The AtlasOrganismsApicomplexan · Plasmodiidae
Plasmodium falciparum
The malaria parasite, and the only single-celled organism the world's cancer agency rates a probable carcinogen — for one lymphoma, in children, in one belt of Africa.
- What it is
- The deadliest human malaria parasite, a single-celled apicomplexan protozoan of the genus Plasmodium 1,2
- Life cycle
- A silent liver stage, a 48-hour red-blood-cell cycle that causes the fever, and a sexual stage taken up by the mosquito 1,3
- Its trick
- Sequestration: PfEMP1 on the infected red cell sticks it to vessel walls, out of the spleen's reach 4,5
- Burden
- About 263 million cases and 597,000 deaths in 2023 (249 million and 608,000 in 2022; 247 million and 619,000 in 2021), 94% of deaths in Africa, 80% of them in children under 5 1,3,6
- IARC rating
- Group 2A, probably carcinogenic to humans, for endemic Burkitt lymphoma in holoendemic areas, Monograph Volume 104 (2012 Lancet Oncology summary; full volume 2014) 7,8
- The cancer
- Endemic Burkitt lymphoma, a fast-growing B-cell cancer and about 90% of childhood lymphomas in sub-Saharan Africa 7,9
- Mechanism
- Chronic malaria keeps the DNA-editing enzyme AID switched on, helping cause the c-myc translocation; Epstein-Barr virus is the established cofactor 10,11,12
- Treatment
- Artemisinin-based combination therapy; intravenous artesunate for severe malaria; partial artemisinin resistance is now emerging in East Africa 2,13,14
In brief
What it is
Plasmodium falciparum is a protozoan, a single-celled parasite, of the apicomplexan group, and it causes the deadliest form of human malaria 1,2. It passes between people only through the bite of a female Anopheles mosquito, then runs a three-act life cycle: a silent stage in the liver, a cyclical stage in the red blood cells that produces the fever, and a sexual stage taken up by the next mosquito 1,3. Its signature trick is sequestration: it studs the infected red cell with its own protein, PfEMP1, which sticks the cell to the lining of small blood vessels so it is hidden from the spleen 4,5.
Why it matters
This page sits in a cancer season because of a single, well-made link. In 2012 the International Agency for Research on Cancer rated P. falciparum malaria in holoendemic areas a Group 2A carcinogen — probably carcinogenic to humans — for its part in endemic Burkitt lymphoma, a fast-growing cancer of B cells in African children 7,8. The mechanism is the page's spine: chronic malaria keeps a DNA-editing enzyme switched on in B cells, which helps produce the chromosome swap that drives the lymphoma, with the Epstein-Barr virus as the established partner 10,11,12. The honest size of this is small — it is one childhood cancer in one belt, not a general cancer effect 9,17.
How you get it
A bite. Sporozoites injected by a female Anopheles mosquito travel to the liver and multiply silently there for about a week, then burst out and invade red blood cells, where the parasite copies itself in 48-hour cycles 1,3. Some become sexual forms that the next mosquito takes up, closing the cycle; people do not pass it to each other directly 1,3. Control therefore aims at the mosquito, with insecticide-treated nets and indoor spraying 18.
What it does to you
Mostly an acute fever. In the blood stage the infected cells sequester in small vessels; in the brain this is cerebral malaria and in the placenta it is placental malaria, and severe disease brings anemia, coma and death, overwhelmingly in young African children 3,5,19. The long, rare tail is cancer: years of repeated infection in the Burkitt belt help set up endemic Burkitt lymphoma in a small number of children 7,12.
Almost everything P. falciparum does, it does within weeks — the lymphoma is a delayed, uncommon consequence of surviving many infections, not the usual outcome 12,17.
What it is
Plasmodium falciparum is a protozoan — a single-celled parasite, not a worm — and the cause of the most dangerous form of human malaria 1,2. It belongs to the apicomplexans, a group of parasites named for a cluster of structures at one end of the cell that they use to force their way into host cells 1. Five Plasmodium species infect people, but this one does nearly all the killing: in the United States, where about 2,000 malaria cases are diagnosed each year in returning travelers, P. falciparum accounts for about 79% of them, and worldwide the case-fatality of malaria is about 0.26% 2.
It is a small organism with a complicated life. The form a mosquito injects, the sporozoite, is a slender thread about 10 micrometers (thousandths of a millimeter) long; the form that invades a red blood cell, the merozoite, is a little over a micrometer across and grows and divides inside a cell about 7 micrometers wide 1. What makes it deadly is not size but behavior: during the blood stage it coats the red cell it occupies with its own protein and sticks that cell to the walls of small blood vessels, a maneuver called sequestration that keeps the infected cell out of the spleen and, in the brain or the placenta, does much of the harm 4,5,19.
For a cancer season, its interest is narrow and specific. It is the one single-celled parasite that the International Agency for Research on Cancer has placed in a carcinogen group, and it earned that place for its part in a single childhood cancer, endemic Burkitt lymphoma, in the part of Africa where malaria never lets up 7,8. The rest of this page is the case for that link, and its limits.
In three sentences each
The bite, the liver, the blood
A female Anopheles mosquito injects sporozoites, which reach the liver and mature there without symptoms; the parasites released into the blood invade red cells and multiply in roughly 48-hour cycles that cause the recurring fever, while some differentiate into the sexual forms a mosquito must pick up to continue the cycle 1,3. The red-cell (asexual erythrocytic) stage is the one that makes people ill, and the stage all first-line drugs target 2,3.
Sequestration: hiding the infected red cell
The parasite exports its own protein, P. falciparum erythrocyte membrane protein 1 (PfEMP1), to the surface of the red cell, where it binds receptors on the vessel wall so the infected cell sticks in small vessels and escapes being filtered by the spleen 4,5. Each genome holds 45-90 var genes encoding different PfEMP1 variants, switched on one at a time, which is how the parasite changes its coat to evade antibody 4. Binding to the endothelial protein C receptor marks the isolates found in cerebral malaria, while a variant called VAR2CSA binds the placenta 5,19,20.
A deaminase left switched on
Endemic Burkitt lymphoma begins in the germinal center, the structure in a lymph node where B cells divide fast and deliberately mutate their own antibody genes using the enzyme activation-induced cytidine deaminase (AID) 10. Chronic malaria keeps AID expressed too long: in Kenyan children, AID stayed high in B cells for up to 8 weeks after the parasite was cleared 21. In mice, chronic Plasmodium infection drove prolonged germinal-center expansion, DNA damage and the chromosome translocations that activate the c-myc growth gene 10,11.
Epstein-Barr virus, the established partner
The virus is the cofactor, not a bystander. Malaria weakens the T cells that police Epstein-Barr virus and reactivates the virus directly, expanding the pool of infected B cells 22,23. The virus then supplies the anti-apoptotic "second hit": its latency program rescues a translocated B cell that would otherwise kill itself, which is why Epstein-Barr virus is present in over 90% of endemic tumors 10,12,24.
Induction is not a fingerprint
That malaria raises a DNA-mutating enzyme is not the same as showing the enzyme cut the tumor's DNA. Epstein-Barr virus and malaria raise several deaminases — AID and the APOBEC3 enzymes — and all can mutate c-MYC in a dish, yet when endemic Burkitt genomes were read, only AID's sequence fingerprint was enriched 24,25. The caution and the evidence point the same way here: of the enzymes switched on, only AID left its mark on the cancer 25.
The words, defined
- Protozoan
- A single-celled organism with a nucleus; a different kind of parasite from the many-celled worms (helminths).
- Apicomplexan
- A large group of parasitic protozoa, including the malaria parasites and Toxoplasma, named for an 'apical complex' of structures they use to enter host cells.
- Sequestration
- The sticking of infected red blood cells to the lining of small blood vessels, so they are hidden from the spleen; a hallmark of P. falciparum.
- Holoendemic
- A level of malaria transmission so constant and intense that essentially every child in the area is infected, repeatedly, from infancy.
