The study found that children younger than 5 years old had the greatest risk of developing leukemia after a CT scan of the head – the most common site scanned in children.
The researchers also discovered that for every 5,250 head scans performed for children younger than age 5, one child developed leukemia. The risk for children ages 10 to 14 was 1 in every 21,160 head scans. Miglioretti says that while the lag time for the onset of disease is longer for other cancers – it may take seven years for a girl who had a chest CT to show signs of breast cancer – leukemia may develop as soon as two years after the scan.
Meanwhile a 2012 study by researchers at the Newcastle University's Institute of Health and Society in England found children who received a dose of at least 50 milligray (a measure of absorbed radiation) were 2.8 times more likely to develop brain cancer; children who received at least 30 milligray were at a 3.2 times greater risk of leukemia. Miglioretti and her team estimated that lowering the highest 25 percent of doses to the median level – which varies based on age and the organ targeted – could prevent about 43 percent of childhood cancers.
The radiation studies, as well as ones that document atomic bomb survivors, provide one clue as to why kids develop leukemia and brain cancer. But what about those who have never been in a CT scanner or nuclear fallout zone yet find out they have malignant cells?
Some researchers, like Jun Li, think geographic location might play a role. Li, an epidemiologist with the Centers for Disease Control and Prevention's cancer prevention division, conducted the first study in the U.S. to determine cancer incidence rates by geographic region for children from birth to age 19. Li drew on 2001-2003 data from the National Program of Cancer Registries and the NCI's Surveillance, Epidemiology and End Results statewide registries, which represent more than 90 percent of the country's population.
Based on the 36,446 cancer cases in the study, Li found the Northeast had the highest incidence of all childhood cancers combined, in addition to the highest incidence of brain tumors. The west coast had the highest rate of leukemia.
U.S. News asked Li to evaluate the regions again to include data through 2009 (the most recent year with reported data), and he found the same results. In 2009, Colorado had highest inland rate of leukemia at 5.5 cases per 100,000 (Hawaii's rate was 5.7), while Utah had the highest rate of brain cancer at 4.8 cases per 100,000. (Rates could not be calculated for some states with fewer than 16 cases.)
Map by Heather Brady
Map by Heather Brady
The trouble is, Li is still struggling to explain the variations. "It could reflect the differences in a lot of things such as the genetic background in the population … it may also reflect some variation in the health care system like access to care and insurance, and the larger level would be environmental," he says.
Kim Ritchey, of the Children's Hospital of Pittsburgh of UPMC and president of the American Society of Pediatric Hematology/Oncology, says each state's caseload is so small that minor variations may appear to be significant when they're not. He also says state-by-state comparisons yield no meaningful patterns.
Li holds out hope that future studies, including a CDC trend analysis study of childhood cancer rates from 2001 to 2008, will yield more useful information. "I'm afraid there's no solid conclusion on childhood cancer [causes]," he says. "If we knew, we would share it with you."
There's been plenty of speculation, often centering on environmental factors – such as exposure to toxins, pesticides and cleaning chemicals – or foods a mother ate during pregnancy. Ritchey dismisses those theories, especially as they relate to leukemia. "The bottom line is there's never been any environmental [factor] that's been proven to be a cause of childhood ALL [acute lymphocytic leukemia]," Ritchey says.