What Causes Leukemia? Genetics, Environment, and Chance

Leukemia is not one disease. It is a group of cancers that begin in the bone marrow, the soft tissue inside bones where blood cells are made. In leukemia, the marrow produces abnormal white blood cells that don't mature properly and crowd out the healthy red cells, platelets, and white cells a body needs.

Doctors split leukemia along two lines. Acute leukemia grows fast, from immature cells, and usually needs urgent treatment. Chronic leukemia develops slowly, from more mature cells, and sometimes doesn't need immediate treatment at all. A second split separates myeloid leukemias, which arise from cells destined to become red cells, platelets, or certain white blood cells, from lymphoid leukemias, which arise from the cells that would normally become lymphocytes, a key part of the immune system. These distinctions matter because the causes behind each type aren't identical. No single trigger explains leukemia. It usually results from some combination of inherited genetics, mutations a person picks up over a lifetime, outside exposures, and simple chance.

The Chromosome Swap That Changed Cancer Genetics

In the early 1970s, researcher Janet Rowley identified something unexpected in chronic myeloid leukemia cells: a piece of one chromosome had traded places with a piece of another. The resulting shortened chromosome became known as the Philadelphia chromosome, and it was the first consistent genetic abnormality ever linked to a specific human cancer. That discovery helped establish that cancer could be understood, at least in part, as a genetic disease playing out inside individual cells.

The swap creates a new, abnormal gene called BCR-ABL1. It isn't inherited; it forms within a single blood cell during that person's lifetime. The gene produces a protein that acts like a switch stuck in the "on" position, pushing the cell to grow and divide without the usual restraints. This kind of change, a mutation that happens in one cell rather than being passed down from a parent, is called a somatic mutation, and it's distinct from mutations present in every cell of the body from birth.

Acute myeloid leukemia has its own common somatic drivers. Mutations in two genes, FLT3 and NPM1, turn up in a meaningful share of cases and influence how the disease behaves and responds to treatment. Researchers first recognized NPM1's role in these leukemias partly through the odd observation that the protein it makes was showing up in the wrong part of the cell. Like the Philadelphia chromosome, these mutations arise during a person's life rather than being inherited.

When Leukemia Risk Runs in Families

Some people start life with a genetic setup that makes leukemia meaningfully more likely. Down syndrome, caused by an extra copy of chromosome 21, is linked to a substantially higher risk of acute leukemia in childhood compared with children who don't have the condition. The exact mechanism connecting the extra chromosome to leukemia risk isn't simple, but the elevated risk itself is well established.

Other leukemia risk is tied to mutations present from birth in every cell of the body, known as germline mutations. Li-Fraumeni syndrome, caused by an inherited change in the TP53 gene, raises susceptibility not just to leukemia but to several other cancers as well. TP53 normally helps cells detect and respond to DNA damage, so an inherited fault in it leaves the body with one less safeguard against cells going wrong. Fanconi anemia works differently but points the same direction: it's an inherited disorder of the systems cells use to repair damaged DNA, and people with it face a markedly elevated lifetime risk of leukemia.

These inherited syndromes are rare. Most people diagnosed with leukemia don't have one of them. But for the small number of families who do carry these mutations, they represent a clear and quantifiable source of elevated risk.

Small Genetic Nudges Versus Rare High-Impact Mutations

Most people's leukemia risk isn't shaped by a single powerful gene at all. Genome-wide studies, which scan the DNA of large numbers of people to spot small differences linked to disease, have found several genetic variants associated with childhood acute lymphoblastic leukemia. Unlike the mutations behind Li-Fraumeni syndrome or Fanconi anemia, none of these variants carries much weight on its own. Each one nudges risk up only slightly.

That distinction matters for how to think about genetic risk generally. A rare, high-impact inherited mutation can dramatically raise an individual's odds. A common variant found through genome-wide scanning does something much subtler: it shifts probabilities slightly across a population, without determining any one person's fate. Most people who carry these common variants never develop leukemia.

Radiation, Chemicals, and the Treatments That Can Cause a Later Leukemia

Outside the body's own genetics, a handful of environmental and medical exposures are established leukemia causes. Ionizing radiation, the high-energy kind found in radiation therapy, certain medical imaging, and nuclear exposure, is strong enough to damage DNA directly and has long been recognized as a leukemia risk factor. Benzene, an industrial chemical found in some fuels and solvents, damages bone marrow cells and raises leukemia risk with heavy, sustained exposure.

A related and somewhat paradoxical cause comes from cancer treatment itself. Certain chemotherapy drugs, particularly alkylating agents and topoisomerase II inhibitors, along with radiation therapy given for an earlier cancer, can damage DNA in healthy blood-forming stem cells. Years later, that damage can give rise to a second, treatment-related leukemia. These drugs work by damaging the DNA of fast-growing cancer cells, which is also how they can occasionally harm normal marrow cells enough to set off a new cancer down the line. This doesn't mean such treatments shouldn't be used; it means the small, delayed risk is a recognized and studied trade-off, not an unexpected failure of therapy.

Viral infection accounts for a small, specific slice of leukemia cases. Infection with human T-cell lymphotropic virus type 1 can lead to adult T-cell leukemia/lymphoma, a rare condition concentrated in certain regions of the world. It is one of the few known infectious causes of leukemia, though the large majority of leukemia cases have no viral trigger at all.

The Blood's Own Aging Process

As people age, individual blood stem cells gradually pick up mutations, the way any cell does over decades of division. Occasionally, one mutated stem cell starts to outcompete its neighbors, expanding until it makes up an unusually large share of a person's blood cells. This state, called clonal hematopoiesis, becomes more common with age and was characterized in large genetic studies, including work led by researcher Siddhartha Jaiswal, around a decade ago.

Clonal hematopoiesis is not leukemia. Most people who have it never develop a blood cancer. But having it does raise the statistical risk of later developing one, including leukemia, which is why researchers now think of it as a precursor state worth understanding rather than a diagnosis to fear.

Smoking and the Limits of Lifestyle Risk

Cigarette smoking shows a modest but real association with acute myeloid leukemia, confirmed across pooled analyses of multiple studies. The size of this effect is far smaller than the risk tied to heavy benzene exposure or therapeutic radiation. It belongs in the picture of leukemia causes, but as a contributing factor among several, not as a dominant one.

What This Adds Up To

Leukemia emerges from several different kinds of risk layered together: rare but powerful inherited syndromes, common genetic variants that barely move the needle individually, somatic mutations that accumulate with age, specific environmental and chemical exposures, occasional viral infection, and the sometimes unavoidable aftereffects of treating an earlier cancer. None of these factors guarantees the disease, and most people who carry one, or even several, never develop leukemia at all.

That's the honest shape of genetic risk in general. It describes probabilities across groups of people, not certainties for any one person. Understanding which factors are well established, like radiation and certain inherited syndromes, and which remain modest associations still being refined, like smoking or common genetic variants, is what lets this picture stay scientifically accurate rather than alarmist.