How studies of coronavirus immunity can inform better vaccines, treatments

Scientists look for broad-acting antibodies that could protect against new variants of SARS-CoV-2 — and maybe a future SARS-3
Illustration of SARS-CoV-2, showing the external spike proteins and internal genetic material
Understanding the interactions between our immune systems and the novel coronavirus' changeability could help us develop vaccines and therapeutic antibodies that offer protection against future variants. Stock photo courtesy of Getty Images

The novel coronavirus is here to stay — that much we know. There’s still a lot we need to find out. How often will vaccines need to be updated? How often will we need to get boosters? We have the power to shape the answers to these questions through the strategies we use to develop vaccines and immunotherapies against the COVID-19 virus.

But to develop vaccines and therapies that protect against a shape-shifting virus and its future variants, we need to understand critical interactions between our immune systems and SARS-CoV-2. Scientists around the world, including at Fred Hutchinson Cancer Research Center, are delving into our immune response and the virus’ attempts to sidestep it.

Here are the key questions they’re working on now to shape the future of how we prevent and treat COVID-19.

How do antibodies recognize SARS-CoV-2?

Researchers are particularly focused on protective immune proteins called antibodies, which are thought to hobble viruses before they manage to infect their target cells.

We already capitalize on antibodies in several efforts to contain the pandemic: Vaccines work by triggering our bodies to produce protective antibodies, while therapeutic antibodies aim to curtail infections already in progress.

Crafting a widely protective vaccine or antibody-based treatment starts with a deeper understanding of how our immune system creates its protective response. In response to a viral infection, we produce a panoply of antibodies that recognize the pathogen, but they’re not scattered uniformly across its surface. And the most protective antibodies may bind certain crucial regions.

Researchers have concentrated primarily on the coronavirus’s spike protein, which acts as a skeleton key to any cell bearing the ACE2 protein on its surface — which includes cells lining our airways, intestines, and heart. Once inside a human cell, the virus quickly co-opts cellular processes to force our cells to churn out more viral particles. The virus can damage the cells it infects, causing a host of issues throughout the body.

Antibodies that prevent SARS-CoV-2 from entering our cells — a phenomenon called neutralization — often bind somewhere on the spike. But scientists still have a lot to learn about how antibodies counteract the virus, which may extend beyond neutralization to other antibody activities.

Schematic of the spike protein from the side and top, showing key antibody-binding regions.
Hutch scientists identified several important antibody-binding regions on the SARS-CoV-2 spike protein, including on the receptor-binding domain where it contacts ACE2 (RBD, aqua), and two areas lower on the stem: the fusion peptide (FP, green) and a segment that links the HR1 (yellow) and HR2 (red) regions. Image courtesy of the Overbaugh Lab, made with BioRender.

There are several important regions on the spike protein where antibodies can bind. Many scientists have focused on the receptor-binding domain, or RBD, where the spike protein directly contacts ACE2. But antibodies can also bind elsewhere, including the fusion peptide, a region of the spike protein’s stem that helps the virus fuse with its target cell after the RBD contacts ACE2.

Researchers in the lab of Hutch evolutionary biologist and flu expert Dr. Jesse Bloom mapped critical antibody-binding sites — and viral escape routes — on RBD. Graduate student Allison Greaney and postdoctoral fellow Dr. Tyler Starr led the work to examine plasma — the fraction of blood in which antibodies are found — from patients who had been infected with SARS-CoV-2.

Greaney found that though people do make a mixture of antibodies (known as polyclonal antibodies), many coronavirus-blocking antibodies from previously infected people cluster at one region on RBD — a region that was quickly changing in the beta and gamma variants that began picking up steam early this year.

In contrast, she saw that antibodies generated by the Moderna vaccine had a broader binding pattern across the RBD.

Meghan Garrett, a graduate student in Hutch HIV-expert Dr. Julie Overbaugh’s group, focused her attention on the segment of the spike protein that lies between its root and the RBD at its tip, a section that includes the fusion peptide.

