Two and a quarter centuries of teaching the immune system to remember an enemy it has never met.
A vaccine is a controlled rehearsal. The body meets a defanged version of a pathogen, mounts an immune response, and files the memory away. When the real thing arrives, the response is already drafted.
The premise extends to populations. Above a threshold of vaccinated individuals — the figure varies by pathogen, roughly 95 percent for measles, 80 percent for polio — the chain of transmission breaks. Unvaccinated neighbours, infants too young to be immunised, the immunocompromised, the elderly: all are protected by the wall of immunity around them. Herd immunity is the public-health argument that turned vaccination from a personal choice into a civic question.
This deck covers vaccines from the variolation practices of 18th-century Constantinople through the mRNA platforms that compressed pandemic response from years to months. Thirty leaves; real names, real dates, real numbers.
Before vaccination there was variolation: deliberately infecting a healthy person with a small dose of smallpox material, usually scratched into the skin, to provoke a milder case and lifelong immunity. Practised in China by the 16th century, in Ottoman Anatolia and West Africa earlier still.
Lady Mary Wortley Montagu, wife of the British ambassador to Constantinople, observed Turkish women conducting variolation parties in 1717. She had her own son inoculated; on her return to London she had her daughter publicly variolated in 1721 and lobbied the Princess of Wales to do the same. Six condemned prisoners and eleven orphans were variolated in royal trials; all survived; the practice spread among the British aristocracy.
Variolation was risky — roughly one in fifty died — but smallpox itself killed one in three. Cotton Mather learned of the practice from his enslaved African Onesimus and pushed Boston physician Zabdiel Boylston to variolate during the 1721 epidemic. Six of 240 variolated patients died; one in seven of the unvaccinated did.
Gloucestershire dairymaids who had contracted cowpox — a mild bovine disease that produced pustules on the hands of those who milked infected cows — appeared not to catch smallpox. The folklore was current in the English countryside long before Edward Jenner (1749–1823) tested it.
On 14 May 1796 Jenner inoculated eight-year-old James Phipps with material from a cowpox lesion on the hand of milkmaid Sarah Nelmes. Six weeks later he challenged the boy with smallpox matter. Phipps did not develop the disease. Jenner repeated the experiment, published An Inquiry into the Causes and Effects of the Variolae Vaccinae in 1798, and named the procedure vaccination from the Latin vacca, cow.
The technique spread within years. By 1801 some 100,000 people had been vaccinated worldwide. The British Parliament awarded Jenner £30,000. Variolation was banned in Britain in 1840.
Jenner's smallpox vaccine was a happy accident — a pre-existing animal pathogen that happened to immunise against a human one. Louis Pasteur (1822–1895) generalised the principle. By weakening pathogens through laboratory passage, he produced vaccines for chicken cholera (1879), anthrax (1881, demonstrated dramatically at Pouilly-le-Fort on 24 sheep), and rabies.
The rabies vaccine was tested on a human for the first time on 6 July 1885: nine-year-old Joseph Meister, mauled by a rabid dog two days earlier and otherwise certain to die. Pasteur, not a physician, administered thirteen injections of progressively stronger preparations of dried rabbit spinal cord. Meister survived. The story made global news; donations from across Europe funded the Institut Pasteur, opened in Paris in 1888.
Pasteur kept Jenner's word, calling the new preparations vaccins in honour of him.
The bacterial diseases of childhood became preventable in succession. Diphtheria antitoxin (Emil von Behring and Shibasaburo Kitasato, 1890) — the first true biologic — preceded the diphtheria toxoid vaccine developed by Gaston Ramon at the Pasteur Institute in 1923. Ramon's insight, that formaldehyde-inactivated toxin retained immunogenicity without toxicity, became a template.
Tetanus toxoid followed (Ramon and Pierre Descombey, 1924). Universally given to wounded soldiers from World War II onward, it eliminated tetanus as a battlefield killer.
