The reported rise in the incidence of brain tumors in American children is most likely a result of better diagnosis and reporting rather than a true increase in cases, according to analyses by National Cancer Institute (NCI) researchers published today. Between 1973 and 1994, the frequency of brain tumor diagnoses in children (incidence) rose 35 percent.

Using mathematical models, NCI researchers showed that brain tumor rates did not increase steadily from 1973 to 1994, but "jumped" to a steady, but higher rate after 1984 to 1985. This jump occurred at the time that magnetic resonance imaging (MRI) became available in the United States. MRI gives pictures of the brain using a powerful magnet linked to a computer. Evidence that the increased use of MRI coincided with the rise in incidence include:

• In 1982, there were only 10 MRI units in the United States, but by the end of 1985 there were 371;
• Scientific papers describing the superiority of MRI for imaging childhood brain tumors appeared first in 1984;
• Children are more likely to be seen at large medical centers, which were most likely to have MRI technology at the earliest spread of the technology; and,
• Insurance reimbursement for MRI was limited until 1985, after which third-party payers, including Medicare, approved its use.

The researchers also looked at the type of brain tumors that were being diagnosed, comparing them before and after the 1984 to 1985 time frame. After 1985, they found more low-grade (slow growing or less aggressive) gliomas in the cerebrum and the brain stem - tumors that have since been shown to be more easily diagnosed by MRI. Additionally, from 1973 to 1994, the death rate from childhood brain tumors did not go up, but actually decreased slightly, another indication that the increased incidence in brain tumors is a difference in diagnosis, not a true increase in cases.

While the reported analyses do not rule out the possibility that environmental agents may influence risk for the disease, they suggest that such factors are likely to be responsible for little, if any, of the observed increase in brain tumor incidence. Direct radiation treatments to the head are the only established risk factor for the disease.

"While this research shows that the incidence of brain cancers in children does not appear to be rising, these tumors remain a major cause of cancer-related death in children," said Malcolm Smith, M.D., Ph.D., head of the Pediatric Section of NCI's Clinical Investigations Branch and lead author of the study. Brain tumors are the second most frequent type of cancer in children, ranking only behind the leukemias. "NCI remains committed to finding the causes of brain tumors, as well as to identifying superior treatments and methods of detection."

NCI is working to establish a Pediatric Brain Tumor Clinical Trials Consortium. The Consortium will include pediatric brain tumor experts from approximately eight institutions that will work together to evaluate innovative treatment approaches for children with brain cancers. These institutions are being selected in a competitive process that will be completed by the Spring of 1999. NCI is also working with other government agencies to develop a National Network for Research on Cancer in Children as part of a presidential initiative on Children's Health and the Environment. The Network will be a national resource to support definitive studies on the causes of childhood cancers.

The report is published in the Sept. 2 issue of the Journal of the National Cancer Institute, and uses data from NCI's Surveillance, Epidemiology, and End Results (SEER) Program. The SEER Program collects cancer incidence data from five state-wide and four metropolitan-area registries, representing 9.5 percent of the U.S. population. SEER uses cancer mortality data from the National Center for Health Statistics, which includes 100 percent of the United States. Primary malignant brain tumors diagnosed in children ages 0 to 14 from 1973 to 1994 were analyzed in this report.