She found that most polyclonal antibodies homed in on just a couple of sites; one in the fusion peptide and one in a region along the spike’s stem that links two repetitive areas. But individual immune responses varied.

“Most people had responses to one site or the other, some had responses to both — but overall, the response was not uniform,” Overbaugh said.

Most people in Garrett’s study also had antibody targets sites that were unique to them as individuals, and not seen in other patients.

The findings illustrate how antibodies can target a variety of locations on the spike protein, Starr noted. Vaccine designs that prompt a broad, not-too-focused, antibody response may be harder for the virus to escape than those that generate strong, but overly focused immunity.

How long does immunity last?

How long immunity lasts will determine how often we’ll need to be re-vaccinated against SARS-CoV-2. Immunity’s “lifespan” depends on several factors, including the strength of the original immune response and how well the virus can evolve to evade it.

The coronaviruses that cause the common cold capitalize on mutation to reinfect us every three to five years. Rachel Eguia, a technician in Bloom’s lab, looked to the history of CoV-229E, one of the coronaviruses that causes the common cold, for insight into SARS-CoV-2. She traced the evolutionary tree of CoV-229E’s RBD back to the 1970s, showing that every few years, a dominant variant “wins” out and displaces other CoV-229E mutants.

Will her findings extend to the pandemic coronavirus?

“Although we can’t be certain, it is likely that new variants [of SARS-CoV-2] will sweep away old ones,” Bloom said. “That has already happened once in SARS-CoV-2 with the D614G mutation [now present in all variants of concern], and may now be in the process of happening again with the delta variant.”

Those dynamics are not necessarily bad news, he noted.

“In some ways, these dynamics make vaccine updates easier than if all the variants are maintained,” Bloom said.

This is because if specific variants sweep old ones away, there’s a lot of similarity among viruses found around the world, which would make designing a vaccine that would be effective globally much less challenging. This is the case for influenza, but not HIV, Bloom said. The extreme diversity of HIV at any given moment is one reason that developing a vaccine for that virus has been so hard, he said.

Work from Kathryn Kistler, a graduate student in the lab of Hutch evolutionary biologist Dr. Trevor Bedford, found that the way that RBDs from CoV-229E and another common coronavirus, CoV-OC43, evolve suggests that they are being shaped by our immunity. According to the study, reinfection may be partly due to changes that make variants harder for our antibodies to recognize.  

Eguia looked at the effects of viral mutation on immune recognition from a different angle. She tested the neutralization capability of plasma collected from people over the years to block variants of CoV-229E that arose at different points in time. She found that plasma was best at binding to CoV-229E variants that circulated shortly before the plasma was collected but became less and less potent when tested against CoV-229E variants that arose years or decades later. She also found that some individuals’ neutralization ability dropped sharply over time, while in others it ebbed slowly.

Some common coronaviruses can reinfect us because they just don’t trigger a particularly robust response to begin with. After a few years, our immunity fizzles and they can waltz back in. This timeframe may be accelerated if the variant is more virulent and able to infect more easily, which could be contributing to the success of delta and other variants of concern, Overbaugh said.

Scientists are still studying this question in SARS-CoV-2. It will take time to know how long-lasting our immune responses to natural infection or vaccination will be. Results are currently mixed — some studies report quick immune decay, while others report a normal amount (immunity always falls to some degree after an infection has been vanquished).

“I think the jury's out. People are definitely looking and following the cohorts [of COVID-19 patients and vaccinated people] is going to be super critical,” Overbaugh said.

How do variants arise and escape the immune response?

But the persistence of our own immune response provides only one part of the answer to the question of our immunity to SARS-CoV-2 in years to come. The other part comes from the virus, and its natural ability to change over time.

'We don’t know where SARS-3 will come from. If we can make antibodies or vaccines that can protect against the diversity of these SARS-related coronaviruses, that could be a long-term solution to preventing future spillovers.'

Dr. Tyler Starr, Bloom Lab

When coronavirus genetic material is copied inside our cells, it’s not perfect — every so often the wrong genetic “letter” gets included. If this change helps the virus, it will get passed along as the virus is transmitted to new hosts. A few transmissions and a few varied letters later, and the virus can collect mutations that work together to help it.

