Stanford Medicine researchers have developed an intranasal liposomal vaccine that protected mice against SARS-CoV-2, Staphylococcus aureus, other coronaviruses, and even common allergens, using a single nasal spray formulation. The study, led by immunologist Bali Pulendran and virologist Haibo Zhang, was published in Science on February 19, 2026, and describes broad, durable protection lasting at least three months in animal models. The results challenge the conventional assumption that vaccines must be tailored to individual pathogens, raising the prospect of a genuinely “universal” respiratory defense.
How a Nasal Spray Trains the Immune System
Traditional vaccines work by teaching the body to recognize a specific virus or bacterium. This approach is effective but inherently limited. Each new pathogen requires a new shot, and novel threats can spread for months before a matched vaccine is ready. The Stanford team took a different route. Their intranasal liposomal formulation combines two classes of immune-activating molecules, a TLR4 agonist and a TLR7/8 agonist, with ovalbumin (OVA), a well-characterized egg protein used as a surrogate antigen. Rather than priming the immune system against one target, the combination appears to reprogram innate defenses in the nasal passages and lungs so they respond faster and more aggressively to a wide range of invaders.
The concept builds on years of research into “trained immunity,” a phenomenon first documented with the century-old tuberculosis vaccine BCG. Work from the Pulendran lab showed that BCG vaccination imprints prolonged innate antiviral resistance through feedback between the adaptive and innate immune systems. A related study demonstrated that BCG induces a specific population of CX3CR1hi effector memory T cells that provide cross-protection through IFN-gamma-mediated trained immunity. The new nasal formulation appears to recapitulate this axis synthetically, using carefully chosen adjuvants instead of a live bacterial vaccine to trigger a similar cascade of innate reprogramming at the mucosal surface where respiratory infections begin.
Ovalbumin Is Not Optional
One of the most striking findings is that the egg protein component is not a passive carrier. When the researchers removed ovalbumin from the formulation, immunity waned significantly. This suggests that OVA is doing more than hitchhiking along with the adjuvants. It may be actively engaging the adaptive immune system in a way that sustains the trained innate response over time, creating a feedback loop between T cells and mucosal macrophages that keeps defenses elevated for months rather than days.
This detail matters because it complicates a straightforward adjuvant-only explanation of the vaccine’s broad protection. If the formulation worked purely by stimulating innate pattern-recognition receptors, removing the protein antigen should not have degraded durability so sharply. The implication is that the vaccine generates a hybrid innate-adaptive state, one that depends on antigen-specific T cell help to maintain nonspecific barrier immunity. That finding aligns with the BCG literature but extends it: the Stanford team has shown that a defined protein antigen, not a whole live microbe, can anchor the same kind of lasting immune vigilance in the airways. It also raises practical questions for translation. Any human-ready version would need to swap OVA for a clinically acceptable protein while preserving the same immunological crosstalk.
Safety Signals From the Adjuvant Components
Any vaccine intended for widespread nasal delivery faces immediate safety questions, particularly around inflammation in sensitive mucosal tissue. The two adjuvant classes in the formulation have independent track records that offer some reassurance. The TLR7/8 agonist 3M-052 is an imidazoquinoline designed for local activity without triggering the systemic cytokine storms that have plagued earlier generations of potent immune stimulants. Its molecular design confines its effects to the tissue where it is delivered, a property that would be especially valuable in a nasal spray meant for repeated use.
The TLR4 agonist component draws on the GLA-SE adjuvant platform, which has already been evaluated in a first-in-human Phase 1 trial alongside the ID93 tuberculosis vaccine. That trial reported safety and immunogenicity data showing the adjuvant improved both the magnitude and quality of immune responses. Neither component has been tested in humans as part of the specific dual-agonist liposomal nasal formulation described in the new paper, however. The jump from individual adjuvant safety profiles to a combined intranasal product is nontrivial, and no clinical trial data yet exist for this particular combination. The NIAID adjuvant compendium catalogs these and related molecules, but standardized entries for the exact TLR4/TLR7-8-OVA nasal configuration have not been confirmed in available sources, underscoring that the current evidence base is still preclinical.
Expert Caution and the Road to Human Trials
External scientists have weighed in with a mix of enthusiasm and restraint. Immunologist Akiko Iwasaki and researcher Zhou Xing both offered expert commentary on the findings, as reported by Nature, noting that the data in mice are unusually broad but still several steps removed from clinical reality. They highlighted unanswered questions about dosing frequency, the potential for chronic inflammation in the nasal mucosa, and whether similar levels of protection can be achieved in larger animals whose respiratory anatomy more closely resembles that of humans. The experts also pointed out that mouse models of allergy and bacterial infection often overestimate vaccine performance when translated to people.
For regulators and clinicians, the path to human trials is likely to proceed in stages. Before a universal-style nasal spray could be tested in healthy volunteers, researchers would need detailed toxicology studies in multiple animal species to look for subtle signs of tissue damage or autoimmunity. Early-phase clinical trials would probably focus on safety and local tolerability, using low doses and short follow-up intervals, with exploratory endpoints such as changes in nasal immune cell profiles or temporary protection against attenuated challenge viruses. Only after those hurdles are cleared could larger trials assess whether trained mucosal immunity translates into fewer respiratory infections or milder disease in real-world settings. The Stanford results do not answer those questions yet, but they provide a mechanistic blueprint that future studies can adapt and test.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
