Respiratory Syncytial Virus in Adults Podcast

50+ Years in the Making, RSV Vaccine Arrives

Forest W. Arnold, DO, MSc; Barney S. Graham, MD, PhD


August 09, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider. 

Forest W. Arnold, DO, MSc: Hello. I'm Dr Forest Arnold. Welcome to Medscape's InDiscussion series on respiratory syncytial virus (RSV) in adults. Today we'll be discussing RSV vaccines with Dr Barney Graham. Dr Graham is a senior advisor for Global Health Equity and professor of medicine and microbiology, biochemistry, and immunology at Morehouse School of Medicine.

Welcome to InDiscussion.

Barney S. Graham, MD, PhD: Thank you. I'm glad to be here.

Arnold: We can't talk about the new RSV vaccines without talking about the beginning of the vaccines, which wasn't 5 years ago. It was 1965, just a decade after the very successful polio vaccine was released. Could you take a couple minutes and review those first years and the children's studies?

Graham: RSV was discovered in 1956, as the polio vaccine was having so much success. The initial attempts at making a vaccine were, at that time, to isolate the virus, grow it to high titer, and inactivate it, and that would be the vaccine. And that was done with RSV. A whole inactivated vaccine was tested.

In 1965, there were studies done in four cohorts of children, and in the youngest age cohort, vaccinations were started before 7 months of age. In children who had not yet experienced an RSV infection, there were some problems. And in that cohort of young, antigen-naive children, 20 out of the 31 vaccinated became infected during that winter season, just after Christmas of the 1966-67 season. Sixteen children had to be hospitalized. Two of those 16 children died.

That was a dramatic setback for vaccines in general. It was a real surprise and put a halt to vaccine development for RSV for several decades until the immunology and pathogenesis were better understood.

Arnold: In the mid-1980s, the fusion glycoprotein (F), which we'll call F protein, was isolated. What is the importance of that F protein?

Graham: In the 1980s, we started getting sequences for RSV monoclonal antibodies and characterization of proteins of the virus. There's three surface proteins: the small hydrophobic (SH), G, and the F glycoprotein. The F is considered to be the most important for vaccines and antibodies. Targeting F can neutralize the virus or prevent virus infection, and F is important because it is the protein that mediates membrane fusion. It helps fuse the viral membrane to the host cell or target membrane, and that's how the viral genome gets into the cell to start the replication process. If you can prevent that F protein from doing its job, then you can largely prevent infection at the next cell.

Arnold: There is a pre-fusion form and a post-fusion form. What is the significance of those two forms, and what are they fusing to?

Graham: The F protein is referred to as a class I fusion protein. It's a type of protein on enveloped viruses like parainfluenza, HIV, Ebola, and viruses like that, including coronaviruses. In RSV, the F protein is anchored in the membrane, and it starts out in one conformation, like a spring-loaded trap. As it approaches the cell, the top of it unfolds, and it inserts itself into the target cell membrane.

For RSV, it's mostly the ciliated, bronchial, or epithelial cells, and the type I pneumocytes in the alveolar space. That attachment then leads to a rearrangement of the protein, where there are three identical parts — it's a trimeric protein. The three identical protomers grab the host cell, and then the two ends — the one end inserted into the host and the other end inserted into the viral membrane — come back together, they're pulled together, and that creates a fusion core that allows the virus replication process to start. When moving from prefusion to postfusion, you're also moving from functional to non-functional form of the protein.

Arnold: It's the postfusion that is short-lived. It's usually in a pre-fusion state, is that correct?

Graham: On the native virus that's budding off of cells, the protein starts as a prefusion. That is the spring-loaded form. The postfusion is a very stable molecule. Before work around 2010 and 2012, we didn't really understand the important differences between these two forms of the protein in terms of what antibodies recognize. The post-fusion form of F is very stable. Once it makes that rearrangement, it is an extremely stable molecule, so it's relatively easy to make, it's easy to store, it's easy from a manufacturing process, and it is what you get when you isolate the protein from a virus or make the protein in cell culture. The postfusion is what people had to work with up until around 2012 or 2013.

Arnold: The new vaccines utilize that F protein as part of their mechanism of action, correct?

Graham: That's correct. Vaccine studies resumed in the 1990s after there was a better understanding of the vaccine-enhanced illness, where we learned only antigen-naïve people are susceptible to that kind of event. Once people are infected with RSV, which starts early in life and then repeated throughout life, they are primed in the right way for an immune response. Vaccination or infection doesn't result in this kind of vaccine-enhanced disease.