Pertussis — whooping cough — was tamed by Pearl Kendrick and Grace Eldering in Grand Rapids, Michigan, in the 1930s. The whole-cell vaccine was combined with diphtheria and tetanus toxoids to produce DTP in 1948. The acellular version (DTaP), with fewer side effects, replaced it from 1991.
Polio epidemics terrified mid-century America. The 1952 epidemic infected 58,000 and killed 3,145; survivors filled wards of iron lungs. The March of Dimes, founded by Franklin Roosevelt — himself paralysed by what was diagnosed as polio in 1921 — funded the research.
Jonas Salk's inactivated polio vaccine (IPV) was tested in the largest medical trial in history: 1.8 million American schoolchildren in 1954. On 12 April 1955 the results were announced and the vaccine licensed within hours. Albert Sabin's oral attenuated vaccine (OPV), easier to administer, followed in 1961 and became the WHO's primary tool for global eradication.
Salk refused to patent the vaccine, telling Edward R. Murrow: "Could you patent the sun?" Cases fell from 350,000 globally in 1988 to 6 wild cases in 2021. Two of three wild polio strains have been eradicated; only Pakistan and Afghanistan still report endemic transmission of the third.
Before the vaccine, measles infected nearly every child on earth. In the United States it caused 3–4 million cases and 400–500 deaths a year. Globally it killed about 2.6 million children annually as late as 1980.
John Enders and Thomas Peebles isolated the measles virus from a sick boy named David Edmonston in 1954. Enders, who had won a Nobel for tissue-culture work in 1954, attenuated the virus through serial passage; the licensed vaccine arrived in 1963 (further attenuated to the Schwarz strain by 1965 and the Moraten strain by 1968).
Measles is the most contagious infectious disease known, with an R-naught of 12–18. The herd-immunity threshold is correspondingly high: about 95 percent. The 2019 global resurgence — 869,000 cases, 207,000 deaths, the highest since 1996 — tracked declining vaccination coverage in conflict zones and pockets of vaccine refusal in wealthy countries.
Three live attenuated viruses combined in one injection. Maurice Hilleman (1919–2005) — the most prolific vaccine developer in history, credited with eight of the fourteen vaccines on the routine US childhood schedule — assembled the components.
Measles: Enders's strain. Mumps: Hilleman swabbed his five-year-old daughter Jeryl Lynn's throat in 1963 when she developed the disease; the Jeryl Lynn strain remains the global standard. Rubella: the Wistar Institute's RA 27/3 strain (1969). The combined MMR was licensed in 1971.
Rubella in particular is a disease whose human cost was almost entirely paid by unborn children. The 1964–65 US rubella epidemic produced an estimated 20,000 cases of congenital rubella syndrome — heart defects, cataracts, deafness, intellectual disability. Rubella vaccination eliminated endemic transmission in the Americas by 2015.
The last case of naturally occurring smallpox was a Somali cook named Ali Maow Maalin, in Merca, on 26 October 1977. The World Health Organization certified eradication on 9 December 1979; the World Health Assembly endorsed it on 8 May 1980.
The campaign began in 1967 under Donald Henderson at the WHO. It rejected the assumption that mass vaccination was the only path; instead the strategy was ring vaccination — rapidly identifying every case, vaccinating every contact, and choking off transmission. The technique exploited smallpox's two-week incubation and its inability to survive long outside a host.
The disease had killed an estimated 300 million people in the 20th century alone before its 1980 extinction. Two laboratory stocks remain — at the CDC in Atlanta and at VECTOR in Russia — and the question of whether to destroy them has been debated, deferred, and re-debated by the World Health Assembly every few years since 1986.
Polysaccharide capsules — the sugar coats of certain bacteria — provoke a weak immune response in young children. The trick of chemically linking the polysaccharide to a protein carrier produces a vaccine that an infant immune system can recognise robustly. The technique was John Robbins and Rachel Schneerson at the NIH in the 1980s.