The delta variant is a perfect example. It’s acquired mutations that are boosting its transmissibility and speeding its transit around the globe.

Viral mutations can also make it harder for the immune systems of vaccinated or previously infected people to recognize the virus. If the changes happen at key sites that our antibodies recognize, they could enable a variant to slip past our protection. The good news is that even though the delta variant makes vaccine protection somewhat less robust, most vaccines still provide a lot of protection against severe disease, as well as symptomatic disease, from delta.

However, people — and our immune responses — vary, too. Our ability to meet viral variability with our own immunological variability could be good news as the vaccinated and previously infected are confronted with new SARS-CoV-2 variants. Because people tend to produce a slightly different complement of antibodies that target slightly different regions of the virus, a given variant’s path to escape will likely differ in different people, Overbaugh said.

“It actually means that if one virus takes hold, it may or may not actually be able to escape in everybody,” Overbaugh said.

The factors that shape how viral variants arise are complex. The fact that a virus mutates is only the first step. SARS-CoV-2 actually mutates at a slower rate than influenza, so it’s likely that its variants are arising from a combination of mutation and opportunity.

“People think that a lot of the really optimized variants, these variants of concern, happen in immunosuppressed people where the virus kept replicating,” Overbaugh said. “If it starts becoming like HIV, where it's a persistent, replicating virus, it can adapt with the immune system. Then you start having this escape problem.”

If a person’s immune response shuts down their infection quickly, the virus has little opportunity to evolve. Hosts who can’t mount an effective defense allow the virus to incubate long enough to throw out new variants that may help the virus transmit better or cause more severe disease. In these instances, SARS-CoV-2 gets the same opportunity that HIV usually gets: a place to hunker down and continually spit out new mutants — some of which could give it the ability to escape antibody defenses.

This means that the virus has more opportunities to morph into a variant that can re-infect previously infected or vaccinated people as it continues to jump between never-infected, unvaccinated hosts.

There are a couple of factors, including how a virus interacts with the immune system, which help determine whether it will produce successful variants, said Greaney. In some cases, mutations that would allow a virus to escape immunity harm it too much in other ways. Measles is a case in point: it also mutates, but these mutations don’t allow it to escape our immunity. We’ve had the same measles vaccine for more than 60 years. The influenza vaccine, in contrast, must be updated yearly to accommodate that virus’ rapidly shifting proteins.

Determining how quickly SARS-CoV-2 escapes immunity will be key to determining how often boosters are needed. It’s too soon to say whether COVID updates will occur on a yearly basis, like flu shots, or on a longer timescale.

So how could various mutations in SARS-CoV-2 help it escape antibodies or immunity triggered by infection or a vaccine? By studying this question, scientists can reveal strategies that might make vaccine less susceptible to viral escape.

Greaney and Starr examined the effects of viral mutations on binding by both monoclonal antibodies — single antibodies that bind just one place on the virus — and polyclonal convalescent plasma, containing whatever mixture of antibodies were produced by a person infected with SARS-CoV-2. They tested mutations that aren’t yet found in prominent viral lineages, but could theoretically arise as the virus continues evolving.

(Neither Bloom’s nor Overbaugh’s groups work with whole viruses. Instead, they employ zoomed-in approaches that allow them to produce specific regions of the spike protein. Starr developed a system to cause yeast to produce the RBD, while Overbaugh’s lab looks at other areas of the spike protein produced by viruses that infect bacteria. Both teams use a technique called deep mutational scanning to introduce different mutations in their spike protein regions of interest.)

Greaney and Starr found that how much an RBD mutation affected antibody binding varied among patients. Even so, a few spots on the RBD grabbed most of the immune system’s attention — including a rapidly changing location called E484. This result was “a bit worrying,” Greaney said, because changes here were most likely to affect how well polyclonal plasma bound RBD — and it had already mutated in the novel coronavirus’ beta and gamma variants.