In the 1990s, immunization studies were resumed, and there were five phase 3 studies with what turned out to be the post-fusion F molecule. And they all worked to a certain extent, but they didn't boost neutralizing activity very much, only maybe two- to three-fold on average. None of those phase 3 trials were able to progress. They didn't meet their primary objectives. The big thing that happened in 2012 and then published in 2013 is that we were able to capture the atomic-level structure of the pre-fusion form of the F protein. Once we had that — and that was done at the Vaccine Research Center at the National Institutes of Health (NIH) in collaboration with a fellow, Jason McLellan, who was in Peter Kwong's lab and my lab — we worked on this protein and got the structure. Were able to stabilize the structure by putting in some internal mutations and holding it together with what's called a trimerization domain on one end and putting in disulfide and cavity-filling mutations inside. It holds it in the right shape and doesn't let it flip around into the post-fusion form.

Once we had the protein in that form, it turned out to be a much better vaccine antigen. In animals, for instance, you could elicit 40-50, sometimes 80-fold higher neutralizing activity from the pre-fusion molecule than you could from a post-fusion molecule.

Arnold: We'll get back to those neutralizing antibodies, but first, we have to sidestep and talk about COVID-19 for a second. There are two vaccines that just came out, and other drug companies are probably going to follow suit with new FDA-approved RSV vaccines, which makes it sound like the technology has happened after COVID-19 vaccines. But COVID-19 in a sense got in the way of making the RSV vaccines, which actually provided the technology and lessons learned to create the COVID-19 vaccine so quickly. Is that giving too much credit to the RSV vaccine?

Graham: No, I don't think it is. The RSV vaccines — from that initial discovery of the pre-fusion conformation and that being a better vaccine antigen — have been on more of a traditional development timeline. Our phase 1 trial that we did and published in 2019, it took that long to evolve. As the companies have developed their products based on that conformation, it's gone through about a 10-year development timeline, which historically is actually relatively fast for vaccine development that has usually been measured in decades.

This is only 1 decade. But the concept of making this fusion protein in its pre-fusion conformation is what led Jason McLellan and I, when Jason was starting his own lab at Dartmouth Hitchcock Medical Center, to extend the RSV findings to another enveloped virus, the coronaviruses.

Working on Middle East respiratory syndrome (MERS) and then the first severe acute respiratory syndrome (SARS), and eventually human coronavirus HKU1, which is one of our endemic coronaviruses, we were finally able, with Andrew Ward's collaboration at Scripps, to get the pre-fusion form of the spike protein of a beta coronavirus. That allowed us to find stabilized mutations. That research was used in prototype vaccines that we knew by 2019 could protect animals from a lethal challenge with the MERS coronavirus.

This concept of the pre-fusion stabilized protein being a better vaccine antigen that came from RSV was extended to coronavirus and is what prepared us in the field for rapidly making coronavirus vaccines in 2020.

Arnold: Let me summarize to this point. In the mid-1960s, we had a very revealing but horrendous study where many children who were vaccinated for RSV were either hospitalized and two died. Twenty years later, there was the isolation of the glycoprotein F, but it really wasn't until it was outlined in an atomic form that we made progress.

Then, you move forward some more, and they're working on RSV vaccines, and here comes COVID-19. The technology gained from RSV gave us the ability to rapidly form a COVID-19 vaccine. Now, we're 3 years out from the COVID-19 vaccine, and we've got our first RSV vaccines. These are indicated for older adults. Why aren't the RSV vaccines that are out now also for children?

Graham: The field spent 20-30 years working out the safety, and then it worked another 20 years or so working out the way to make vaccines effective. These new vaccines, based on subunit proteins, are being used initially to boost adults who have already been primed with prior infection.

The vaccine for adults, which we know can be given safely because everyone's already been infected with RSV before, are being used to boost immunity in adults. They can boost immunity to supernormal levels — levels that are found in very few people from natural infection. They can provide a level of protection usually for severe disease of up above 90% and even above 70% for mild disease. These vaccines are effective in the elderly. They're also effective in boosting women of childbearing age or pregnant women enough that their antibodies can be transferred to babies and protect the babies for around 6 months. We do have an approach that will be available for babies, but it comes through the mother or through antibodies made in a factory and given directly to babies at birth. That can protect them for around 6 months as well. The thing we still don't have, and the thing that is still being worked on and will probably take a format other than a purified protein, is a vaccine for the 6-month to the 5-year-old child.

