What actually raises the risk of leukemia
Last reviewed November 16, 2026. OwieClear editorial team. Waiting for clinician review. Not medical advice. Talk with your doctor about your own care.
People want a reason, and for most individual cases there is not one. The mutations that cause leukemia usually arise from ordinary copying errors in dividing cells.
That said, some exposures genuinely do raise risk, and they are worth knowing, partly because a few of them are avoidable and a few of the things people worry about are not on the list.
Related pages: Leukemia (blood cancer)
The bottom line
- Most leukemia mutations happen by chance during cell division, and rise with age.
- Benzene exposure is an established cause of acute myeloid leukemia.
- High-dose ionizing radiation and some previous chemotherapy raise risk years later.
- A few inherited syndromes and one virus, HTLV-1, raise risk substantially but account for a small share of cases.
- Whether low-level environmental exposures contribute meaningfully to overall risk.
- What causes the childhood ALL peak, where an infection-related hypothesis is leading but unproven.
The story, step by step
- 1. Chance and age ()
- 2. Benzene ()
- 3. Radiation and prior treatment ()
- 4. Viruses and inherited syndromes ()
- 5. The childhood question ()
What the evidence labels mean
The biggest factor is the dullest one
What happens
Risk rises with age for most leukemias because mutations accumulate as cells divide.
What the research shows
- Each cell division carries a small chance of a copying error. Marrow divides constantly for a lifetime.
- By their seventies, a large share of people carry a detectable mutated clone in their blood without any disease.
- This is why AML, CLL and CML are predominantly diseases of older adults.
- Childhood ALL is the clear exception, with a peak between ages two and five that age-related mutation does not explain.
Why it matters
For most people diagnosed, the honest answer to why me is bad luck in a system that has to copy itself billions of times a day.
What's still uncertain
- How much of overall cancer risk is attributable to random copying error versus environment remains actively debated.
Read more
- NCI: LeukemiaGovernment
Next: Benzene
The clearest chemical link
What happens
Benzene is one of the best-established chemical causes of cancer in humans.
What the research shows
- It is a component of crude oil, gasoline and cigarette smoke, and is used industrially as a solvent.
- The liver converts it into reactive compounds that reach the marrow and damage DNA in blood-forming cells.
- The link is strongest for acute myeloid leukemia and myelodysplastic syndromes.
- Smoking is a genuine and avoidable risk factor for AML, largely through benzene and other chemicals in smoke.
- Occupational exposure limits exist for this reason; historic exposures in some industries were far higher than anything permitted now.
Why it matters
Smoking is the one leukemia risk factor most people can actually do something about, and it is rarely mentioned in that context.
What's still uncertain
- Risk from low-level everyday exposure, such as at a filling station, is far smaller than occupational exposure and hard to quantify.
Read more
- NCI: Benzene and cancer riskGovernment
- NCI: LeukemiaGovernment
Therapy-related leukemia
What happens
Treatments that damage DNA to kill cancer cells can, years later, cause a new one.
What the research shows
- High-dose ionizing radiation raises leukemia risk, demonstrated most clearly in atomic bomb survivors and in some occupational and medical exposures.
- Alkylating chemotherapy and topoisomerase II inhibitors given for an earlier cancer can cause therapy-related AML or myelodysplastic syndrome, typically several years afterwards.
- Therapy-related leukemias often carry worse genetics and are harder to treat than those arising on their own.
- This risk is known and factored into treatment decisions. It is small compared with the benefit of treating the original cancer.
Why it matters
If you had chemotherapy or radiation years ago, new unexplained fatigue, bruising or infections is a reason for a blood count rather than a wait.
What's still uncertain
- Diagnostic imaging doses are far lower than therapeutic doses, and attributing individual cases to medical scans is not possible.
Read more
- NCI: LeukemiaGovernment
Real, but uncommon
What happens
Two categories are genuinely established and together account for a small minority of cases.
What the research shows
- HTLV-1, a retrovirus, integrates into T-cell DNA and can cause adult T-cell leukemia/lymphoma. It is concentrated in southwestern Japan, the Caribbean, parts of West Africa and South America, and most people infected never develop it.
- Inherited conditions including Fanconi anemia, Li-Fraumeni syndrome, ataxia-telangiectasia, Bloom syndrome and some familial platelet disorders raise risk substantially.
- Down syndrome raises the risk of both ALL and AML in children, through the extra copy of chromosome 21 and mutations in GATA1, not through faulty DNA repair.
- Epstein-Barr virus is linked to several lymphomas rather than to the common leukemias, and the two are often conflated.
Why it matters
These groups need different surveillance, which is why family history and syndrome diagnosis are asked about at the first appointment.
What's still uncertain
- Why only a small percentage of people with HTLV-1 develop leukemia, usually decades later, is not understood.
Read more
- NCI: LeukemiaGovernment
- OwieClear: How EBV hijacks your cellsOwieClear
Next: The childhood question
The peak nobody has fully explained
What happens
The sharp rise in ALL in young children is one of the open questions in cancer epidemiology.
What the research shows
- Studies of identical twins show that the first mutation often occurs before birth, so the process starts in the womb.
- Only a small fraction of children carrying that pre-leukemic clone ever develop leukemia, so a second event is required.
- The leading hypothesis, from Mel Greaves, proposes that an immune system with too little early microbial exposure reacts abnormally to a later common infection, and that this reaction triggers the second mutation.
- Supporting observations include lower rates in children who attended daycare early or had older siblings, and clusters following unusual population mixing.
- It remains a hypothesis. It has not been proven, and no preventive measure follows from it yet.
Why it matters
Nothing a parent did causes childhood leukemia. That point is worth stating plainly, because parents routinely search for something they did wrong.
What's still uncertain
- The hypothesis is supported by consistent epidemiology but has no direct experimental proof in humans.