Garrett found less consistency among mutations that conferred antibody escape along the stem of the spike protein. These areas are less changeable among viral variants, which means they may be attractive additions to future vaccines designed to provide people with broader protection against many different variants, Overbaugh said.

If a variant is too different from its original strain, an updated vaccine may not boost prior immunity — instead, it may start a new immune response from scratch.

“That’s why we've been so interested in regions that are more [evolutionarily] constrained to add into the mix,” Overbaugh said. Among vaccine developers “there is interest in this idea for really a conserved target, maybe optimizing how the immune system responds.”

Both Overbaugh’s and Bloom’s groups have also begun asking similar questions of antibodies generated after vaccination against COVID-19. Garrett’s early results looking at the fusion peptide and linker region on the spike protein’s stem suggest that vaccines may elicit more-consistent immunity, but also direct antibodies to focus on a couple new sites.

Consistent with her findings that vaccination generates a broader antibody response against the RBD than infection, Greaney found that single viral mutations had less of an effect on neutralization by polyclonal antibodies from vaccinated people than from previously infected people. But a couple of mutations seen in variants of concern, including E484K in beta and gamma variants, and the L425R mutation seen in the delta variant, did reduce neutralization by vaccine-elicited antibodies by a modest-to-moderate amount, she said.

But vaccines aren’t the only anti-coronavirus strategy that are vulnerable to viral mutations. Therapeutic antibodies have been developed to treat COVID-19 patients in hopes of lessening their disease. So far, five monoclonal antibodies have received emergency use authorization to be used as COVID-19 treatments, either individually or in two-antibody mixtures.

Unfortunately, these are also susceptible to viral escape. For example, viral variants rendered the monoclonal antibody bamlanivimab ineffective as a single therapeutic antibody because it happened to target some of the spike protein’s most variable sites. In June, the F.D.A. paused distribution of the combination of bamlanivimab and another monoclonal antibody, etesevimab.

Greaney and Starr had begun their work on SARS-CoV-2 in 2020 by examining how SARS-CoV-2 mutations could affect neutralization by monoclonal antibodies. They’d found that many individual antibodies have different escape mutations, underlining how two-antibody mixtures could limit the potential for viral escape.

Unfortunately, Starr said, variants such as beta and gamma have arisen that combine two mutations that knock off both antibodies in one of the first authorized therapeutic antibody mixtures. And as evidenced by Greaney’s and his work, the same mutations that help SARS-CoV-2 shrug off natural immunity also help it dodge commercial antibodies. This means that some genetic changes could give the virus a two-for-one deal by rendering vaccines less effective and therapeutic antibodies ineffective in one evolutionary swoop.

Part of the reason that commercial antibodies may be so vulnerable to escape is due to how they were selected and developed, Starr noted. In the desperate rush to develop therapeutics that could help treat COVID-19, some companies chose the most potent to bring to the clinic. So far, some of the most potently neutralizing antibodies have also been those that target the virus’ most fast-changing regions. Apparently the RBD can change a lot while retaining its ability to connect with ACE2 (as Starr and Greaney showed in a separate study mapping mutations that affect how RBD interacts with ACE2).

How might we limit the coronavirus’s powers to escape immunity?

Starr wanted to seek out the antibodies and epitopes that could limit viral escape. (Epitopes are the structures to which antibodies bind.) Are there antibodies that can neutralize a broad swath of SARS-CoV-2 variants? If so, what epitopes would they bind, and how can we better identify these antibodies for therapeutic development?

To do this, he teamed up with researchers at San Francisco, California-based Vir Biotechnology, who had been studying coronaviruses since SARS-CoV-1 emerged in 2002. Starr established assays that would allow him to map which mutations allowed SARS-CoV-2 RBD to escape when pitted against 12 monoclonal antibodies. To study how well the antibodies retained their ability to neutralize across extremely divergent SARS-related coronaviruses, he also tested the antibodies against an array of RBDs from 45 SARS-related coronaviruses, including some isolated from bats and pangolins.