How you get it: the mosquito, the liver, the blood
Malaria is not caught from another person; it is injected by a mosquito 1,3. When a female Anopheles mosquito feeds, it injects a small number of sporozoites with its saliva. These travel in the blood to the liver and enter liver cells, where over about a week each one multiplies silently into thousands of new parasites — the pre-erythrocytic, or liver, stage, which causes no symptoms 1,3. This is the stage the leading vaccines try to stop, because the parasite is few in number and out of the blood 1,3.
Then the liver cells rupture and release merozoites into the bloodstream, and the illness begins. Each merozoite invades a red blood cell, feeds on its hemoglobin, the oxygen-carrying protein, and divides; roughly every 48 hours the infected cells burst in unison to release a new wave of parasites, and that synchronized rupture is what produces malaria's recurring fever 2,3. This asexual erythrocytic stage — the blood stage — is the one that makes people ill and the one that every first-line drug is built to kill 2,3. A few parasites instead become gametocytes, the sexual forms; when another mosquito bites, it takes these up, they mate in its gut, and new sporozoites travel to its salivary glands, closing the cycle 1,3.
The blood stage is where sequestration happens. The parasite exports PfEMP1 to the surface of the red cell, and PfEMP1 binds receptors on the endothelium, the single-cell lining of blood vessels, so the infected cell adheres in small vessels rather than circulating to the spleen, which would destroy it 4,5. The genome carries 45-90 var genes, each coding a different PfEMP1, and the parasite switches them one at a time, changing its surface faster than antibodies can keep up 4. Which receptor a variant binds decides where the trouble is. Isolates from children with cerebral malaria carry more PfEMP1 that binds the endothelial protein C receptor (EPCR), the one brain-vessel receptor the evidence agrees on 5,20. A possible second receptor, ICAM-1 (intercellular adhesion molecule 1, another adhesion protein on the vessel wall), is less certain: it was enriched in cerebral isolates in one study 5 but no different between cerebral and uncomplicated isolates in another, which questioned its role 20. A separate variant, VAR2CSA, binds chondroitin sulfate A in the placenta and causes placental malaria in pregnancy 19.
Because the parasite can only move from person to person through the mosquito, control has always meant attacking the insect. Insecticide-treated bed nets and indoor residual spraying are the mainstays, and where the Anopheles vector has grown resistant to one insecticide, programs switch to another to which it is still susceptible 18.
The words, defined
- Sporozoite
- The thread-like parasite stage a mosquito injects, which travels to and infects the liver.
- Merozoite
- The parasite form released from the liver (and later from each infected red cell) that invades red blood cells.
- Gametocyte
- The sexual form of the parasite in human blood, taken up by a mosquito to continue the cycle.
- Endothelium
- The single layer of cells lining the inside of every blood vessel; the surface infected red cells stick to in sequestration.
- PfEMP1
- P. falciparum erythrocyte membrane protein 1: the parasite protein placed on the infected red cell that both anchors it to vessel walls and helps it evade the immune system.
How many it sickens, and how many it kills
Malaria is still one of the great infectious killers. The World Health Organization estimated about 263 million cases and 597,000 deaths in 2023 — more cases than the year before but slightly fewer deaths — against about 249 million cases and 608,000 deaths in 2022, and about 247 million cases and 619,000 deaths in 2021; these are modeled annual estimates, not a direct count 1,3,6. The burden is overwhelmingly African and overwhelmingly young: about 94% of malaria deaths occur in Africa, and about 80% of those are in children under 5 years old 3.
Nearly all of this is the work of P. falciparum, the species responsible for severe disease 1,2. Severe malaria — the form that kills — follows from the blood stage and its sequestration: cerebral malaria, when infected cells pack the small vessels of the brain and the child falls into coma; severe anemia, too little hemoglobin, as waves of red cells are destroyed; and respiratory and metabolic failure 3,5. In pregnancy, parasites sequester in the placenta and harm both mother and fetus 19.
Progress against this burden, rapid in the 2000s, has slowed, and two newer threats — covered below — now shadow it: parasites resistant to the artemisinin drugs that treat the disease, and mosquitoes resistant to the insecticides that prevent it 14,18. Against that, the arrival of the first malaria vaccines is the hopeful counterweight 15,16.
The words, defined
- Cerebral malaria
- The most lethal form of severe malaria, in which infected red cells sequester in the small vessels of the brain, causing coma and, often, death.
- Anemia
- Too little hemoglobin in the blood, the protein that carries oxygen; a major cause of death in young children with malaria.
- Placental malaria
- Sequestration of infected red cells in the placenta during pregnancy, mediated by the VAR2CSA variant of PfEMP1, which harms both mother and fetus.
The one protozoan with a cancer rating
Most of the parasites in this season earned their place through flukes and worms. P. falciparum is the exception: it is the only single-celled parasite the International Agency for Research on Cancer, the World Health Organization's cancer arm, has classified as a probable human carcinogen 7,8. The evaluation is Monograph Volume 104, whose findings the Working Group summarized in The Lancet Oncology in 2012 and whose full volume PubMed dates to 2014, which is why both years circulate 7,8.
The exact classification matters, and it is narrower than "malaria causes cancer." The agency rated malaria caused by P. falciparum in holoendemic areas — not malaria in general — as Group 2A, probably carcinogenic to humans, and it did so for one cancer: endemic Burkitt lymphoma 7. Group 2A is the category the agency uses when the evidence in humans is limited but the mechanistic evidence is strong; "limited evidence in humans" is a term of art meaning a positive association was seen that chance, bias or confounding could not fully rule out 7. The specific wording "Group 2A" and "holoendemic areas" comes from the agency's public listing rather than the primary volume, which this environment cannot reach; the volume number, pages and years are from the PubMed record 7,8.
So the rating is real and it is official, but it is a hazard classification for one lymphoma under conditions of relentless childhood exposure, resting partly on mechanism — which is exactly why a later meta-analysis finding no individual-level association is not a contradiction of it, a point the rest of this page returns to 7,17. For the rating itself, the reference is the IARC Monographs Volume 104 entry, reachable through the agency's monographs database and its online Atlas of carcinogenic hazards 7,8.
The words, defined
- IARC
- The International Agency for Research on Cancer, the World Health Organization body that reviews evidence and classifies agents by how likely they are to cause cancer.
- Group 2A
- IARC's category for an agent 'probably carcinogenic to humans,' typically used when human evidence is limited but mechanistic or animal evidence is strong.
- Endemic Burkitt lymphoma
- A fast-growing cancer of B cells, the antibody-making white blood cells, that is the commonest childhood cancer in parts of equatorial Africa and classically appears as a swelling of the jaw.
How a fever helps make a lymphoma
The link's strength is its mechanism, and the mechanism is the spine of this page. Endemic Burkitt lymphoma is a cancer of B cells that begins in the germinal center, the temporary structure inside a lymph node where B cells divide at great speed and deliberately rewrite their own antibody genes to make better antibodies 10. To do that rewriting they switch on activation-induced cytidine deaminase, or AID, an enzyme that chemically edits DNA — a controlled mutation machine that, when it fires in the wrong place, can break a chromosome 10,25. The defining event of Burkitt lymphoma is one such break: a translocation that moves the c-myc growth gene next to an antibody gene, locking c-myc permanently on 10,26.
Chronic malaria feeds this process at several points. The parasite is a powerful, indiscriminate stimulant of B cells — one region of PfEMP1, CIDR1alpha (the cysteine-rich interdomain region 1-alpha of the protein), directly binds and activates them — so more B cells pass through the germinal center 10,27. More important, malaria keeps AID switched on too long: in Kenyan children with uncomplicated malaria, AID-expressing B cells stayed elevated for up to 8 weeks after the parasite had been cleared, which is how an episodic infection can leave a cumulative mark 21. The clearest experiment is in mice: chronic Plasmodium infection drove prolonged germinal-center expansion, widespread DNA damage and chromosome translocations, and shifted the lymphomas that arose toward mature, AID-dependent, translocation-bearing tumors 11.
