How Vaccines Are Tested: From Lab Work to FDA Approval
Each stage of vaccine testing answers a different question, from basic safety in a few dozen volunteers to rare side effects that only show up after millions of doses.
A new vaccine is tested in stages: lab and animal studies first, then three phases of human trials that grow from about 20 to 100 volunteers in Phase 1 to thousands in Phase 3, according to the US Food and Drug Administration (FDA). If the data hold up, the FDA reviews the results and inspects the factory before licensing the vaccine, and safety monitoring continues after approval.
Each step answers a narrower question than the one before, from basic safety to the right dose to whether the vaccine prevents disease in large numbers of people. This article explains the process. Whether a particular vaccine suits you is a question for your own clinician.
What happens before a vaccine is tried in people?
Before any person receives a candidate vaccine, it goes through preclinical testing in the lab and in animals. These studies show how the vaccine provokes an immune response and whether it is likely to be safe enough to try in humans.
The developer then files an Investigational New Drug application, or IND, with the FDA. According to the FDA’s Vaccine Development 101 page, the IND contains the preclinical results plus details of how the vaccine is made and quality-checked, and the agency looks at whether the lab work followed Good Laboratory Practices. Human trials start only after this step.
Many candidates that look promising in animals fail in people, which is why the clinical phases exist. For what a vaccine is trying to train, see how the immune system works.
What are Phase 1, 2 and 3 trials?
The three clinical phases test the vaccine in progressively larger groups, and each phase has its own main question. The sizes below are the FDA’s general descriptions; individual trials can be smaller or larger.
| Phase | Typical size (FDA) | Main question |
|---|---|---|
| Phase 1 | About 20 to 100 healthy volunteers | Is it safe, and what reactions come with each dose level? |
| Phase 2 | Hundreds of people, in randomized studies with a control group | Which dose and schedule work, how strong is the immune response, and is there early evidence it protects? |
| Phase 3 | Thousands of people | Does it prevent disease compared with a control group, and are there less common side effects? |
| Phase 4 (after approval) | Varies | Studies the FDA may require to look at known or possible serious risks |
Phase 3 is the expensive part. Researchers need enough people, and enough cases of the disease in the control group, to see a difference. For a common infection that can happen in a few months. For a rare disease it can take years, or a much bigger trial. The 2020 Phase 3 trial of the Pfizer-BioNTech COVID-19 vaccine, for example, randomized 43,548 people, according to the results published in the New England Journal of Medicine.
Why do trials use placebos and blinding?
Placebos and blinding exist so that the only real difference between the groups is the vaccine itself. In a randomized trial, chance decides who gets the vaccine and who gets the control, so age, health and behavior end up spread evenly between the groups.
The control is usually a placebo injection, though some trials compare against a different, already licensed vaccine. In a double-blind trial neither the volunteers nor the staff who assess them know who received what. Otherwise expectations could color how symptoms are reported and how borderline cases are judged.
Volunteers in both groups record reactions such as a sore arm, tiredness or headache. Comparing the two groups shows which reactions come from the vaccine and which are just the ordinary aches people get anyway. Short-lived fever after a dose is the immune system at work, which we explain in why we get a fever.
How is vaccine efficacy calculated?
Vaccine efficacy is the percentage by which the vaccine cut the rate of disease in the vaccinated group compared with the placebo group during the trial. The basic formula is simple:
Efficacy = (rate in placebo group minus rate in vaccine group) divided by rate in placebo group
In the COVID-19 trial above, 21,720 people received the vaccine and 21,728 the placebo. Counting from 7 days after the second dose, 8 cases of COVID-19 occurred in the vaccine group and 162 in the placebo group. The groups were almost the same size, so the calculation comes out close to 1 minus 8/162, or about 95%. The paper reported 95% efficacy with a credible interval of 90.3% to 97.6%.
A 95% efficacy does not mean that 5% of vaccinated people will get sick. It means vaccinated volunteers had about 95% fewer cases than unvaccinated ones over the period studied. Efficacy is measured in a trial. Effectiveness is the same idea measured later in the real world, in observational studies of people who did or did not get the vaccine, and it can differ because real populations, variants and timing differ from trial conditions.
What does the FDA check before approval?
To market a vaccine in the US, the maker files a Biologics License Application, or BLA. The FDA says it includes the preclinical and clinical data plus a description of the manufacturing process and facility. FDA investigators also inspect the factory and its operations.
In some cases the FDA asks its Vaccines and Related Biological Products Advisory Committee, an independent panel of outside experts, to discuss the data at a public meeting. The FDA considers that advice but is not bound by it. After licensing, the maker must submit test results and samples for each lot, and the FDA says a lot cannot be distributed until the agency releases it. A separate body, the CDC’s Advisory Committee on Immunization Practices, advises on how a licensed vaccine should be used.
How are vaccines watched after approval?
Safety monitoring continues for as long as a vaccine is in use, because even a trial of tens of thousands of people can miss a side effect that strikes 1 in several hundred thousand. Two US systems do most of this work:
- VAERS, the Vaccine Adverse Event Reporting System, is co-managed by the CDC and FDA and has run since 1990. Anyone can file a report on the VAERS website. A report means something happened after vaccination, not that the vaccine caused it, so VAERS works as an early-warning system that flags patterns for study.
- The Vaccine Safety Datalink, set up by the CDC in 1990 with large health care organizations, uses electronic health records to check whether a health problem occurs more often after vaccination than in a comparison group. It can test whether a signal from VAERS is real.
If a real risk turns up, regulators can change the label or limit use, and the FDA can require Phase 4 studies.
Bottom line
Vaccines move from lab and animal work to Phase 1 trials of about 20 to 100 people, Phase 2 trials of hundreds and Phase 3 trials of thousands, with placebo groups and blinding to keep the comparison fair. Efficacy is the drop in disease in the vaccine group compared with the placebo group. The FDA reviews the data and the factory before licensing, and systems such as VAERS and the Vaccine Safety Datalink keep watching for rare problems afterward.
FAQ
How long does it take to test a vaccine?
It usually takes years, depending on how common the disease is and how fast trials can enroll volunteers. Timelines can shrink when phases overlap and when a disease is spreading widely, because the cases needed to measure efficacy accumulate faster.
What is the difference between efficacy and effectiveness?
Efficacy is measured in a controlled trial with a placebo group. Effectiveness is measured after approval, in real-world observational studies, and it can be higher or lower than the trial result.
Does a VAERS report prove that a vaccine caused a problem?
No. VAERS collects reports of health events that happen after vaccination, whether or not the vaccine was the cause. Researchers use those reports to spot unusual patterns and then test them with other data, such as the Vaccine Safety Datalink.
Who decides whether a vaccine is approved in the US?
The FDA decides whether to license a vaccine, sometimes after hearing from its outside advisory committee. Questions about whether a specific vaccine is right for you are best taken to a doctor or other clinician who knows your health history.
