The Everything Vaccine: How Scientists Are Trying to Build One Shot to Rule Them All

For decades, vaccine development has operated on a one-pathogen, one-vaccine model. Measles gets a shot. Polio gets a shot. COVID-19 arrived and, after a frantic global sprint, it got several shots. But a growing cadre of researchers is now pursuing something far more ambitious — vaccines designed not against a single disease, but against entire families of viruses, or even against threats that haven’t emerged yet.

The goal sounds almost absurdly expansive. And yet the science is catching up to the ambition.

According to The Economist, laboratories around the world are working on what some have informally dubbed “everything vaccines” — immunizations engineered to provide broad protection across multiple strains, multiple variants, and in some cases, multiple species of pathogen. The concept isn’t entirely new. Scientists have long dreamed of universal flu vaccines that could eliminate the need for annual reformulation. But the tools available today — particularly mRNA technology, structural biology, and computational protein design — have made what was once a theoretical exercise into an active, well-funded research program.

The implications for public health are enormous. So are the scientific hurdles.

From Narrow Targets to Broad Shields

Traditional vaccines work by training the immune system to recognize a specific feature of a specific pathogen — usually a surface protein. The immune system sees the protein, generates antibodies, and remembers the encounter. If the real virus shows up later, the body is ready. This approach has been spectacularly successful against diseases like smallpox and polio, where the virus doesn’t change much over time.

But many of the most dangerous pathogens are shape-shifters. Influenza mutates its surface proteins constantly. Coronaviruses, as the world learned during COVID-19, can evolve new variants with alarming speed. HIV is so mutationally prolific that a single infected person may harbor thousands of slightly different viral strains at once. Against these targets, traditional vaccines are perpetually playing catch-up.

The new approach flips the script. Instead of targeting the parts of a virus that change, researchers are looking for the parts that don’t — conserved regions of viral proteins that remain stable across strains and even across species. These conserved regions are often structurally essential to the virus. They can’t mutate without crippling the pathogen’s ability to function. Hit those targets, and you’ve got a vaccine that works whether the virus is this year’s flu strain or next year’s.

It’s an elegant idea. The problem is that conserved regions are often hidden deep inside the viral structure, shielded from the immune system by the very surface proteins that traditional vaccines target. Getting the immune system to see and respond to these buried targets requires creative molecular engineering.

This is where the mRNA platform, proven during the COVID-19 pandemic, becomes particularly useful. mRNA vaccines can be designed and manufactured quickly, and they can encode complex protein structures that present conserved viral regions to the immune system in precisely the right configuration. Researchers at the National Institutes of Health, for instance, have been developing nanoparticle-based vaccines that display fragments from multiple coronavirus strains simultaneously, training the immune system to focus on the features they all share rather than the features that distinguish them.

As The Economist reports, this multi-display strategy has shown promising results in animal models, generating antibodies that neutralize not just known coronaviruses but also related viruses that the vaccinated animals were never exposed to. That kind of cross-reactive immunity is exactly what a pan-coronavirus vaccine would need to provide.

Similar work is underway for influenza. The stalk of the hemagglutinin protein — the “H” in designations like H1N1 — is far more conserved than the protein’s head, which is what current flu vaccines primarily target. Multiple research groups, including teams at the Icahn School of Medicine at Mount Sinai and at the NIH’s Vaccine Research Center, have been engineering immunogens that expose the stalk while hiding the head, forcing the immune system to generate antibodies against the stable part of the molecule. Early-stage clinical trials have been encouraging, though the antibody responses have sometimes been weaker than researchers hoped.

Weak responses are a recurring challenge. The conserved regions of viral proteins tend to be less immunogenic than the variable regions — meaning the immune system doesn’t naturally mount a strong response against them. Evolution has, in a sense, optimized these viruses to hide their vulnerabilities. Overcoming that evolved stealth is one of the central technical problems in broad-spectrum vaccine design.

Computation, AI, and the New Toolkit

What’s changed most dramatically in recent years isn’t just the vaccine platforms. It’s the tools for understanding viral structure and predicting immune responses.

Computational protein design, powered in part by machine-learning systems like DeepMind’s AlphaFold, has given researchers an unprecedented ability to model how viral proteins fold, how antibodies bind to them, and how mutations might alter those interactions. This matters enormously for broad-spectrum vaccine design, because the entire strategy depends on identifying molecular targets that are structurally constrained — regions where mutations would be so disruptive to the virus that they’re effectively forbidden by natural selection.