For the most part, Starr saw that antibodies traded potency for breadth: Those that neutralized SARS-CoV-2 RBD very effectively were less able to recognize RBD from more distantly related viruses. But included in the 12-antibody panel was one that bucked this trend.

This antibody, dubbed S2H97, had been isolated by VIR scientists from a patient who’d contracted SARS-CoV-2. And it was a standout, Starr said. Though the RBDs they tested ranged from 95% similarity to SARS-CoV-2, to just 60% similarity, and were separated by sometimes hundreds of years of evolution, this antibody bound them all. This included a divergent group of RBDs from bat coronavirus that have never spilled over into humans.

The key may be in what S2H97 targets, a so-called cryptic epitope: deeply hidden within the closed RBD, only accessible when RBD opens as it contacts ACE2.

Though rarely recognized by our immune system, this epitope is incredibly similar across SARS-related coronaviruses, Starr said. If an antibody binds this site, it can block a wide range of those viruses.

There are other antibodies that neutralize SARS-CoV-2 more potently than S2H97, but none that can neutralize so broadly. And S2H97’s potency is nothing to sneeze at: Starr and his collaborators at Vir showed that it protects hamsters against SARS-CoV-2 infection.

Starr hopes that findings like this inform future vaccine design by highlighting immune targets that could trigger broad, hard-to-escape immunity.

“If you know all the antibodies that are out there and what kinds you want, that can help you be a little bit more principled in designing vaccines that give you that type of antibody response [you’re looking for],” he said. “And we don’t know where SARS-3 will come from. If we can make antibodies or vaccines that can protect against the diversity of these SARS-related coronaviruses, that could be a long-term solution to preventing future spillovers.”

The findings could also be used to improve strategies for developing therapeutic antibodies for COVID-19 — whether or not you’re bearing future pandemics in mind, Starr said.

“Even if you only care about SARS-CoV-2, it can help to identify these epitopes that are not changing over moderate to long-term evolution,” he said. “They're probably not going to change in short term evolution as well.”

Several of the antibody-binding spike regions that Garrett and Overbaugh identified fit this bill, and they’re now working to isolate the antibodies that bind at those sites.

Starr also argued that assessing potential antibodies using breadth in addition to potency could help companies identify more robust antibodies.

The emerging picture of how our immune systems interact with SARS-CoV-2 could help us develop better COVID-19 immunotherapies and better vaccines. Enduring antibody responses that target more than a few areas and focus on parts of SARS-CoV-2 that the virus can’t afford to modify may be the key to reducing how often we update vaccines and get booster shots. Such vaccines, combined with therapeutic antibodies that block a range of variants yet to come, could help make SARS-CoV-2 less of a threat in future years.

“We need a lot more antibodies,” Starr said. “The way that this virus is going to continue to evolve — you don't know. … We need to develop a diverse arsenal of antibodies to target many epitopes.”

The projects discussed in this story were funded by a combination of federal grants, awards from private foundations and philanthropic donations.

Sabrina Richards, a staff writer at Fred Hutchinson Cancer Center, has written about scientific research and the environment for The Scientist and OnEarth Magazine. She has a PhD in immunology from the University of Washington, an MA in journalism and an advanced certificate from the Science, Health and Environmental Reporting Program at New York University. Reach her at

Are you interested in reprinting or republishing this story? Be our guest! We want to help connect people with the information they need. We just ask that you link back to the original article, preserve the author’s byline and refrain from making edits that alter the original context. Questions? Email us at

Related News

All news
'Everything we’ve been training for is happening now' How early-career virologists dropped everything to seize the opportunity, and face the challenges, of studying the coronavirus September 16, 2020
What happens if the coronavirus’s spikes mutate? First-of-its-kind study explores how changes in the virus’s ‘lock picks’ could help or hinder its ability to sneak into our cells August 13, 2020
Antibodies a hot topic in COVID-19 research Hutch scientists tap their antibody expertise for new tests, treatments, vaccines June 24, 2020

Help Us Eliminate Cancer

Every dollar counts. Please support lifesaving research today.