How Vaccines Are Developed and Tested: From Lab to Approval

What Vaccines Are Trying to Do

The immune system has a remarkable feature: it remembers pathogens it has encountered before and responds far faster and more powerfully upon re-exposure. This immunological memory is the biological foundation on which all vaccines rest. A vaccine introduces the immune system to something derived from — or resembling — a pathogen, training it to mount a defense without the danger of actual infection. When the real pathogen later appears, the pre-trained immune system eliminates it before disease can develop or, in cases of partial protection, before severe illness occurs.

The practical challenge is choosing what to introduce. The immune system can learn from whole killed pathogens, from live attenuated (weakened) versions, from isolated protein subunits, from polysaccharide coats, or from genetic instructions that cause the body's own cells to produce a harmless antigen. Each approach has distinct advantages in terms of immune response quality, manufacturing complexity, storage requirements, and safety profile. Selecting the right platform is one of the earliest and most consequential decisions in vaccine development.

Vaccines also need adjuvants — substances that amplify the immune response. Aluminum salts (alum) have been used for nearly a century to boost antibody production. Newer adjuvants like AS04 and MF59 stimulate additional innate immune pathways, generating stronger and more durable protection. The combination of the right antigen with the right adjuvant can mean the difference between a vaccine that provides years of protection and one that requires frequent boosting.

Discovery and Antigen Identification

Vaccine development begins in research laboratories long before any human receives an injection. Scientists first study the pathogen in detail: How does it infect cells? Which proteins on its surface are recognized by antibodies? Which proteins are essential for its survival and thus unlikely to mutate away from vaccine pressure? These questions are answered through structural biology, genomics, proteomics, and experiments in cell culture systems.

For bacterial vaccines, polysaccharides — complex sugars on the bacterial surface — are often excellent antigens. For viruses, surface proteins that mediate cell entry are prime targets because antibodies that bind these proteins can physically prevent infection. The spike protein of SARS-CoV-2 is a canonical example: COVID-19 vaccines from multiple platforms all targeted this single protein because it was essential, exposed, and antibody-accessible. Once candidate antigens are identified, researchers design vaccine constructs and test them in cell culture to confirm that immune cells recognize and respond to them appropriately before moving into animal studies.

Preclinical Studies: Animal Trials and Safety Screens

Preclinical studies serve two main purposes: demonstrating that a vaccine candidate generates the desired immune response and identifying safety signals that would make human testing unjustifiable. Researchers test candidates in mice first — their small size, short lifespan, and the availability of genetic tools make them invaluable for early screening. Promising candidates then move to non-human primates, whose immune systems more closely resemble humans, or to other relevant animal models depending on the pathogen.

In these animal studies, researchers measure antibody titers, the types of antibodies produced, and the cellular immune response — particularly the activation of T cells, which are critical for clearing infected cells and for generating long-lived memory. For pathogens where an animal model of disease exists, challenge studies expose vaccinated animals to the actual pathogen to directly measure protective efficacy. Safety screening examines toxicity, local injection-site reactions, effects on reproduction and development, and any evidence of immune pathology such as vaccine-enhanced disease. A vaccine that passes preclinical screening with a compelling immune response and acceptable safety profile advances to an Investigational New Drug (IND) application submitted to regulatory agencies, allowing controlled human testing to begin.

Phase 1 and Phase 2 Clinical Trials

Human clinical trials are divided into three phases, each with a different primary purpose and escalating in scale. Phase 1 trials enroll a small number of healthy adults — typically 20 to 100 — and focus on safety and dosing. Researchers test different dose levels, examine local reactions at the injection site, monitor for systemic symptoms such as fever and fatigue, assess laboratory values for signs of immune activation or organ stress, and measure the basic immunogenicity of the vaccine. Phase 1 rarely answers questions about efficacy; it establishes a safety foundation and identifies the dose or doses to carry forward.