The Epstein-Barr virus is the established partner, and the order of events is now thought to run opposite to the old slogan. The virus infects almost everyone, usually harmlessly, and hides for life inside B cells; in the Burkitt belt children acquire it in infancy, most having seroconverted by about 3 years of age, years before any tumor 23,28. Malaria then does two things to it: it suppresses the T cells that normally police virus-infected cells — in children aged 5 to 9, exactly the peak age of the tumor, Epstein-Barr-virus-specific T-cell responses were blunted in holoendemic areas 22, though a 2025 study of Kenyan children found that in acute malaria this blunting is a systemic shift toward interleukin-10 (a signaling molecule that damps immune responses) rather than a specifically anti-virus defect, complicating the "virus-specific" reading 29 — and it reactivates the latent virus directly, as the parasite protein CIDR1alpha was shown to do in culture 23. The virus's role at the end is to keep the damaged cell alive: a B cell that has just suffered a c-myc translocation would normally destroy itself, and Epstein-Barr virus blocks that self-destruction, which is why the virus is found in over 90% of endemic tumors 10,24. The current synthesis calls endemic Burkitt "a tumor of malaria survivors": the virus holds the cell steady while recurrent malaria, over years, fires the mutating enzyme that eventually breaks the chromosome 10,12.
Here is the caution the brief insists on, stated precisely. Showing that malaria raises a DNA-editing enzyme is not the same as showing that enzyme cut the tumor's DNA — an induced enzyme is not yet a fingerprint. That caution is not idle: Epstein-Barr virus and malaria raise several related deaminases, AID and the APOBEC3 family, and all of them can mutate c-MYC when forced to in a dish 25. But when researchers read the genomes of real endemic Burkitt tumors, the excess mutations fell in AID's sequence motifs and not the others' — so of the enzymes switched on, only AID actually left its mark on the cancer 24,25. In this case, unusually, the induced enzyme and the fingerprint point to the same culprit.
The words, defined
- Germinal center
- A temporary structure inside a lymph node where B cells divide rapidly and deliberately mutate their own antibody genes to improve the antibodies they make.
- Activation-induced cytidine deaminase (AID)
- An enzyme B cells switch on in the germinal center to edit their own DNA; a controlled mutation machine that, misdirected, can break a chromosome.
- c-myc translocation
- The chromosome swap that defines Burkitt lymphoma: the c-myc growth gene is joined to an antibody gene and switched permanently on.
- APOBEC3
- A family of enzymes related to AID that also edit DNA, mostly as an antiviral defense; induced in malaria and Epstein-Barr infection but, unlike AID, not found to have mutated the tumors.
- Latency
- The quiet state in which a virus such as Epstein-Barr sits inside a cell making almost no new virus; in Burkitt cells this latent program blocks the cell's self-destruct.
How strong is the link, really
A strong mechanism is not the same as a proven cause, and the honest answer is that the individual-level epidemiology is genuinely mixed while the genetic evidence is strong. The weakest link is the plainest test: does a child's own malaria status predict Burkitt lymphoma? A 2021 meta-analysis reviewed ten case-control studies and found it did not: pooling the five that measured malaria infection gave an odds ratio of 0.87 (95% CI 0.54-1.39), and pooling the five that measured a raised anti-malarial antibody titer gave only a borderline 1.50 (95% CI 1.00-2.25) 17. But the infection studies disagreed so violently (a heterogeneity statistic, I-squared, of 93.5%) that pooling them is questionable, and case-control serology captures recent, not lifelong, exposure 17.
The largest field study shows why a simple test comes out flat. In the EMBLEM study of 862 suspected cases and 2,934 controls across Uganda, Tanzania and Kenya, a history of inpatient malaria more than a year before enrollment raised Burkitt risk, while a positive malaria test at enrollment and a recent malarial fever lowered it 30. That mirror-image pattern — old severe malaria harmful, current infection apparently protective — is what a cumulative-damage mechanism with a long delay looks like, and it is exactly why snapshot serology averages out to nothing 30.
The strongest evidence is genetic, because genes cannot be confounded by poverty or recall. Sickle cell trait — carrying one copy of the sickle hemoglobin gene — protects powerfully against malaria and causes no illness, so it is a natural experiment: if it also protects against Burkitt, malaria must be doing something causal. It does. In northern Uganda, sickle cell trait was associated with a 63% lower risk of Burkitt (odds ratio 0.37, 95% CI 0.21-0.66), and a larger study of 800 cases across four countries replicated it (0.687, 95% CI 0.533-0.885) 31,32. The authors of the larger study are careful: the geographic overlap alone "remains insufficient for causal inference," and this is what strengthening it looks like 32.
The geography and a stray signal fill in the edges. In Malawi, district Burkitt incidence rose with local malaria prevalence (8.2 versus 2.9 cases per million, correlation 0.77), though an ecological comparison of districts cannot speak to individuals 9. And in a UK primary-care database, malaria exposure was tied to Burkitt in adults aged 20 to 59 (odds ratio 8.00), a hint that the mechanism is not peculiar to African childhood — but on so few people, with a confidence interval running from 1.46 to 43.7, that it is a curiosity, not a proof 33. Set side by side, these pieces neither make malaria an open-and-shut cause nor let anyone dismiss it: the mechanism is strong, the genetics support cause, and the individual-level epidemiology is unsettled.
| Study and year | Design | What it found | What it cannot show |
|---|---|---|---|
| Kotepui 2021 17 | Meta-analysis, 10 case-control studies (5 of infection, 5 of antibody) | No rise with malaria infection (OR 0.87, from 5 studies); antibody titer borderline (OR 1.50); extreme disagreement between studies | Whether lifelong exposure matters; it measures recent infection only |
| Peprah 2019 (EMBLEM) 30 | Population case-control, 3 countries | Old inpatient malaria raised risk; current infection lowered it | Causation; the pattern fits long latency but cannot prove it |
| Legason 2017 31 | Mendelian randomization | Sickle cell trait (anti-malarial) cut Burkitt risk (OR 0.37) | How much of the effect is malaria vs other gene actions |
| Hong 2024 32 | Genetic, 800 cases, 4 countries | Sickle allele cut both malaria and Burkitt risk (OR 0.687) | Individual timing; the across-loci correlation was modest |
| Gondwe 2022 9 | Ecological, Malawi districts | Burkitt incidence tracked malaria prevalence (r = 0.77) | Anything about individual children, only places |
| Karimi 2018 33 | Nested case-control, UK | Malaria tied to adult Burkitt (OR 8.00) | Much at all — tiny numbers, confidence interval 1.46-43.7 |
No single design settles it. The null meta-analysis and the positive genetics are not in conflict: one tests whether a child's current malaria status predicts the tumor, the other tests whether malaria is in the causal chain at all.
The words, defined
- Case-control study
- A study comparing people who have a disease with similar people who do not, looking back at their exposures.
- Odds ratio
- How much more likely an exposure is among cases than controls; 1.0 means no difference, below 1 means lower risk.
- I-squared
- A measure of how much the studies in a meta-analysis disagree; above about 75% they are so inconsistent that averaging them is doubtful.
- Mendelian randomization
- Using inherited gene variants, fixed at conception and so immune to lifestyle confounding, as a natural experiment to test whether an exposure truly causes a disease.
- Ecological study
- A study comparing whole populations or districts rather than individuals; it can show two things occur in the same places without showing they occur in the same people.
The belt, and the men who mapped it
The whole story began with a surgeon noticing a pattern. In 1958 Denis Burkitt, working at Mulago Hospital in Kampala, described an unusual cancer of the jaw in African children in The British Journal of Surgery — the founding clinical description of what became Burkitt lymphoma 34. The geography came four years later: in 1962, in The British Medical Journal, Burkitt mapped where the tumor did and did not occur and showed its limits followed temperature and rainfall, defining the "lymphoma belt" 35. That second paper, not the first, is the origin of the belt idea, and the climatic boundary was the clue that the cause might be something an insect carried 35.
