Public Release: 

Large-Scale Screening For Childhood Cancer

University of Minnesota

Infant screening for neuroblastoma--one of the most common forms of solid tumors in young children--fails to detect the most severe form of the disease, according to a group of international researchers led by University of Minnesota faculty. After screening nearly 500,000 Canadian infants for neuroblastoma, the group found that screening failed to detect the most severe neuroblastomas even though it detected twice as many cancers as had been expected. The researchers conclude that, if left to run their course, some neuroblastomas detected by screening in infancy either go away or fail to progress. Results of the study appear in the December 21 issue of The Lancet.

"Widespread infant screening for neuroblastoma does not save lives and could subject infants to unnecessary cancer treatments," said Mendel Tuchman, one of three principal investigators in the study, who is also a professor of pediatrics and member of the University of Minnesota Cancer Center. He added that no such routine screening programs are currently in place in the United States. Other principal investigators were William Woods, professor of pediatrics and director of the South Carolina Cancer Center at the University of South Carolina, and Bernard Lemieux, professor of pediatrics at the University of Sherbrooke in Quebec. Leslie Robison, holder of the Cancer Center Children's Cancer Research Fund chair in pediatric cancer at the University of Minnesota, and Russell Luepker, professor and head of the division of epidemiology in the University of Minnesota School of Public Health, also were involved in the study.

"When the disease develops before the child is a year old, it is often successfully treated," said Tuchman, whose laboratory developed and performed the screening test. "After that age, it could evolve rapidly into an advanced stage disease with a very high mortality. It is extremely unfortunate that infant screening does not detect the severe neuroblastomas that develop in older infants."

Researchers now believe that there are at least two distinct types of neuroblastoma. Those that appear at birth or shortly thereafter may resolve spontaneously without leaving any trace of the disease. A more advanced form appears in older infants and carries a high mortality. One in 7,000 children is diagnosed with neuroblastoma, making it one of the most common malignant tumors in young children. In 90 percent of cases, the disease affects those under five years of age.

The researchers screened children born between 1989 and 1994 in the province of Quebec for neuroblastoma. Quebec was chosen for the study because it had an established program to screen for metabolic defects at the age of three weeks. Parents submitted urine samples from their children at the ages of three weeks and six months. Researchers analyzed samples from 425,816 3-week-olds (91 percent of those eligible). Of those, samples from 349,706 (75 percent) were again analyzed at 6 months of age. The 118 detected neuroblastoma cases were more than twice the 54 expected when compared to newborns in Ontario and Minnesota, where screening was not performed.

"While screening for cancer has great appeal, these results demonstrate the importance of large-scale studies to make sure we don't do more harm than good and end up wasting millions of dollars," said Robison. Prior to the study, many people had advocated for widespread infant screening in the United States, based on preliminary reports from programs in Japan.

The $6.6 million Quebec Neuroblastoma Screening Project was funded by the National Cancer Institute. The NCI grant was based on the results of a pilot study in the Twin Cities funded by the American Cancer Society. The University of Minnesota, a leader in cancer research since 1911, formed the Cancer Center in 1991 to coordinate cancer-related programs. The university receives more than $20 million a year in research grants from the National Cancer Institute and other federal sources.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.