Hib (Haemophilus influenzae type b) — the leading cause of bacterial meningitis in children before 1990. The conjugate vaccine, licensed in 1987 and refined by 1990, eliminated Hib disease in vaccinated populations within five years.
Pneumococcal conjugate (Prevnar, 2000; PCV13, 2010; PCV20, 2021) prevents Streptococcus pneumoniae infection — pneumonia, meningitis, otitis media. Meningococcal conjugate (against Neisseria meningitidis) addresses the rare but devastating bacterial meningitis of adolescents. All three have shifted childhood mortality globally.
The first vaccine ever to prevent a cancer. Harald zur Hausen proposed in the 1970s that human papillomavirus caused cervical cancer; he isolated HPV-16 in 1983 and HPV-18 the next year. He won the 2008 Nobel.
Ian Frazer and Jian Zhou in Brisbane developed the virus-like particle technology that became the platform for the vaccines. Gardasil (Merck, 2006) initially covered HPV-16 and 18 plus the wart-causing 6 and 11; Gardasil 9 (2014) covers nine strains. Cervarix (GSK, 2007) followed.
The data are now beginning to come in. A 2020 Swedish study of 1.7 million women found a 63 percent reduction in invasive cervical cancer among those vaccinated; for those vaccinated before age 17, an 88 percent reduction. Australia, where HPV vaccination has been universal since 2007, is on track to be the first country to eliminate cervical cancer by 2035.
Live attenuated. The pathogen, weakened so it cannot cause disease but can replicate enough to provoke immunity. MMR, varicella, yellow fever, oral polio. Strongest, longest-lasting immunity; cannot be given to immunocompromised patients.
Inactivated. The pathogen, killed. Salk polio, hepatitis A, rabies, most flu vaccines. Safer for immunocompromised; usually requires boosters.
Subunit, recombinant, polysaccharide, conjugate. Pieces of the pathogen — a surface protein, a capsule sugar — produced in yeast, bacteria, or insect cells. Hepatitis B (the first recombinant vaccine, 1986), HPV, pertussis acellular, Hib.
Toxoid. Inactivated bacterial toxin. Tetanus, diphtheria.
Viral vector. A harmless virus carries the pathogen's gene into cells, which then display the antigen. Ebola (rVSV-ZEBOV, 2019), AstraZeneca and Johnson & Johnson COVID-19 vaccines.
mRNA. Instructions, encapsulated in lipid, that the body's own cells translate into the antigen. Pfizer-BioNTech and Moderna COVID-19. The newest platform, with the broadest implications.
Most modern vaccines need help. An adjuvant is a substance that amplifies the immune response to the antigen — it does not provide immunity itself but ensures the body takes the antigen seriously enough to remember it.
Aluminum salts (alum) have been used since 1926. They remain the most common adjuvant by volume in the global vaccine supply. The mechanism — depot effect, immune-cell recruitment, activation of the NLRP3 inflammasome — was only fully elucidated in the 2000s.
Newer adjuvants include MF59 (squalene-based, in the Fluad influenza vaccine), AS04 (in Cervarix and Fendrix), AS01 (Shingrix, RTS,S), and the lipid nanoparticles of mRNA vaccines, which serve both as delivery vehicle and adjuvant.
The cultural reading of adjuvants — as foreign agents introduced into a "natural" immune response — has made them a focus of vaccine-hesitant suspicion. The biological reading is that they are what allow non-replicating vaccines to work.
A vaccine that has been allowed to warm too far is no longer a vaccine. The cold chain — the unbroken sequence of refrigeration from manufacturing line to upper-arm injection — is the largest logistical undertaking in global health.
Most vaccines require 2–8°C. The oral polio vaccine is the fragile end; varicella and zoster need −15°C. The Pfizer-BioNTech COVID-19 vaccine famously required −70°C, which prompted the procurement of ultra-cold freezers, dry-ice supply chains, and time-temperature indicators worldwide in 2020–21.