That clue handed virology a target. Because Burkitt's map looked like the map of an insect-borne infection, Anthony Epstein and colleagues examined tumor cells under the electron microscope, and in 1964, in The Lancet, reported virus particles in cultured Burkitt lymphoblasts — the first virus ever linked to a human cancer, later named the Epstein-Barr virus 36. It is the cleanest case in medicine of epidemiology pointing virology at a discovery. The malaria half of the belt took far longer to pin down, and the mechanism — the deaminase, the translocation, the viral second hit — was only assembled in the 2010s and is still being refined 10,11,12.
The geography that defines endemic Burkitt is the geography of holoendemic malaria overlaid on early-childhood Epstein-Barr infection. Burkitt lymphoma is about 90% of childhood lymphomas in sub-Saharan Africa, and Epstein-Barr virus is found in over 90% of the tumors there, against up to 30% in the sporadic form seen elsewhere 9,24. What makes the belt is not that the virus is present — it is present almost everywhere — but that children in it acquire the virus in infancy and are bitten, and infected, all year round 12,28.
1958Seen
The jaw tumor is described
Denis Burkitt described a distinctive cancer of the jaws in African children in The British Journal of Surgery, the founding clinical account of the disease that bears his name 34.1962Explained
The lymphoma belt is mapped
Burkitt showed in The British Medical Journal that the tumor's limits followed climate, defining the geographic belt and pointing to an infectious, insect-linked cause 35.1964Seen
The first human tumor virus
Epstein, Achong and Barr found virus particles in cultured Burkitt lymphoblasts — the Epstein-Barr virus, the first virus tied to a human cancer, found because the geography suggested an infection 36.2007Explained
2012Policy
2014Explained
The germinal-center model
Torgbor and colleagues set out the three-part model — malaria deregulates AID, swells the germinal center, and raises the fraction of B cells carrying the apoptosis-blocking virus 10.2015Explained
The mouse experiment
Chronic Plasmodium infection in mice was shown to drive AID-dependent, translocation-bearing B-cell lymphomas — without changing the overall lymphoma rate 11.2021Approved
2022Explained
Only AID leaves a fingerprint
Epstein-Barr virus and malaria were shown to induce several deaminases, but only AID's mutational signature was enriched in actual endemic Burkitt tumors 25.2023Approved
2024Overturned
Artemisinin resistance reaches Africa
Partial resistance to artemisinins, long confined to Southeast Asia, was confirmed emerging in several East African countries, mediated by kelch13 mutations 14.2025Overturned
The immune block looks systemic, not virus-specific
A study of Kenyan children found that in acute malaria the T-cell change was a systemic shift toward interleukin-10, a response-damping signal, rather than a specifically anti-Epstein-Barr defect, refining the 2007 picture of a virus-specific loss of surveillance 29.2026Explained
A tumor of malaria survivors
The integrated model framed endemic Burkitt as a cancer of children who survive repeated malaria, with Epstein-Barr virus as the second hit, and argued that malaria control is a form of cancer control 12.
Treatment, resistance, and the vaccines
The drugs that treat malaria come, in part, from a plant. Quinine, from cinchona bark, was the old standby; the modern first-line drug, artemisinin, was isolated in 1971 from Artemisia annua, sweet wormwood, a herb used for more than 2,000 years in China against intermittent fever, work that later won the 2015 Nobel Prize 13,38. Today the recommended first-line treatment for P. falciparum is an artemisinin-based combination therapy — artemisinin paired with a longer-acting partner drug — which clears parasites and fever quickly; P. falciparum is now resistant to the older drug chloroquine across most of the world 2,13. For severe malaria, intravenous artesunate is first-line 2.
The new threat is that the parasite is learning to withstand even these. Partial resistance to artemisinins — slower clearance of parasites, mediated chiefly by point mutations in the parasite's kelch13 gene — was for years confined to Southeast Asia, but since about 2020 it has emerged independently in several East African countries 14. A 2026 meta-analysis of 24 East African studies put the pooled frequency of resistance-linked kelch13 mutations at 5.0% (95% CI 3-7%), rising to about 10% in Rwanda and Uganda, with the variants R561H and A675V the commonest and carriers more likely to fail treatment 39. A 2024 survey in western Uganda found such mutations in 4.8% of samples, between the known hotspots 40. This is the early stage of a problem that, in Southeast Asia, eventually broke the partner drugs too 14.
Against that backdrop the arrival of vaccines is the hopeful news, though their efficacy is modest and both target the same stage. Both WHO-recommended vaccines are aimed at the circumsporozoite protein on the sporozoite, to stop the parasite before it leaves the liver 3,41. RTS,S/AS01 (Mosquirix), which the World Health Organization recommended in October 2021 37 on a four-dose schedule from 5 months of age 15, showed in its pivotal trial 36.3% efficacy against clinical malaria in children with a booster (95% CI 31.8-40.5) and 32.2% against severe malaria, with an unexplained excess of meningitis and febrile convulsions that was watched closely 42. R21/Matrix-M, recommended in October 2023 and prequalified that December 37, is cheaper and easier to supply and reported higher trial efficacy — 75% against first clinical malaria at seasonal sites and 68% at standard sites over 12 months 16. Seen together and honestly, the two gave 72% and 55% protection against repeated clinical episodes in the first year of their African trials, and both wane, so they are a real but partial tool added to nets, spraying and drugs, not a replacement for them 41,42.
The words, defined
- Artemisinin-based combination therapy (ACT)
- The standard malaria treatment: a fast-acting artemisinin drug paired with a longer-acting partner drug, to cure quickly and slow resistance.
- Artesunate
- A water-soluble artemisinin derivative given intravenously; the first-line treatment for severe malaria.
- kelch13
- A parasite gene whose point mutations are the main markers of partial artemisinin resistance, used to track its spread.
- Circumsporozoite protein
- The protein coating the sporozoite; the target of both licensed malaria vaccines, which try to stop the parasite before it establishes in the liver.
What malaria does, and does not, cause
The honest shape of malaria's cancer burden is this: it is essentially one lymphoma, in children, in one belt — not a general carcinogenic effect across the body. The IARC classification is for endemic Burkitt lymphoma and that alone; malaria has not been established as a cause of other cancers, and the mechanism that ties it to Burkitt — a mutating enzyme switched on in dividing B cells, with a virus to keep the damaged cell alive — is specific to that setting, not a general poison 7,10,12. By the numbers, the overwhelming harm malaria does is the acute disease: the hundreds of thousands of children who die each year of cerebral malaria and anemia dwarf the lymphomas 3,9.
At the same time the link should not be waved away, because for a parasite-cancer claim it is unusually well built. Many such claims have risen on a plausible mechanism and then failed the hard test — the protozoan Trichomonas vaginalis and prostate cancer, for instance, looked promising and then did not replicate in a prospective cohort with pre-diagnostic blood 43. Malaria's case is stronger precisely because it does not rest on serology alone: it has a worked-out enzymatic mechanism, a tumor-genome fingerprint, an animal experiment, and — the part that carries the most weight — a genetic natural experiment in which a gene that blocks malaria also blocks the lymphoma 11,25,31,32.
The practical hope folds the two halves together. If malaria is in the causal chain of endemic Burkitt, then preventing malaria should, over time, prevent some of these cancers; the integrated model calls this "malaria control is cancer control" 12. That remains more a mechanistic inference than a measured result, but it is no longer wholly untested: a single-hospital series from coastal Kenya found pediatric Burkitt admission incidence falling sharply as P. falciparum admissions and parasite density declined over three decades — suggestive but ecological, built on only 95 cases, so it supports rather than settles the inference 44. It is still the reason this parasite belongs in a cancer season at all 12,17. Separately, and not part of the carcinogenicity case, recent work is exploring the opposite direction — that malaria infection can prime anti-tumor immunity in animal models, and that the placental binding protein VAR2CSA, which recognizes a chondroitin-sulfate form also displayed by many cancer cells, is being turned into a tool for cancer detection and treatment 45. For the rating itself, the reference is the IARC Monographs Volume 104 entry and the agency's online Atlas of carcinogenic hazards 7,8.