Most hospitalization and lung damage happens in children less than 6 months of age, but there's still a lot of disease and hospitalization in children between 6 months and 5 years of age that we would like to eventually have something for. That is still something that's being worked on.

Arnold: Can you describe disease enhancement? By that, I mean RSV vaccine-associated enhanced respiratory disease? Is that a problem that children or infants experience that adults don't because they've already been exposed to RSV?

Graham: Yes, the vaccine-enhanced illness associated with RSV and immunization in infants is called vaccine-associated enhanced respiratory disease (VAERD). That is a syndrome that has an immunological basis. Over the years, we've found two problems. One was an antibody problem, meaning that there were a lot of antibodies induced that could bind the virus protein that did not neutralize the virus infection.

Because the virus could still grow, and because the antibody levels were high, it was found in those early studies to have caused an immune complex-mediated disease where you could find immune complexes in the small airways that activated complement. We think that is part of what happened originally.

Part of the problem to fix that vaccine-enhanced disease is to keep the protein in the conformation that will elicit more potent, effective antibodies. There was also a T-cell problem, and it was found in animal models with some corroboration in people who are primed with a protein or whole inactivated virus-type vaccine that, when they are challenged or infected with RSV, they get more of an allergic inflammation. They get what's called a Th2 response or a CD4 T-cell response. That includes IL-4, IL-5, and IL-13, which means you get a lot more mucus, you get a lot more bronchoconstriction, you get a lot of eosinophils and neutrophils.

And that kind of response is not as effective at clearing RSV, and it causes a lot of problems with airway obstruction. That kind of immune response is something that needs to be avoided, especially in the very young child. Making sure that we have the vaccines set just right and all the children primed in just the right way to create the best possible immune response is going to take a little longer than it has for the adults. In the adults, the T-cell response is already set, and all you have to do is make sure you're boosting the right antibodies, which we can now do with this pre-fusion F protein.

Arnold: The immune-mediated response in children is what causes them a problem if they receive the same kind of vaccine that we're giving to older adults. Was the immune-mediated response the problem with the whole inactivated vaccine back in the 1960s?

Graham: Well, what we know is that the way they made that vaccine in the 1960s was to heat it at 36 degrees for 72 hours. And we know that if you do that with virus, all of the F protein flips into the post-fusion form within about 24-48 hours. That vaccine that was given back in the 1960s was a post-fusion F antigen, it didn't have any pre-fusion F on it. So, it's possible that the pre-fusion F protein by itself could be a solution for young children. If we have the antibody right, the T-cell problem won't be an issue, but because there's a little uncertainty about that, there's going to have to be a lot more work. Most of the vaccines being developed for the young child are live attenuated viruses, which would mimic better the natural infection, or potentially gene-based delivery of vaccines that would mimic infection more.

They also elicit a Th1-type of CD4 T-cell response that would have more interferon gamma and less IL-4. Those kinds of vaccines are coming along, and it's possible that just the protein that's in the right shape would work. But because most developers are mostly focused on safety first and efficacy second, this is going to be done with a lot of caution.

Arnold: Today we've had Dr Graham discussing several important parts of the RSV vaccine. We've talked about its beginning in the 1950s. We've talked about the importance of the F protein, especially the pre-fusion form. We've talked about its impact on the COVID-19 vaccine and how it made that possible quickly, and we've talked about disease enhancement as well as neutralizing antibodies that would be present in mothers who would be vaccinated, which may be available soon, and the protection it could offer fetuses in the first 6 months of life. Thank you so much for joining us. This is Dr Forest Arnold for InDiscussion.


Respiratory Syncytial Virus Infection

Field Evaluation of a Respiratory Syncytial Virus Vaccine and a Trivalent Parainfluenza Virus Vaccine in a Pediatric Population

Structure and Function of Respiratory Syncytial Virus Surface Glycoproteins

Identification and Characteristics of Fusion Peptides Derived From Enveloped Viruses

Brief History and Characterization of Enhanced Respiratory Syncytial Virus Disease

Structure-based Design of a Fusion Glycoprotein Vaccine for Respiratory Syncytial Virus

FDA Approves First Respiratory Syncytial Virus (RSV) Vaccine

Vaccine-associated Enhanced Disease: Case Definition and Guidelines for Data Collection, Analysis, and Presentation of Immunization Safety Data

Th1 and Th2 Responses: What Are They?

Cytokine Pathways in Allergic Disease

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.