Foods can be vaccines. It’s an exciting idea at a basic scientific level, but the best part is that it means vaccines can be regulated as foods. It means vaccine development can be democratized. In the inspirational words of Chef Gusteau in the movie Ratatouille:
Great cooking is not for the faint of heart. You must be imaginative! You must try things that may not work. You must not let anyone define your limits because of where you come from. What I say is true: anyone can cook. But only the fearless can be great.
The first objection I hear from some scientific colleagues comes along the lines of “you’re not fooling anyone with this food thing - we all know vaccines are drugs.” While it’s true that all previous vaccines have been marketed with the intent of preventing disease - a regulatory bright line that makes them drugs - that doesn’t mean that all future vaccines can only be considered drugs.
My colleagues are missing the point made in my carrageenan post, where I illustrate the fact that a given ingredient can simultaneously exist as a food in some products and a drug in other products. Another example is artemisinin, which is separately marketed either as an FDA-approved anti-malarial drug or as a food supplement product. We all know artemisinin is a drug, but that doesn’t mean it can’t also be a food. A third example is vitamin C - that fact that it exists as an FDA-approved injection drug used for treating scurvy doesn’t automatically turn limes and multivitamins into drugs.
The bright line is that food products only become drugs if the manufacturer makes marketing claims indicating that the product is intended to prevent or treat disease. Foods can be advertised as having general effects on normal healthy bodily structures and functions, they just can’t be advertised with the intent of treating disease.
An important point is that the food versus drug distinction is focused on manufacturer intent, as revealed by marketing claims. At the consumer level, people are free to read medical literature and choose foods they hope might treat a specific disease. If you know you have vitamin A deficiency, you’re allowed to go buy some carrots with the explicit intent of curing yourself of the disease. You could also choose to eat Golden Rice (a food product intentionally engineered to provide vitamin A). The law also allows physicians and scientists to advise patients and the general public that carrots and Golden Rice are known cures for vitamin A deficiency. At a marketing level, it’s only the manufacturer whose hands are tied.
In the Discussion section of our vaccine beer manuscript, we draw an analogy to spinach - which contains an Investigational New Drug (IND) called zeaxanthin. If you wanted to test the hypothesis that the zeaxanthin in spinach can prevent a form of blindness called macular degeneration (which extensive scientific literature suggests is possible) you would need to alert the FDA with an IND application before your spinach product could go into controlled clinical trials. The IND filing would make the specific spinach product being used in the clinical trial a not-yet-approved drug. Spinach on supermarket shelves would still just be spinach. A food manufacturer could also lawfully sell a GMO spinach strain engineered to have ten times more zeaxanthin [1], so long as the super-spinach food product doesn’t carry the marketing claim that the added zeaxanthin is intended to prevent macular degeneration. The point is that foods - including foods with ingredients known to have possible disease-fighting effects - are regulated differently than drugs.
A challenging marketing question is whether the word “vaccine,” all by itself, can only be construed as an intent to prevent or treat disease. The Merriam-Webster’s definition of the word vaccine describes a methodology that can have various possible intents:
A preparation that is administered (as by injection) to stimulate the body's immune response against a specific infectious agent or disease.
A definition from the National Cancer Institute also describes multiple possible intents:
A substance or group of substances meant to cause the immune system to respond to a tumor or to microorganisms, such as bacteria or viruses. A vaccine can help the body recognize and destroy cancer cells or microorganisms.
The word “or” in the definitions highlights the fact that infectious agents aren’t the only possible target for vaccines, and it highlights the fact that infectious agents aren’t necessarily synonymous with disease. Viruses often cause disease, but infections can also be asymptomatic. In the case of the BK polyomavirus (BKV) vaccine beer [2] I drank, my consumer intent reflected a hope for general health benefits. Although BKV infection is widely described in scientific literature as “asymptomatic” or “harmless” in healthy individuals, the virus can sometimes cause diseases of the kidneys, bladder, gut, brain, and cardiovascular system in immunosuppressed people. I also worry that BKV may be modifying normal bodily structure/function in other ways we simply haven’t figured out yet. The BKV vaccine project was essentially just a philosophical “no thanks” to BKV as a general health proposition - as opposed to a specific intent to prevent or treat any particular disease. It could be broadly described as a way of supporting the normal healthy functions of my immune system. This is a general-health structure/function claim can be found on approximately 25,000 food product labels.
In summary, having the term “BKV vaccine” on the label isn’t claiming an intent for therapeutic efficacy for any particular disease - the intent is simply to inform consumers about the presence of a specific component of the food product that could have possible health effects (including hypothetical hazards). It’s no different from telling people limes contain vitamin C.
In the vaccine beer manuscript, we report the first commercial sales food-grade vaccine-yeast products. The starting goal wasn’t to be scalable or profitable - the major initial aim is to satisfy specific aspects of food and drug law. As described in detail in Appendix B (reproduced verbatim below), if the first appearance of a substance is on the FDA IND track then it can never later be marketed as a food supplement product - it can only ever be considered a drug. The initial sales of the yeast as a food (to a set of customers with expertise in vaccine design) secures its status as a product that any company can sell as a general health food product even if, in parallel, it goes through clinical trials as a possible disease-preventing drug in the future.
When I imagine future sales to less expert customers, I start to worry that omitting the word “vaccine” from the product label could make some consumers think we’re trying to furtively sneak vaccines into the food supply. That’s the exact opposite of what we’re trying to do. The goal of this work is to promote open source vaccine development in a way that will increase everybody’s freedom of choice. Tricking people into doing things absolutely is not on the agenda.