The words, defined
- Prospective cohort with pre-diagnostic blood
- A study whose participants' blood was banked before anyone fell ill, so a laboratory result cannot have been caused by the disease; the strongest design for an infection-cancer claim.
What is still unknown
Whether malaria control has actually lowered Burkitt incidence. The claim that "malaria control is cancer control" is still largely a mechanistic inference; what would settle it is a cancer-registry time series of childhood Burkitt across an African region through the 2000-2015 scale-up of insecticide-treated nets and artemisinin drugs, with malaria transmission measured alongside. The closest evidence so far is a single-hospital series from coastal Kenya, where the ten-year cumulative incidence of pediatric Burkitt admissions ran 93.1 per 100,000 in 1990-1999 and 130.3 in 2000-2009 before falling to 10.2 in 2010-2020 as P. falciparum admissions and parasite density declined (correlation 0.53, p=0.0024) — but it is ecological, rests on only 95 cases (just 4 in the last decade), and is confounded by changing diagnosis and referral, so it supports rather than settles the inference 12,17,44.
Why individual malaria status does not predict the tumor while the genetics say malaria is causal. The leading explanation is a long latency and cumulative damage — old severe malaria matters, current infection does not — but the exact exposure window, and how many infections over how many years it takes, are not known 17,21,30.
How much of the mechanism rests on a few hands. The AID germinal-center model traces largely to two laboratories, and the mouse experiment used a rodent parasite in a tumor-prone, p53-deficient background rather than P. falciparum in a normal animal; the human AID evidence, though growing, is still a handful of cohorts 10,11,21,46.
Whether resistance will outrun the gains. Partial artemisinin resistance is spreading in East Africa and the leading vaccines are only partly effective and wane, so whether the recent tools will hold the line against the acute disease — which is, after all, the vast majority of malaria's harm — is an open question 14,39,41.
The tests, one by one
- Giemsa-stained blood filmA drop of blood is spread, stained and read under the microscope for parasites inside red cells; a thick film finds them and a thin film names the species and counts the share of cells infected 2.The reference method almost everywhere, and the way species and parasite load are established 2Depends on the reader and the parasite load; low-level infection can be missed on a single film 2.Light infections, and the sequestered parasites hiding in deep vessels, which may not show in peripheral blood 2,5.
- Rapid diagnostic test (antigen)A dipstick detects a parasite protein in a finger-prick of blood in minutes, without a microscope 2; rapid-test positivity was the on-the-spot measure of current infection in the EMBLEM study 30.Point-of-care diagnosis where microscopy is not available 2Good for P. falciparum at moderate loads, but parasites with deletions of the target protein can read falsely negative 2.Very low parasite loads, and strains that have lost the antigen the test looks for 2.
- PCR and kelch13 genotypingMolecular tests use PCR (the polymerase chain reaction) to copy and read parasite DNA, detecting low-level or mixed infection and, by sequencing the kelch13 gene, the mutations that mark partial artemisinin resistance 14,39.Reference confirmation and the backbone of resistance surveillance across Africa 14,39The most sensitive way to detect the parasite and the only way to read resistance markers; not a bedside test 39.Nothing of the parasite at the limits of detection, but it is slow, needs a laboratory, and is not standardized everywhere 14.
- Tumor biopsy and the MYC rearrangementEndemic Burkitt lymphoma is diagnosed not from the parasite but from the tumor: a biopsy showing sheets of B cells carrying the IG-MYC translocation, the chromosome swap that is the disease's hallmark 24,26.To diagnose the lymphoma the parasite is tied to, and to classify it by Epstein-Barr-virus status 24,26The translocation is near-universal in Burkitt, but it is not specific to the malaria-linked form — sporadic tumors carry it too 26.It cannot say how a given tumor arose; the parasite is long gone by the time the cancer appears 12.
Where it connects
In the Atlas
Topics on the map
On the map
A star in Protozoa, one of 16. Five species infect people, and two of them park a dormant form in the liver that no blood-stage drug touches.
Sources
46 sources, numbered as they are cited. Every one was checked against PubMed or its publisher before it was cited here; the note under each says what it shows and what it does not.
- 1Richie TL, Church LWP, Murshedkar T, et al. Sporozoite immunization: innovative translational science to support the fight against malaria.doi:10.1080/14760584.2023.2245890 · PMID 37571809
Review by the group developing a whole-sporozoite vaccine. Read for the life cycle and burden it states: the sporozoite is the stage Anopheles mosquitoes inject into people, and the only vaccine immunogen reaching over 90% efficacy against P. falciparum; it cites the World Health Organization figure of 247 million malaria cases and 619,000 deaths in 2021. Conflict of interest: most authors are employed by Sanaria Inc., a company developing a sporozoite malaria vaccine.
- 2Daily JP, Minuti A, Khan N. Diagnosis, Treatment, and Prevention of Malaria in the US: A Review.doi:10.1001/jama.2022.12366 · PMID 35916842
Review. First-line therapy for P. falciparum is an artemisinin-based combination; intravenous artesunate is first-line for severe malaria; P. falciparum is resistant to chloroquine in most of the world; worldwide malaria mortality is about 0.26%; about 2,000 cases are diagnosed in the US each year, mostly in returning travelers, P. falciparum accounting for about 79% of them, with a US case-fatality of about 0.3%.
- 3Khan S, Patel MP, Patni AD, Cha SJ. Targeting Plasmodium Life Cycle with Novel Parasite Ligands as Vaccine Antigens.doi:10.3390/vaccines12050484 · PMID 38793735
Review. States the World Health Organization estimate of 249 million cases and 608,000 deaths in 2022, with 94% of deaths in Africa and 80% of those in children under 5; that RTS,S/AS01 uses the circumsporozoite protein to target the liver stage; that the WHO Malaria Vaccine Implementation Program, completed at the end of 2023, found the vaccine reduced mortality by 13%; and that the three vaccine-target stages are the pre-erythrocytic (sporozoite/liver), asexual erythrocytic (merozoite/red cell) and sexual (transmission) stages. Published in an author-pays (MDPI) journal.
- 4Cubillos EFG, Prata IO, Fotoran WL, Ranford-Cartwright L, Wunderlich G. The Transcription Factor PfAP2-O Influences Virulence Gene Transcription and Sexual Development in Plasmodium falciparum.doi:10.3389/fcimb.2021.669088 · PMID 34268135
States that each parasite genome carries 45-90 var genes encoding PfEMP1, that they are expressed one at a time (mono-allelically) to produce antigenic variation, and that PfEMP1 on the infected red cell binds endothelial receptors and drives sequestration; identifies PfAP2-O as a regulator. Mechanistic knockdown study, not a clinical one.
- 5Storm J, Jespersen JS, Seydel KB, et al. Cerebral malaria is associated with differential cytoadherence to brain endothelial cells.doi:10.15252/emmm.201809164 · PMID 30610112
Malawian pediatric isolates: infected red cells from cerebral malaria bound brain microvascular endothelium more strongly than those from uncomplicated malaria, and carried more var-gene transcripts predicted to bind endothelial protein C receptor (EPCR) and ICAM-1. Supports sequestration in the brain as a basis of cerebral disease; an ex vivo binding assay, not an in vivo causal proof.
- 6Legesse F, Berhane N. Analysis of a seven-year trend in malaria prevalence at Pawe General Hospital, Northwestern Ethiopia: a retrospective cross-sectional study.doi:10.1186/s12879-026-12546-w · PMID 41507864
A hospital trend study whose introduction quotes the most current global figure: an estimated 263 million malaria cases and 597,000 deaths worldwide in 2023, citing the World Health Organization's World Malaria Report 2024. Used here only for that up-to-date burden number — as the other review sources are used for the earlier-year WHO figures; the local Ethiopian prevalence data are not drawn on.