The WHO's Performance, Quality, Safety prequalification programme certifies cold-chain equipment used in low-income countries. Solar refrigerators, vaccine carriers with phase-change material, and electronic temperature loggers have been transformative for last-mile delivery in Africa and South Asia. Gavi and UNICEF estimate that ~30 percent of vaccines were spoiled in transit before cold-chain reform; the figure is now under 5 percent in most middle-income countries.
The Wuhan SARS-CoV-2 sequence was published online by Yong-Zhen Zhang on 11 January 2020. By 13 January, Moderna had finalised the mRNA sequence for what would become its vaccine. Two days. The compression of vaccine development from a process measured in years to one measured in months is the central biomedical event of the early 21st century.
BioNTech (Mainz, Germany) — founded by Uğur Şahin and Özlem Türeci — partnered with Pfizer in March 2020. Phase III enrolment of 43,548 participants began on 27 July; the primary efficacy readout, on 8 November, showed 95 percent efficacy. Emergency authorisation followed in the UK on 2 December and the US on 11 December 2020.
Moderna's mRNA-1273 followed by a week. By the end of 2022 over 13 billion doses of COVID-19 vaccine had been administered globally; an estimated 14.4 million deaths were prevented in the first year alone (Lancet, 2022).
Katalin Karikó spent decades unable to win major grants. A Hungarian biochemist who emigrated to the US in 1985, she was demoted at the University of Pennsylvania in 1995 for failing to attract funding for her work on therapeutic mRNA. She kept going.
The breakthrough — co-authored with the immunologist Drew Weissman, whom she met at a Penn photocopier in 1997 — came in 2005. They showed that substituting the nucleoside pseudouridine for uridine in synthetic mRNA evaded innate immune detection while preserving translation. The modification turned mRNA from an inflammatory misfire into a viable therapeutic platform.
Karikó moved to BioNTech in 2013 as senior vice-president. The 2005 paper underwrote both Pfizer-BioNTech and Moderna COVID-19 vaccines. Karikó and Weissman shared the 2023 Nobel Prize in Physiology or Medicine.
Andrew Wakefield's 1998 paper in The Lancet, claiming a link between MMR vaccination and autism, was based on twelve children. It contained fabricated data and undisclosed conflicts of interest — Wakefield had been paid £435,000 by lawyers preparing to sue vaccine manufacturers.
The journalist Brian Deer, in a series of Sunday Times investigations from 2004 onwards, demonstrated that Wakefield had altered the medical histories of all twelve children. The General Medical Council struck him off the medical register in May 2010. The Lancet retracted the paper the same year.
The damage outlasted the retraction. UK MMR coverage, 92 percent in 1996, fell to 80 percent by 2003. Measles, declared eliminated in the UK in 2017, has since returned. The largest peer-reviewed studies — a 2019 Danish cohort of 657,461 children, a 2014 meta-analysis of 1.26 million — find no association whatsoever between MMR and autism.
The WHO's 2019 list of top ten threats to global health included, for the first time, vaccine hesitancy. The term covers a continuum: outright refusers (a small minority), the hesitant who delay or pick selectively, and the simply absent — those whose hesitance is access, not belief.
The Vaccine Confidence Project, led by Heidi Larson at LSHTM, has surveyed confidence in 149 countries since 2015. France ranks consistently among the lowest in trust globally. Bangladesh and India among the highest. Confidence is not strongly correlated with national wealth.
The literature is clear that mandates work where information campaigns do not. California's 2015 SB277, removing personal-belief exemptions for school entry, raised kindergarten MMR coverage from 90 to 95 percent. France's 2018 expansion of the mandate from three to eleven required vaccines for children under two raised coverage of all eight added vaccines above 90 percent within two years.
The US childhood schedule (CDC/ACIP) protects against 14 diseases by age two. Birth: hepatitis B. Two months: rotavirus, DTaP, Hib, PCV13, IPV. Four and six months: repeats. Twelve months: MMR, varicella, hepatitis A. Annual influenza from six months. HPV at 11. Meningococcal at 11 and 16. Tdap booster at 11.