Phase 2 trials expand to hundreds of participants and begin to include populations of interest: the elderly, children, or immunocompromised individuals if they are the target of vaccination. Phase 2 refines dosing, assesses immunogenicity in greater detail — comparing the vaccine's antibody responses to those seen in people who recovered naturally from infection — and continues safety monitoring in a larger, more diverse group. Statistical power is still insufficient to measure protection against disease directly, but Phase 2 data inform the design of the pivotal Phase 3 trial.

Phase 3 Trials: Measuring Efficacy

Phase 3 trials are randomized controlled trials enrolling thousands to tens of thousands of participants, typically comparing the vaccine to a placebo under double-blind conditions. Participants are followed for months to years, during which researchers count cases of the disease occurring in vaccinated versus unvaccinated groups to calculate vaccine efficacy — the percentage reduction in disease incidence attributable to vaccination. A VE of 90 percent means vaccinated individuals are 90 percent less likely to develop the disease than unvaccinated controls. Regulatory agencies typically require at least 50 percent efficacy for licensure, along with a lower bound of the confidence interval comfortably above 30 percent, to guard against statistical uncertainty.

Safety monitoring in Phase 3 can detect adverse events occurring in as few as one per ten thousand recipients. The COVID-19 vaccine trials of 2020 achieved remarkably rapid Phase 3 completion through massive simultaneous investment across multiple candidates, pre-built clinical trial infrastructure, and extraordinarily high disease prevalence that allowed efficacy signals to emerge faster than in normal circumstances — demonstrating that the timeline can be dramatically compressed without sacrificing scientific rigor under extraordinary conditions.

Regulatory Review and Approval

After Phase 3 completes, the developer submits a Biologics License Application (BLA) or equivalent package to regulatory authorities. This submission contains every piece of data generated during development: preclinical studies, all three clinical phases, manufacturing process descriptions, quality control data, and proposed labeling. Regulatory reviewers — scientists, physicians, statisticians, and manufacturing experts — conduct an independent analysis of all submitted data, often requesting additional analyses from the developer.

Regulatory agencies also conduct facility inspections to verify that manufacturing processes meet Good Manufacturing Practice (GMP) standards. Advisory committees of independent experts typically convene publicly to review the data and vote on whether the evidence supports approval, though their recommendations are advisory rather than binding. The entire review process for a standard vaccine commonly takes 6 to 12 months. Emergency Use Authorization pathways allow earlier authorization based on promising but still accumulating data during public health emergencies, with full approval following as the dataset matures. Different agencies — the FDA, EMA, Health Canada, and MHRA — each conduct independent reviews, though they increasingly cooperate through information-sharing arrangements.

Manufacturing, Cold Chain, and Post-Market Surveillance

Regulatory approval authorizes a vaccine but does not deliver it. Manufacturing at the scale required for national or global vaccination campaigns is a massive logistical and engineering undertaking. For protein subunit vaccines, large bioreactors grow cells expressing the antigen protein; for mRNA vaccines, enzymatic synthesis produces RNA strands and lipid nanoparticle machines package them; for live attenuated viral vaccines, eggs or cell culture systems propagate the virus. Most vaccines require refrigeration at 2 to 8 degrees Celsius throughout storage and distribution. Some mRNA vaccines originally required ultra-cold storage, spurring major formulation advances in thermostability through freeze-drying and novel lipid compositions.

Phase 4 post-market surveillance continues after approval, monitoring for rare adverse events that Phase 3 trials were too small to detect, assessing effectiveness in real-world conditions, and studying long-term protection. Surveillance systems such as the U.S. Vaccine Adverse Event Reporting System collect reports that, analyzed with sophisticated statistical methods, can detect genuine safety signals against the background of illness that would occur in any large population regardless of vaccination. The entire journey from pathogen identification to widespread distribution typically takes 10 to 15 years and costs hundreds of millions to over a billion dollars — a timeline and cost structure that explains why accelerated development during emergencies requires unprecedented coordinated investment from governments, industry, and multilateral institutions.