- 7Bouvard V, Baan RA, Grosse Y, et al. Carcinogenicity of malaria and of some polyomaviruses.doi:10.1016/S1470-2045(12)70125-0 · PMID 22577663
The IARC Working Group's own published summary of the Volume 104 evaluation; indexed as a News item (an expert hazard-evaluation summary, not a research review). Its MeSH set names Burkitt Lymphoma, Epstein-Barr Virus, Malaria and Plasmodium falciparum. The summary is 2012 and the full volume dated 2014, which is why both years are cited. No abstract in PubMed.
- 8IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Malaria and some polyomaviruses (SV40, BK, JC, and Markel cell viruses).PMID 26173303
The monograph volume that evaluated malaria; PubMed carries a misspelling of "Merkel" in the title and holds no abstract; MeSH terms include Malaria and Neoplasms. The specific classification string "Group 2A" could not be read from IARC's own site, which this environment cannot reach, and is taken from IARC's public listing; the volume, pages and years are from the PubMed record. No DOI in PubMed (a primary document PubMed does not fully index).
- 9Gondwe Y, Salima A, Manda A, et al. Spatial distribution of incident pediatric Burkitt lymphoma in central and northern Malawi and association with malaria prevalence.doi:10.1002/pbc.29867 · PMID 35731580
220 confirmed cases. District Burkitt incidence was 8.2 per million in central districts versus 2.9 per million in the north and tracked P. falciparum prevalence (Pearson r = 0.77), each 1% rise in parasite prevalence predicting 0.2 more cases per million, with referral distance controlled; Burkitt is about 90% of pediatric lymphomas in sub-Saharan Africa. An ecological design: it speaks to places, not to individuals.
- 10Torgbor C, Awuah P, Deitsch K, Kalantari P, Duca KA, Thorley-Lawson DA. A multifactorial role for P. falciparum malaria in endemic Burkitt's lymphoma pathogenesis.doi:10.1371/journal.ppat.1004170 · PMID 24874410
Proposes and supports the three-part germinal-center model: P. falciparum deregulates activation-induced cytidine deaminase (AID), raising the chance of a c-myc translocation; increases the number of B cells passing through the germinal center; and raises the fraction of them infected with Epstein-Barr virus and so protected from c-myc-driven apoptosis. The paper that frames the virus's role as apoptosis rescue.
- 11Robbiani DF, Deroubaix S, Feldhahn N, et al. Plasmodium Infection Promotes Genomic Instability and AID-Dependent B Cell Lymphoma.doi:10.1016/j.cell.2015.07.019 · PMID 26276629
Mouse study using Plasmodium chabaudi, not P. falciparum. Chronic infection produced prolonged germinal-center expansion, widespread DNA damage and chromosome translocations; crucially, the abstract states infection "does not change the overall rate" of lymphoma but shifts the spectrum toward mature, AID-dependent, translocation-bearing B-cell tumors, in a p53-deficient tumor-prone background. Animal evidence for the mechanism, not evidence that malaria raises cancer rates.
- 12Rochford R, Mbulaiteye SM. The Complex Interplay of Malaria and EBV in Burkitt Lymphoma.doi:10.3390/cancers18132146 · PMID 42449688
The current integrated model, co-written by the NCI epidemiologist behind the EMBLEM studies. Malaria drives the AID-mediated c-myc translocation in proportion to recurrent parasite burden; Epstein-Barr virus provides the anti-apoptotic "second hit"; endemic Burkitt is "a tumor of malaria survivors," peaking years after peak malaria mortality, and "malaria control is a form of cancer control." A synthesis and a mechanistic inference, not a measured outcome; author-pays (MDPI) journal.
- 13Kano S. Artemisinin-based combination therapies and their introduction in Japan.PMID 24979951
Review. Artemisinin was discovered in 1971 from Artemisia annua, used for more than 2,000 years in China against intermittent fever; artemisinin-based combinations are the recommended first-line treatment for all human malarias and clear parasite and fever rapidly, including in cerebral malaria. No DOI in PubMed.
- 14Rosenthal PJ, Asua V, Bailey JA, et al. The emergence of artemisinin partial resistance in Africa: how do we respond?doi:10.1016/S1473-3099(24)00141-5 · PMID 38552654
Review. Partial resistance to artemisinins, long confined to Southeast Asia, has now emerged in several East African countries, mediated chiefly by point mutations in the P. falciparum kelch13 protein; the authors call for expanded genomic surveillance and new regimens. An authoritative statement of the threat.
- 15Syed YY. RTS,S/AS01 malaria vaccine (Mosquirix): a profile of its use.doi:10.1007/s40267-022-00937-3 · PMID 36093265
Product profile. WHO recommends a 4-dose RTS,S schedule from 5 months of age in areas of moderate-to-high transmission, with an optional 5-dose seasonal schedule; pilot implementation showed a favorable safety profile, improved equity of access and reduced severe malaria. The author is employed by the publisher; a secondary summary.
- 16Datoo MS, Dicko A, Tinto H, et al. Safety and efficacy of malaria vaccine candidate R21/Matrix-M in African children: a multicentre, double-blind, randomised, phase 3 trial.doi:10.1016/S0140-6736(23)02511-4 · PMID 38310910
More than 4,800 children. Twelve-month efficacy against first clinical malaria was 75% (95% CI 71-79) at seasonal sites and 68% (61-74) at standard sites; antibody to the circumsporozoite NANP repeat correlated with protection; the vaccine received a WHO recommendation and prequalification. Funded by the Serum Institute of India (the manufacturer) and others.
- 17Kotepui KU, Kotepui M. Malaria Infection and Risk for Endemic Burkitt Lymphoma: A Systematic Review and Meta-Analysis.doi:10.3390/ijerph18115886 · PMID 34070881
Ten case-control studies were reviewed — five measuring malaria infection, five measuring anti-malarial IgG titer. Pooling the five infection studies, current or past malaria infection did not raise the odds of endemic Burkitt lymphoma (odds ratio 0.87, 95% CI 0.54-1.39; I-squared 93.5%; malaria in 604/1,506 eBL vs 2,117/4,549 non-eBL), and pooling the five antibody studies gave a borderline raised IgG titer (1.50, 95% CI 1.00-2.25). Heterogeneity is extreme, and the authors call for longitudinal studies; this refutes individual-level prediction, not the geography or the genetics.
- 18Alonso S, Chaccour CJ, Wagman J, et al. Cost and cost-effectiveness of indoor residual spraying with pirimiphos-methyl in a high malaria transmission district of Mozambique with high access to standard insecticide-treated nets.doi:10.1186/s12936-021-03687-1 · PMID 33691706
An economic evaluation in a high-transmission district where the Anopheles vector was resistant to pyrethroids but susceptible to pirimiphos-methyl; indoor residual spraying added to treated nets averted cases and was cost-effective. Used here only for the point that control rests on attacking the mosquito; a single-district model, not a global estimate.
- 19Doritchamou JYA, Renn JP, Hviid L, Duffy PE. A conformational epitope in placental malaria vaccine antigen VAR2CSA: What does it teach us?doi:10.1371/journal.ppat.1011370 · PMID 37228009
Perspective. VAR2CSA is the PfEMP1 variant that binds chondroitin sulfate A and sequesters infected red cells in the placenta, causing placental malaria; its sequence polymorphism has hindered a vaccine. Shows sequestration has a pregnancy-specific form.
- 20Joste V, Guillochon E, Fraering J, et al. PfEMP1 A-Type ICAM-1-Binding Domains Are Not Associated with Cerebral Malaria in Beninese Children.doi:10.1128/mBio.02103-20 · PMID 33203751
73 cerebral and 98 uncomplicated malaria children: EPCR-binding PfEMP1 domains were overexpressed in cerebral isolates, but the ICAM-1-binding motif appeared in both groups, so its role was questioned. A reminder that the sequestration-pathology mechanism is partly established (EPCR) and partly unresolved (ICAM-1).