The UK schedule diverges slightly — no varicella as routine, no hepatitis A or B except for risk groups. Most European countries cover similar ground. The WHO's Expanded Programme on Immunization recommends a minimum schedule for low-income countries: BCG, hepatitis B, polio, DTP, Hib, pneumococcal, rotavirus, measles, rubella, HPV.
The schedule is the most carefully designed product of public health. Each interval is a compromise between maternal-antibody decay, infant immune-system maturation, age-specific disease risk, and parental tolerance for needles. The arguments about whether it is "too crowded" are not arguments physicians or epidemiologists are having.
The traditional vaccine prevents. A therapeutic vaccine treats — typically by training the immune system against an existing tumour or chronic infection.
Sipuleucel-T (Provenge, 2010) was the first FDA-approved therapeutic cancer vaccine, for advanced prostate cancer. Modest survival benefit; expensive; not widely adopted. Talimogene laherparepvec (T-VEC, 2015) for melanoma uses an engineered herpes virus.
The mRNA platform has reopened the field. Moderna and Merck's mRNA-4157/V940 — a personalised neoantigen vaccine for melanoma — encodes 34 peptides specific to a patient's own tumour. The 2024 Phase IIb readout showed a 49 percent reduction in recurrence or death versus checkpoint inhibitor alone. BioNTech's parallel pancreatic cancer trial (autogene cevumeran) showed eight of sixteen patients still recurrence-free at three years — a remarkable signal in a cancer with a 12 percent five-year survival rate.
Malaria has resisted vaccination for a century. The parasite — Plasmodium falciparum chiefly — has thousands of antigens, three life-cycle stages in the human host, and the ability to vary its surface proteins. It kills around 600,000 people a year, mostly African children under five.
RTS,S/AS01 (Mosquirix, GSK), the result of 30 years of work led by Joe Cohen at GSK, was approved by the WHO for childhood use in October 2021 — the first malaria vaccine, and the first vaccine ever for any human parasitic disease. Efficacy is partial: about 40 percent against clinical malaria over four years.
R21/Matrix-M (University of Oxford, Serum Institute of India), recommended by the WHO in October 2023, achieves 75 percent efficacy in seasonal settings and is producible at $4 a dose at the Serum Institute's manufacturing scale. Rollout across 18 African countries began in 2024.
The Expanded Programme on Immunization, launched at the World Health Assembly in May 1974, is the global public-health success of the past half-century. At launch, fewer than 5 percent of children in low-income countries were vaccinated against the original six EPI diseases (TB, polio, diphtheria, tetanus, pertussis, measles).
By 2019, coverage of the third diphtheria-tetanus-pertussis dose (DTP3, the standard global immunisation indicator) had reached 86 percent. The COVID-19 pandemic dropped it briefly to 81 percent — the largest sustained decline in 30 years; recovery to 84 percent was reported by 2023.
The programme now covers 14 diseases. Gavi, the Vaccine Alliance, founded in 2000 with seed funding from the Gates Foundation, finances vaccine procurement for 78 of the world's poorest countries. Gavi-supported immunisation has prevented an estimated 17.3 million deaths since 2000.
Real adverse events exist and are tracked. Vaccine-Associated Paralytic Polio from oral polio vaccine: ~1 in 2.7 million doses; the reason high-income countries switched to IPV in the 1990s. Anaphylaxis from any injected vaccine: ~1 in a million doses. Myocarditis from mRNA COVID-19 vaccines, especially in young men after second dose: ~1–2 per 100,000, almost always mild and self-limiting; the rate from COVID-19 infection itself is several times higher. Guillain-Barré syndrome from the 1976 swine-flu vaccine: ~1 in 100,000.
