Children get cancer for fundamentally different reasons than adults do. While adult cancers typically result from decades of accumulated damage from smoking, sun exposure, or other lifestyle factors, childhood cancers arise mostly from random errors during the rapid cell growth and development that happens before and shortly after birth. In the United States, roughly 9,550 children from birth to age 14 will be diagnosed with cancer in 2025, and the most common types are leukemias, brain tumors, and lymphomas.
How Childhood Cancer Differs From Adult Cancer
Adult cancers accumulate hundreds or thousands of mutations over a lifetime. Childhood cancers are strikingly different. Pediatric solid and brain tumors carry six to 16 times fewer mutations than their adult counterparts. Instead of many small DNA changes piling up, childhood cancers are often driven by large structural rearrangements in the genome, where big chunks of DNA get shuffled, deleted, or fused together. These structural variants account for over 60 percent of the genetic changes that drive pediatric cancers.
This distinction matters because it explains why children’s cancers can’t be blamed on environmental exposures the way many adult cancers can. A child simply hasn’t lived long enough to accumulate that kind of damage. Instead, something goes wrong during the frantic pace of cell division that builds a human body from a single fertilized egg into trillions of specialized cells.
Errors During Fetal Development
Many childhood cancers begin before a child is even born. As a fetus develops, cells multiply rapidly and transform into specialized types: brain cells, kidney cells, immune cells. Occasionally, some of these embryonic cells fail to fully mature or end up in the wrong place. They persist after birth as leftover fetal cells that retain their ability to keep dividing, and if additional genetic errors occur, they can become cancerous.
This is the origin of a category called embryonal tumors, which includes some of the most recognizable childhood cancers. Medulloblastoma, the most common malignant brain tumor in children, forms from embryonic cells that remain in the brain after birth. Wilms tumor arises from kidney cells that didn’t finish developing. Neuroblastoma grows from immature nerve cells. Retinoblastoma starts in developing cells of the retina. These cancers are virtually nonexistent in adults because adult bodies no longer contain these types of immature cells.
The Role of the Immune System’s Own Machinery
One of the more surprising causes involves the immune system itself. In developing immune cells, a molecular tool called RAG acts like a pair of scissors, cutting and rearranging DNA segments to create the enormous diversity of antibodies and immune receptors a child needs. This process is essential for building a working immune system, but it’s inherently risky. Sometimes RAG cuts in the wrong spot, disrupting genes that control cell growth.
Researchers have identified recurring patterns of these misdirected cuts in pediatric acute lymphoblastic leukemia, the most common childhood cancer. In children, RAG-mediated errors can disrupt both immune system genes and cancer-driving genes simultaneously. In adults with lymphoid cancers, the same machinery tends to damage only immune genes. This helps explain why leukemia is so disproportionately common in young children: their immune systems are actively under construction, and the construction process itself carries cancer risk.
Inherited Genetic Predisposition
About 10 percent of children diagnosed with cancer have an underlying genetic predisposition syndrome, a condition they were born with that significantly raises their cancer risk. Some of these are inherited from a parent, while others arise as brand-new (de novo) mutations that weren’t present in either parent’s DNA.
Several well-characterized syndromes fall into this category. Li-Fraumeni syndrome, caused by mutations in the p53 gene (one of the body’s most important tumor suppressors), dramatically increases risk for multiple cancer types starting in childhood. Beckwith-Wiedemann syndrome, an overgrowth condition, raises the risk of Wilms tumor and other embryonal cancers. Noonan syndrome and related conditions affecting growth-signaling pathways increase susceptibility to certain leukemias. DICER1 mutations predispose children to rare tumors of the lungs, kidneys, and other organs.
In retinoblastoma, the genetic picture is especially clear. Studies have found germline mutations in the RB1 gene in 24 to 52 percent of cases, and a large share of those mutations are de novo, meaning the child is the first in the family to carry them. For osteosarcoma (bone cancer), about 5 percent of cases involve a germline p53 mutation, with nearly half arising de novo. These numbers illustrate an important point: having no family history of cancer doesn’t rule out a genetic predisposition.
Epigenetic Changes Without DNA Mutations
Not all the changes that drive childhood cancer involve alterations to the DNA sequence itself. Some involve epigenetic changes, which affect how genes are read and expressed without changing the underlying genetic code. Think of it as the difference between changing the words in a recipe versus covering some of them with tape. The recipe is intact, but instructions are missing.
Epigenetic disruption is a recurring theme across several pediatric brain tumors. In certain aggressive childhood gliomas, a single mutation alters a protein that packages DNA, which in turn silences or activates hundreds of other genes at once. In some infant ependymomas (tumors of the brain and spinal cord lining), epigenetic alterations alone, without many traditional DNA mutations, are sufficient to drive lethal cancers. This is part of why pediatric brain tumors can be so different from adult brain tumors even when they appear in the same location.
Environmental Factors
Compared to adult cancers, the evidence linking environmental exposures to childhood cancer is much weaker and more uncertain. High-dose ionizing radiation is the best-established environmental risk factor, but most children are never exposed to doses high enough to matter. At the lower doses encountered in everyday life, the role of radiation remains debated.
Pesticide exposure has been investigated across several childhood cancer types, and some studies suggest a link, particularly for leukemia and brain tumors. Living near high-traffic roads or high-voltage power lines has also been explored as a potential leukemia risk factor. However, the evidence for all of these remains too inconsistent to translate into specific prevention guidelines.
One intriguing hypothesis involves delayed exposure to common infections. Children who don’t encounter typical childhood infections early in life may have an immune system that responds abnormally when those infections finally arrive, potentially triggering leukemia in children who already carry a genetic first hit. This could partly explain why leukemia rates tend to be higher in more affluent, more hygienic environments, though the theory is still being tested.
Prenatal and Early Life Factors
Researchers have looked closely at whether a mother’s diet or habits during pregnancy influence childhood cancer risk. A review by the International Agency for Research on Cancer found a possible association between high maternal caffeine intake and increased risk of childhood acute leukemia, though the link is far from proven. The WHO already recommends pregnant women limit caffeine to under 300 mg per day for other reasons (miscarriage risk and low birth weight), not because of cancer concerns.
One early-life factor does show a consistent protective effect: breastfeeding. Multiple studies have found abundant evidence that breastfeeding reduces the risk of childhood acute leukemia. The mechanism likely involves its role in supporting healthy immune system development during a critical window. On the other hand, researchers have thoroughly debunked the once-circulating concern that vitamin K injections given to newborns might cause leukemia. There is no convincing evidence of any such link.
Screening for Children at Higher Risk
Because most childhood cancers arise from random developmental errors rather than preventable exposures, there’s no broad screening program the way there is for adult cancers like breast or colon cancer. Screening is reserved for the roughly 10 percent of children who have a known genetic predisposition syndrome.
For these children, expert guidelines developed by the American Association for Cancer Research recommend surveillance when the risk of developing cancer in a given age window reaches 5 percent or higher, though some conditions warrant monitoring at risk levels as low as 1 percent. The specific surveillance plans vary by syndrome and can include regular imaging and blood tests starting in infancy. These protocols are complex enough that guidelines recommend families be connected to specialized centers with expertise in pediatric cancer genetics, even if routine monitoring happens closer to home.
For the majority of childhood cancers that arise without a known predisposition, the most important factor is recognizing symptoms early. Persistent unexplained fevers, unusual lumps, sudden vision changes, unexplained weight loss, or persistent headaches with vomiting are among the warning signs that warrant prompt medical evaluation.