- 21Ariera B, Guyah B, Rahkola J, et al. Sustained activation induced cytidine deaminase (AID) expression in B cells following Plasmodium falciparum malaria infection in Kenyan children.doi:10.1093/jimmun/vkaf005 · PMID 40085439
Children with uncomplicated malaria had elevated CD19+AID+ B cells that stayed high up to 8 weeks after the parasite was cleared, with 52% of AID in the nucleus; in culture, Epstein-Barr virus, CpG (mimicking parasite DNA) and the B-cell factor BAFF together raised AID+ frequency about 30% above any single stimulus. The cohort part is human observational; the synergy figure is in vitro, and the authors claim only a "potential" influence on the Burkitt translocation.
- 22Moormann AM, Chelimo K, Sumba PO, Tisch DJ, Rochford R, Kazura JW. Exposure to holoendemic malaria results in suppression of Epstein-Barr virus-specific T cell immunosurveillance in Kenyan children.doi:10.1086/511984 · PMID 17299709
Human observational. Children aged 5-9 in a holoendemic area had significantly fewer Epstein-Barr-virus-specific interferon-gamma responses than peers in an area of unstable transmission, while responses to other antigens were unchanged; the gap coincides with the peak age of endemic Burkitt lymphoma. A specific loss of viral surveillance, not generalized immune suppression.
- 23Chêne A, Donati D, Guerreiro-Cacais AO, et al. A molecular link between malaria and Epstein-Barr virus reactivation.doi:10.1371/journal.ppat.0030080 · PMID 17559303
In vitro. CIDR1alpha of PfEMP1 drives latent Epstein-Barr virus into its lytic cycle in an EBV-positive B-cell line and in blood cells from healthy donors and children with Burkitt lymphoma; the first report that a microbial protein can reactivate the virus. Cell culture; the authors say only that it "may increase the risk."
- 24Grande BM, Gerhard DS, Jiang A, et al. Genome-wide discovery of somatic coding and noncoding mutations in pediatric endemic and sporadic Burkitt lymphoma.doi:10.1182/blood-2018-09-871418 · PMID 30617194
Whole-genome and transcriptome analysis showed a genome-wide excess of aberrant somatic hypermutation in Epstein-Barr-virus-positive tumors (an AID signature) and fewer driver mutations, especially in apoptosis genes; EBV is positive in over 90% of cases in malaria-endemic regions versus up to 30% elsewhere, and tumor EBV status, not geography, defines the phenotype. Funded in part by a Burkitt-specific research charity.
- 25Summerauer AM, Jäggi V, Ogwang R, et al. Epstein-Barr virus and malaria upregulate AID and APOBEC3 enzymes, but only AID seems to play a major mutagenic role in Burkitt lymphoma.doi:10.1002/eji.202249820 · PMID 35503749
Epstein-Barr virus raised the related enzymes APOBEC3B and APOBEC3G and malaria raised APOBEC3A, and all three mutated c-MYC when overexpressed, yet mutational enrichment in actual endemic Burkitt tumors was found only in AID sequence motifs. The discipline the page turns on: inducing a deaminase is not the same as leaving its fingerprint, and only AID's fingerprint is in the tumor.
- 26López C, Kleinheinz K, Aukema SM, et al. Genomic and transcriptomic changes complement each other in the pathogenesis of sporadic Burkitt lymphoma.doi:10.1038/s41467-019-08578-3 · PMID 30926794
Whole-genome and transcriptome sequencing of 39 sporadic Burkitt lymphomas, with no malaria involved, which still carried the defining IG-MYC translocation. Shows the translocation is the shared hallmark reachable by several routes, so malaria is one route to it, not the only one.
- 27Donati D, Zhang LP, Chêne A, et al. Identification of a polyclonal B-cell activator in Plasmodium falciparum.doi:10.1128/IAI.72.9.5412-5418.2004 · PMID 15322039
In vitro, with cells from non-immune donors. The cysteine-rich interdomain region 1alpha (CIDR1alpha) of PfEMP1 binds and activates human B cells, driving proliferation, size increase, IgM secretion and release of tumor necrosis factor-alpha and interleukin-6, in a pattern like staphylococcal protein A; the authors argue the var/PfEMP1 family evolved to manipulate the immune system, not only to sequester. Cell culture, not a tumor-scale demonstration in people.
- 28Muckian MD, Shi T, Qarkaxhija V, Kapoor S, Morgan T, Stagg HR. Equity in protection: bridging global data gaps for an EBV vaccine - a systematic review and meta-analysis.doi:10.1136/bmjgh-2024-015534 · PMID 40813096
The first meta-analysis of Epstein-Barr-virus seroprevalence by age: among 0-4 year olds, 59% were seropositive in middle-income countries (95% CI 28-91%) versus 29% in high-income countries (95% CI 16-41%), both with I-squared of 99%. Shows why "EBV is everywhere" misleads — age at infection, not adult ubiquity, is what matters for Burkitt. Framed around a commercial EBV-vaccine pipeline; the estimates are very heterogeneous.
- 29Ariera BO, Guyah B, Onditi I, et al. CD4 and CD8 T-cell response is dominated by IL-10-secreting cells in children with uncomplicated Plasmodium falciparum malaria.doi:10.1093/immhor/vlaf045 · PMID 41285032
A small human study: 10 Kenyan children with acute malaria, matched recovery samples, and 10 controls. Clinical malaria did not impair T-cell activation but shifted both CD4 and CD8 cytokine output toward interleukin-10, and the authors conclude the malaria-induced change is systemic immune suppression, not an Epstein-Barr-virus-specific defect — a 2025 refinement of the 2007 "EBV-specific surveillance loss" framing. Small sample, and a shift in culture-stimulated cytokines rather than a measured change in tumor risk.
- 30Peprah S, Ogwang MD, Kerchan P, et al. Risk factors for Burkitt lymphoma in East African children and minors: A case-control study in malaria-endemic regions in Uganda, Tanzania and Kenya.doi:10.1002/ijc.32390 · PMID 31054214
The population-based EMBLEM study, 862 suspected cases and 2,934 controls. A history of inpatient malaria more than 12 months before enrollment raised risk, while malaria-attributed fever in the prior 6 months and a positive rapid test at enrollment lowered it; living in a mass-malaria-suppression area carried higher risk, most likely confounding by baseline transmission. The paradox that fits a cumulative-damage mechanism with long latency.
- 31Legason ID, Pfeiffer RM, Udquim KI, et al. Evaluating the Causal Link Between Malaria Infection and Endemic Burkitt Lymphoma in Northern Uganda: A Mendelian Randomization Study.doi:10.1016/j.ebiom.2017.09.037 · PMID 29033373
202 cases and 624 controls. Sickle cell trait, which protects against malaria, was associated with lower Burkitt risk (adjusted odds ratio 0.37, 95% CI 0.21-0.66). Because genotype is fixed at conception, this supports a causal role for malaria; a variant (SEMA3C) protected against lymphoma but not malaria, an honest loose end. One author's listed address is a company (cgix.com).
- 32Hong HG, Gouveia MH, Ogwang MD, et al. Sickle cell allele HBB-rs334(T) is associated with decreased risk of childhood Burkitt lymphoma in East Africa.doi:10.1002/ajh.27149 · PMID 38009642
800 cases and 3,845 controls across four countries, scanning 22 malaria-risk genes. HBB-rs334(T) lowered both P. falciparum infection (odds ratio 0.752) and Burkitt risk (0.687); the authors state plainly that the geographic association alone "remains insufficient for causal inference." The Mendelian-randomization result at scale; the correlation across loci was modest (Spearman's rho 0.37).