Before tools like AlphaFold, identifying those regions required painstaking experimental work — crystallizing proteins, imaging them with X-ray diffraction or cryo-electron microscopy, and testing hypotheses one at a time. Now, researchers can screen thousands of potential vaccine targets computationally, narrowing the field before any lab work begins. The speed increase is staggering.

And it’s not just about structure. Machine-learning models are also being trained to predict which protein fragments will be most effective at stimulating T-cell responses — a critical component of lasting immunity that antibody measurements alone don’t capture. T cells don’t prevent infection the way antibodies do, but they’re essential for controlling infections once they start and for providing long-term immune memory. A vaccine that generates strong T-cell responses against conserved viral proteins could provide durable protection even if antibody levels wane over time.

This dual focus on antibodies and T cells represents a significant shift in vaccine design philosophy. For years, the field was dominated by what immunologists call the “correlate of protection” problem — the question of what measurable immune response actually predicts whether a vaccine will work. Antibody titers were the go-to metric, partly because they’re relatively easy to measure. But the COVID-19 experience demonstrated that T-cell immunity plays a critical role in preventing severe disease, even when antibody levels drop. Broad-spectrum vaccines are being designed with both arms of the immune system in mind.

There’s also growing interest in mucosal vaccines — formulations delivered to the nose or throat rather than injected into muscle. The respiratory tract is where most airborne viruses first establish infection, and mucosal immunity can potentially stop a pathogen before it gains a foothold. Several groups are working on intranasal broad-spectrum vaccines, though the challenges of generating strong, durable mucosal immunity remain substantial.

Not everyone is convinced the everything-vaccine vision is realistic. Some immunologists argue that the immune system has inherent biases — shaped by prior infections and vaccinations — that make it difficult to redirect toward conserved but poorly immunogenic targets. This phenomenon, sometimes called “original antigenic sin” or, more recently, “immune imprinting,” means that the body tends to recall and boost responses to the first version of a pathogen it encountered, even when presented with a new variant. Overcoming imprinting may require entirely new priming strategies, not just better antigens.

Others point to practical concerns. Regulatory pathways for traditional vaccines are well established: you pick a target pathogen, run clinical trials, and measure whether the vaccine prevents disease caused by that pathogen. But how do you run a clinical trial for a vaccine designed to protect against a virus that doesn’t exist yet? The regulatory frameworks for broad-spectrum and pandemic-preparedness vaccines are still being worked out, and they’ll need to accommodate novel endpoints like cross-reactive antibody breadth and T-cell functionality rather than simple efficacy against a single circulating strain.

Funding is another variable. The COVID-19 pandemic unleashed unprecedented investment in vaccine technology, but that funding has already begun to recede. Operation Warp Speed and its international equivalents demonstrated what’s possible when money is essentially no object. Sustaining that level of investment for threats that are theoretical — the next pandemic coronavirus, the next novel influenza strain — requires political will that tends to evaporate once the immediate crisis passes.

Still, the momentum is real. The Coalition for Epidemic Preparedness Innovations (CEPI) has committed hundreds of millions of dollars to broad-spectrum vaccine programs. BARDA, the U.S. Biomedical Advanced Research and Development Authority, has funded multiple universal flu vaccine candidates. And private-sector players, including Moderna and BioNTech, have invested heavily in mRNA platforms that could be adapted quickly if a new threat emerges — a form of preparedness even if a single universal vaccine proves elusive.

The most realistic near-term outcome may not be a single shot that protects against everything, but rather a new generation of vaccines that are broader than what we have now. A flu vaccine that covers all Group 1 influenza viruses instead of just this season’s predicted strains. A coronavirus vaccine that protects against SARS-CoV-2 and its variants but also against SARS-CoV-1 and potentially MERS. An enterovirus vaccine that covers multiple serotypes simultaneously.

Incremental breadth, compounded over time, could transform the economics and logistics of vaccination programs worldwide. Fewer shots, broader protection, less dependence on annual reformulation. That’s not a silver bullet. But it’s a significant advance.

The question isn’t really whether broad-spectrum vaccines are possible. The animal data and early clinical results suggest they are. The question is whether the scientific community, the regulatory apparatus, and the funding environment can sustain the effort long enough to get them across the finish line — ideally before the next pandemic, not after.

History suggests we’re better at reacting than preparing. The scientists working on everything vaccines are betting that this time can be different.