins, director of clinical strategy for RSV at Bavarian Nordic. The company is “hoping that that’s more advantageous than just focusing on the narrower immune response against just one protein.” Results are due in mid-2023, although preliminary findings in a human challenge trial have suggested promising efficacy, says Jenkins.
Employing a different approach, Moderna is using the same mRNA technology that’s in its SARS-CoV-2 vaccine. In early 2022, the company launched a Phase 3 trial in adults 60 and older to test the efficacy of mRNA coding for the pre-fusion F protein. “Like for [the] COVID-19 [vaccine], the protein is produced in the body,” says Francesca Ceddia, senior vice president of respiratory vaccines at Moderna. “It’s an advantage because it mimics natural infection.” The company released topline results in January showing that the vaccine is 84 percent effective against RSV-associated lower respiratory tract disease with two or more symptoms in this older adult population. The FDA has given the vaccine a Breakthrough Therapy Designation.
See “The Promise of mRNA Vaccines”
A step toward changing medical care?
Despite the swift pace of progress in RSV vaccine development, clinical research to date has also highlighted gaps in scientists’ knowledge about how these vaccines might protect the wider population.
For example, GSK’s and Pfizer’s trials of older adults both lacked sufficient data to determine the vaccine’s efficacy among people over 80 years old, even though this age group is one of the most vulnerable. Michael Ison, an infectious disease researcher who was a physician at Northwestern University when he ran the GSK trial at this site and presented its findings last October, says that the challenge in determining this was that relatively few participants in that age group were included in the trial, and thus both the vaccine and placebo groups experienced very few cases of RSV. He points out that the GSK trial did find that the vaccine induced strong antibody responses to both RSV-A and RSV-B subtypes and T cell response regardless of a person’s age (even over 80 years) or frailty.
Some trials have also turned up concerns around safety that will require further study. For example, GSK had been running a Phase 3 trial in pregnant people until February 2022, but halted the study early because of a safety concern. “We made the decision to stop enrollment and vaccination . . . following a recommendation from the [trial’s] Independent Data Monitoring Committee,” GSK’s Hunt writes in an email. A February 2023 abstract published in the ReSViNET Conference proceedings explained the study was halted because of “an imbalance in the proportions of preterm births and neonatal deaths between the vaccine and the placebo groups. The imbalance in preterm births was statistically significant. . . . The imbalance in neonatal deaths was considered to be a consequence of the imbalance in preterm births.” The authors wrote that the imbalance was more associated with low- and middle-income countries (relative risk: 1.57) than high income countries (relative risk: 1.04). Phil Dormitzer, Head of Vaccine R&D at GSK, writes in a statement, “We continue to work with study investigators to ensure the best care possible for the women and children involved. These initial findings may be useful for understanding the risks and benefits of RSV maternal immunization more broadly. We continue to collect data and further analysis is ongoing. We are committed to share updates as they become available.”
A more pressing challenge for widespread RSV immunization is that even as several vaccine candidates have shown promise in trials with older adults and with pregnant people, there’s no such vaccine close to being available for children. The biggest need is in newborns, and their immature immune systems likely wouldn’t mount a large enough response to a vaccine, says Bont; only a couple of vaccines (BCG, HBV) work in newborns. Graham notes that there were some live-attenuated virus vaccines being evaluated in the 1990s that are still in development for young infants. Vaccinating pregnant women or giving infants a dose of monoclonal antibodies before their first RSV season could protect them to around six months, Graham says, but “there is still lot of work to do, I think, in vaccine protection of the six-month-old to the five-year-old.” Live-attenuated virus vaccines are moving through development in this age group, but haven’t yet reached Phase 3 (several are in Phase 2). Logistical difficulties loom on the horizon, too. Researchers who spoke to The Scientist emphasize that vaccines, whether for mother or child, will need to be made accessible in low- and middle-income countries, where 99 percent of the infant deaths from RSV occur.
Nevertheless, with vaccine approvals likely imminent, it looks as though years of work on basic RSV structure is about to pay off, says Graham. “I’ve been working on RSV since 1985,” he says. “The first twenty years of my career were spent trying to understand why the RSV vaccine safety problem occurred back in the mid-sixties. . . . So now, getting to this point where we think we can make a vaccine for RSV safely and also now have efficacy is very exciting. It’s gratifying—it’s the end of a long journey.”
Sidebar: Monoclonal Antibodies Protect Infants from RSV
While vaccines against RSV have been progressing through trials in the past year, another, passive form of protection has also been making headlines. AstraZeneca/Sanofi has developed a monoclonal antibody called nirsevimab (Beyfortus) that targets what’s known as the pre-fusion F protein site 0, and that could be given in a shot to babies before or during their first RSV season. (Another monoclonal antibody called palivizumab—which targets both the post-fusion F protein and the pre-fusion F—has been in use since 1998, but must be given monthly, and is only approved for high-risk kids.)
Company researchers reported peer-reviewed Phase 2b study results in 2020: preterm infants born between 29 and 34 weeks given one injection of nirsevimab before their first RSV season had a 78 percent lower incidence of hospitalization than infants in the control group. The company reported further promising results of a Phase 3 trial published in 2022, this time in infants born after 35 weeks gestation. If the FDA approves nirsevimab, the Centers for Disease Control and Prevention’s Advisory Committee on Immunization Practices may vote on whether to recommend it as soon as June 2023. The drug is already approved by the EU’s European Medicines Agency for the prevention of RSV lower respiratory tract disease in newborns and infants during their first RSV season.
Despite the excitement, researchers caution that it’s important to keep collecting data. Pediatric infectious disease specialist Louis Bont of University Medical Center Utrecht says that RSV may evolve resistance to the antibodies, though it’s unlikely. Tonya Villafana, the global franchise lead for infectious diseases at AstraZeneca who headed the studies on nirsevimab, says that so far, this doesn’t seem to be a problem. “We know that site 0 remains highly conserved,” she says.
Another question is whether babies given nirsevimab might have a harder time with the virus when they encounter it later on. Villafana says AstraZeneca has data suggesting that babies given nirsevimab can mount an immune response when they encounter RSV later. James Campbell, a pediatric infectious disease specialist at the University of Maryland School of Medicine, says that “we won’t know for sure until we have longer-term data,” but does note that data recently presented to the CDC’s ACIP show that children who received nirsevimab did not experience a more severe second season of RSV compared to those who had received placebo.