- 33Karimi P, Birmann BM, Anderson LA, et al. Risk factors for Burkitt lymphoma: a nested case-control study in the UK Clinical Practice Research Datalink.doi:10.1111/bjh.15229 · PMID 29676453
156 cases and 608 controls in UK primary care. Malaria exposure was associated with Burkitt lymphoma at ages 20-59 (odds ratio 8.00, 95% CI 1.46-43.7), outside Africa entirely, while nothing predicted childhood Burkitt in the UK. Tiny numbers and an enormous confidence interval: suggestive and underpowered, not a replication.
- 34Burkitt D. A sarcoma involving the jaws in African children.doi:10.1002/bjs.18004619704 · PMID 13628987
The founding clinical description, by a surgeon at Mulago Hospital, Kampala. PubMed holds no abstract; index terms are JAWS/neoplasms and SARCOMA/in infant and child. Cited for the description only; case counts were not verified against the original and are not quoted.
- 35Burkitt D. Determining the climatic limitations of a children's cancer common in Africa.doi:10.1136/bmj.2.5311.1019 · PMID 14017064
Where the geographic and climatic boundary of the tumor — the "lymphoma belt" — was established, four years after the 1958 description; this, not the 1958 paper, is the origin of the belt idea that led to the malaria hypothesis. No abstract in PubMed; no figures quoted from it here.
- 36Epstein MA, Achong BG, Barr YM. Virus particles in cultured lymphoblasts from Burkitt's lymphoma.doi:10.1016/s0140-6736(64)91524-7 · PMID 14107961
Electron microscopy found herpes-type virus particles in cultured Burkitt lymphoblasts — the first human tumor virus, later named Epstein-Barr virus. The virus was looked for because Burkitt's geography suggested an infectious cause. No abstract; index terms call the agent a herpesvirus, since it was not yet known to be new.
- 37Osoro CB, Ochodo E, Kwambai TK, et al. Policy uptake and implementation of the RTS,S/AS01 malaria vaccine in sub-Saharan African countries: status 2 years following the WHO recommendation.doi:10.1136/bmjgh-2023-014719 · PMID 38688566
An implementation review that dates the two World Health Organization recommendations precisely: RTS,S/AS01 was recommended in October 2021, and R21/Matrix-M in October 2023, the latter prequalified in December 2023. Cited for those dates, which the vaccine trial and profile papers do not themselves state; the country-rollout detail is not drawn on here.
- 38Tu Y. The discovery of artemisinin (qinghaosu) and gifts from Chinese medicine.doi:10.1038/nm.2471 · PMID 21989013
Historical article by the discoverer. Artemisinin (qinghaosu) was isolated from Artemisia annua (sweet wormwood), a plant long used in Chinese medicine; the work later won the 2015 Nobel Prize in Physiology or Medicine. No abstract in PubMed.
- 39Kapesa A, Baraka V, Mwaiswelo R, et al. Plasmodium falciparum non-synonymous Kelch13 mutations mediating artemisinin resistance in East Africa: A systematic review and meta-analysis: 2014-2024.doi:10.1371/journal.pone.0354429 · PMID 42520053
Twenty-four studies. Pooled kelch13 non-synonymous mutation prevalence was 5.0% (95% CI 3-7%) in East Africa, rising to about 10% in Rwanda and Uganda, with R561H and A675V the commonest; carriers were significantly more likely to fail treatment. Heterogeneity was high (I-squared over 95%).
- 40van Loon W, Schallenberg E, Mande E, et al. Plasmodium falciparum artemisinin partial resistance markers in Fort Portal, Western Uganda, 2024.doi:10.1128/aac.01755-24 · PMID 40265952
kelch13 resistance-associated mutations, including A675V and R561H, were found in 4.8% of 126 samples in western Uganda, between the Rwandan and Ugandan resistance hotspots. A small local survey showing incipient spread.
- 41Macià D, Pons-Salort M, Moncunill G, Dobaño C. The effect of disease transmission on time-aggregated treatment efficacy estimates: a critical analysis of factors influencing the RTS,S and R21 malaria vaccine phase 3 trials.doi:10.1016/S1473-3099(25)00090-8 · PMID 40286802
Analysis. Both WHO-approved subunit vaccines target the circumsporozoite protein; in African phase 3 trials R21 and RTS,S showed 72% (95% CI 69-76) and 55% (51-59) efficacy against multiple clinical episodes in the first year, and both wane, with estimates depending on transmission intensity. Useful for an honest side-by-side rather than a single headline figure.
- 42RTS,S Clinical Trials Partnership. Efficacy and safety of RTS,S/AS01 malaria vaccine with or without a booster dose in infants and children in Africa: final results of a phase 3, individually randomised, controlled trial.doi:10.1016/S0140-6736(15)60721-8 · PMID 25913272
The pivotal RTS,S trial, 8,922 children and 6,537 young infants. With a booster, efficacy against clinical malaria over the full follow-up was 36.3% in children (95% CI 31.8-40.5) and 25.9% in young infants (19.9-31.5), and 32.2% against severe malaria in children (13.7-46.9); an excess of meningitis (11 vs 10 vs 1 across arms) and of febrile convulsions after the booster was noted. Funded by GlaxoSmithKline (the manufacturer) and the PATH Malaria Vaccine Initiative.
- 43Nagata M, Tome A, White K, et al. No Association of Trichomonas vaginalis Seropositivity with Advanced Prostate Cancer Risk in the Multiethnic Cohort: A Nested Case-Control Study.doi:10.3390/cancers15215194 · PMID 37958367
A prospective cohort with pre-diagnostic blood: the protozoan Trichomonas vaginalis was not associated with advanced prostate cancer (adjusted odds ratio 1.31, 95% CI 0.67-2.53; 470 cases, 470 controls). Included only as contrast — a once-publicized protozoan-cancer link that did not hold up, which is why malaria's unusually well-supported case stands out.
- 44Mwaniki MK, Mohammed S, Kariuki N, et al. A "Familiar Foe Revisited": examining the relationship between endemic Burkitt's lymphoma and changing malaria admissions in the coastal region of Kenya.doi:10.1186/s12885-025-15450-9 · PMID 41413501
A single-hospital pediatric series in coastal Kenya across three decades: among 95 Burkitt cases, the ten-year cumulative admission incidence was 93.1 per 100,000 in 1990-1999 and 130.3 in 2000-2009, then fell to 10.2 in 2010-2020 as P. falciparum admissions and median parasite density declined, with a positive correlation between endemic Burkitt and malaria (r = 0.53, P = 0.0024). The first longitudinal, if ecological, support for "malaria control is cancer control"; built on very few cases (only 4 in the last decade) and confounded by changing diagnosis and referral, so it is suggestive, not decisive.
- 45Su XZ, Stadler RV, Teklemichael AA, Xu F, Wu J. Malaria and cancer: common features and interactions.doi:10.1016/j.pt.2026.07.004 · PMID 42527283
Review from a National Institutes of Health malaria laboratory. Alongside malaria's role in promoting Burkitt lymphoma, it describes the opposite research strand: in animal models malaria infection can activate anti-tumor immunity and improve survival, and the placental PfEMP1 variant VAR2CSA binds a chondroitin-sulfate form found on metastatic cancer cells, which is being developed for cancer detection and therapy. Included only to mark that separate, non-carcinogenic strand; animal-model and tool-development work, not a human cancer finding.
- 46O'Connor E, Scanlan P, Smith OP, Halasz M. Clinical and molecular variations in Burkitt lymphoma.doi:10.1016/j.tranon.2025.102611 · PMID 41297313
Review arguing that EBV-positive and EBV-negative Burkitt are distinct entities and that aberrant AID expression, in the setting of Epstein-Barr virus and chronic malaria, is the most likely cause of endemic Burkitt. Recent, independent statement that the AID model has become the consensus.
This is education, not medical advice. Nothing on this page is written with knowledge of your history, your medications or your risks, and nothing here is a dose. Do not start or stop any treatment on the basis of it — talk to your own physician. Read the full medical disclaimer.