The US Vaccine Adverse Event Reporting System (VAERS), the WHO Global Vaccine Safety database, and the EU's EudraVigilance log adverse events for population-level analysis. The signals that emerge are real; the signals that are claimed (autism from MMR, infertility from HPV) are not detected in the data.
The National Vaccine Injury Compensation Program (US, 1986) provides no-fault compensation for proven vaccine injuries. It has paid roughly $5 billion across 9,500 claims since 1989, against 4–5 billion vaccine doses administered.
What is in late-stage trials, as of mid-decade.
Universal influenza. mRNA candidates encoding multiple haemagglutinin subtypes; the goal is a single annual jab that covers strains beyond the matched WHO forecast. Moderna's mRNA-1010 Phase III readouts began in 2024.
RSV. The 2023 approvals of Arexvy (GSK) and Abrysvo (Pfizer) for older adults, and the maternal RSV vaccine for newborn protection, ended a half-century gap since the disastrous 1966 RSV vaccine trials.
HIV. The Moderna-IAVI mRNA candidate (mRNA-1644) entered Phase I in 2022, using germline-targeting immunogens designed by Scripps. The path to broadly neutralising antibodies remains long.
Tuberculosis. M72/AS01E (GSK and Gates MRI) showed 50 percent efficacy in Phase IIb (2018); the Phase III is enrolling 20,000 participants across seven countries. A modern TB vaccine would be the most consequential single vaccine of the decade.
↑ A very short history of vaccines in America · FRONTLINE
Watch · mRNA vaccines, explained
Watch · The history of vaccination · Smallpox
Start with Oshinsky's Polio: An American Story (2011, Pulitzer) for the texture of mid-century American epidemic medicine. Eula Biss's On Immunity (2014) is the literary essay on parental hesitancy. For the science, Plotkin's Vaccines is the working textbook.
Vaccines work only when they reach the people who need them. The COVID-19 rollout exposed the fault lines. By the end of 2021 high-income countries had administered more than 150 doses per 100 people; sub-Saharan Africa, fewer than 10. COVAX, the WHO-led global allocation mechanism, was outbid in the 2020 procurement market by direct bilateral deals between manufacturers and rich-country governments.
The TRIPS waiver debate at the WTO ran for two years and produced a narrow, late, and largely ineffective patent flexibility in 2022. The argument continues over whether intellectual property is the binding constraint or whether technology transfer, manufacturing capacity, and cold-chain build-out are.
Africa has begun building its own vaccine manufacturing. The WHO mRNA technology-transfer hub in Cape Town (Afrigen, 2021) is producing a COVID-19 candidate without Pfizer or Moderna's cooperation; spinoff hubs are operating in Senegal, Egypt, Kenya, Nigeria, Rwanda, South Africa, and Tunisia.
Vaccination is the most cost-effective intervention in modern medicine. The CDC's analysis of the US childhood schedule for the 2009 cohort estimated a $13.5 billion direct cost saving and $68.8 billion in societal savings, against a vaccination cost of $7.5 billion. Each dollar spent returns ten.
It is also the most quietly successful. Diseases that vaccinated populations no longer remember — diphtheria choking children, polio paralysing them, congenital rubella robbing them of sight before they were born, measles killing them before they finished the year — are the success the discipline rarely receives credit for. The argument that vaccines are unnecessary is most loudly made in the populations they have made the safest.
The next decade's frontier is therapeutic — cancer, autoimmune disease, perhaps even allergy. The mRNA platform is the through-line. The 250-year arc from Jenner's eight-year-old patient to a cancer vaccine designed for a single tumour is short, by the timescale of medicine, and not yet finished.
Vaccines — Volume VI, Health, of The Deck Catalog. Set in IBM Plex Sans and Tiempos Text. Paper at #f5f3ed; ink in deep navy; accent in warm clinical orange.
Thirty-one leaves on the slowest revolution in medicine: the gradual transfer of the immune system's lessons from the species's chance encounters to a ledger written by hand.
↑ Vol. VI · Health · Vaccines