The fact that the word vaccine is intended to describe an ingredient can be emphasized by referring to the emerging product category, in general, as “vac-foods” — where vac stands for “vaccine antigen-containing.” The possible vaccine antigen would be listed as a bioengineered component of the food product. My sense is that this compromise approach should be readily intelligible to consumers, while also lowering the hackles of scientific colleagues and regulators. We all know that familiar vaccines are drugs, but maybe we can all agree that vac-foods are foods [3].
This is a controversial new area of legal interpretation, and I’m not a lawyer. Comments - especially constructive criticism or cautions - are most welcome [4].
Footnotes
[1] GMO tomatoes engineered to produce a class of drugs called anthocyanins are marketed as a food with a range of possible general health benefits.
[2] Grapes, apples, and honey don’t have enough maltose to trigger the production of the vaccine antigen. All you beer-haters out there will have to develop your own expression systems. Or you could learn to love beer? It built the pyramids! How bad could it be.
[3] In the surprising event that courts eventually rule the category name “Vac-Yeast” an impermissible disease claim, manufacturers could resort to a category name like “Immune Support Yeast” (with the expressed antigen simply listed as a bioengineered component). I like this market approach a lot less, because it risks tricking consumers into accidentally drinking vaccines. But it’s good to know there’s technically a backup plan that only uses a permissible general health structure/function claim.
[4] In response to reader comments, I extensively revised this post several times during the last week of December 2025 and again in April 2026. Thanks, readers!
Song of the day: With apologies to Uncle Byron, who didn’t like war metaphors, I say again: Viruses. Must. Die.
Appendix B. Food structure-function studies do not require IND
This study does not require an Investigational New Drug (IND) application because it is an investigation of the effects of a food on the normal healthy structure and function of the human body, which is explicitly allowed under 21 U.S.C. § 321(g)(1)
(g)(1) The term “drug” means (A) articles recognized in the official United States Pharmacopoeia, official Homoeopathic Pharmacopoeia of the United States, or official National Formulary, or any supplement to any of them; and (B) articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals; and (C) articles (other than food) intended to affect the structure or any function of the body of man or other animals; and (D) articles intended for use as a component of any article specified in clause (A), (B), or (C). A food or dietary supplement for which a claim, subject to sections 343(r)(1)(B) and 343(r)(3) of this title or sections 343(r)(1)(B) and 343(r)(5)(D) of this title, is made in accordance with the requirements of section 343(r) of this title is not a drug solely because the label or the labeling contains such a claim. A food, dietary ingredient, or dietary supplement for which a truthful and not misleading statement is made in accordance with section 343(r)(6)* of this title is not a drug under clause (C) solely because the label or the labeling contains such a statement.
*343(r)(6) a statement for a dietary supplement may be made if the statement… characterizes the documented mechanism by which a nutrient or dietary ingredient acts to maintain such structure or function, or describes general well-being from consumption of a nutrient or dietary ingredient
The statutory definition indicates that a food can be intended to affect the structure and function of the body without being considered a drug, so long as the food is not intended to prevent or treat disease. I do not have, and I am not at risk of developing, any of the diseases BK polyomavirus is known to cause. The stated intent of the self-experiment was not to study disease, but rather to measure the effects of a food on the normal healthy structure and function of my immune system.
The scientific literature includes many examples of IRB-approved clinical investigations evaluating the effects of yeast-based food products on the structure and function of the human immune system, including measurements of virus-specific serum antibody responses (Jensen et al., 2011; McFarlin et al., 2023; Moreno et al., 2025; Roos et al., 2018). Investigations of this type are never conducted under IND because they are food studies, not drug studies.
Andrew and I are still wrestling with what I call the "only-possible-intent" problem. A funny thing about the law is that it defines drugs based on manufacturer *intent*, rather than claims. The specific language is:
"articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease"
A regulator might try to argue that the only conceivable intent for adding a vaccine antigen to yeast is the prevention and treatment of disease caused by the virus from which the antigen was derived. It's essentially the argument my scientific colleagues are making - everybody knows what I'm doing when I add a viral antigen to yeast. My intent - whether spoken or not - is obviously to prevent or treat disease.
Fortunately, a judge would laugh this argument out of court for BK polyomavirus (BKV), because the manufacturer has no idea which diseases might or might not be prevented (see footnote [1]). The BKV yeast are truly just pursuing a possible general health effect - which is allowed for foods. Unfortunately, the only-possible-intent argument starts to become more plausible for something like influenza - where there's obviously one specific disease in the crosshairs.
My sense is that legal precedents would make an only-possible-intent argument a nonstarter in court. Historically, marketing claims are the only thing regulatory agencies have ever used as evidence of manufacturer intent. In the case of Activia yogurt, a court ruled that the manufacturer couldn't have a label claim saying that eating the added Bifidobacterium animalis DN-173 010 ingredient is clinically proven to relieve temporary irregularity (i.e., constipation). Activia complied with the court ruling by changing the label to the general health claim "helps regulate your digestive system."
In the Activia case, the manufacturer essentially announced on its original label that the only-possible-intent for adding Bifidobacterium animalis DN-173 010 is relieving constipation. The point is that the court didn't attempt to psychoanalyze the manufacturer to infer an only-possible-intent - the ruling only used the text of the label claims to infer disease-treatment intent. We all know what Activia is trying to do when they add Bifidobacterium animalis DN-173 010 to the yogurt - but it doesn't matter, so long as it's not in the sales pitch.
With the Activia case in mind, it seems to me that yeast with influenza antigens would be just fine, so long as the label doesn't state an intent to foil the flu. The label could, on the other hand, lawfully have a general health claim along the lines of "promotes healthy immune function."