Some Variations in Childhood Cancers Throughout the World

111

Data from the first four volumes of Cancer Incidence in Five Continents (CI-5) and from the first 5 years of the US Surveillance, Epidemiology and End Results (SEER) program were analyzed for evidence of geographical and temporal variations in the incidence of selected childhood tumors. Only lymphoid leukemia and glial neoplasms are common enough for the observed differences between US registries to be distinguished from sampling variation. Internationally, kidney and eye tumors and leukemia show less geographical variation than do lymphomas and brain tumors, but for none of the tumors examined is the incidence constant. Wilms' tumor rates among Japanese, Singapore Chinese and Indians (Bombay) are approximately 60% of the rates in North America and Britain, whereas in Scandinavia the rates are up to 30% higher. This lessens the status of Wilms' tumor as an "index tumor" of childhood. Areas or countries with especially high or low rates of other tumors are identified. Rates for glial neoplasms (SEER data) and Hodgkin's disease (CI-5) are increasing with time in the US, while brain tumors are being diagnosed more frequently worldwide. However, the results for brain tumors may largely reflect changes in pathology diagnosis or reporting practices, and those for Hodgkin's disease may reflect improvements in case ascertainment. Otherwise, there is a remarkable stability in the incidence of selected childhood cancers over time.

Childhood cancer incidence: Geographical and temporal variations

Breslow

Langholz

1983

111

The international incidence of childhood cancer

Stiller²,

Draper³

et al. 1988

632

484

The International Agency for Research on Cancer has coordinated a worldwide study of the incidence of cancer in childhood. Contributors from over 50 countries have provided data. This paper presents a summary of some of the major results. The incidence rates and relative frequencies of childhood cancers are described according to 12 diagnostic groups, defined mainly in terms of tumour morphology. Variations in the risk of those tumours between different countries and different ethnic groups provide important information on the relative importance of environmental and genetic factors in their aetiology.

International variations in the incidence of neuroblastoma

Stiller

Parkin

1992

119

The International Agency for Research on Cancer has coordinated a worldwide study of childhood cancer incidence, with data from over 50 countries. We present here the results for neuroblastoma. In predominantly white Caucasian populations the age-standardized rate was 7-12 per million, and 6-10% of all childhood cancers were neuroblastomas. Rates were highest in the first year of life (25-50 per million, 30% of total neuroblastoma incidence), and decreased with age to 15-20 per million (50% of the total) at age 1-4, 2-4 per million (15%) at 5-9 and 1-1.5 per million (5%) at 10-14. In the United States, black children had an incidence of 8.5 per million compared with 11.5 among Whites; Blacks tended to be older than Whites at diagnosis. The highest rate in Africa was in Bulawayo, Zimbabwe (8.0 per million) and the lowest in West Nile, Uganda, with no cases registered. Incidence in Israel was similar to that in many white populations, with Jews having a particularly high rate. In other parts of West Asia neuroblastoma had a low relative frequency, suggesting that incidence is low. Rates were also low throughout much of southern and eastern Asia, including India and China. Incidence in Japan was somewhat higher, though less than in Western countries, with the deficit most pronounced in the first year of life; these data relate to the period before mass screening of infants for neuroblastoma in the regions concerned. Incidence was generally higher in regions and among ethnic groups enjoying a higher standard of living, though previous studies within single countries had suggested that neuroblastoma is more common among less affluent groups. Blacks in Africa and the United States may have a weaker genetic predisposition to neuroblastoma, but some of the deficit in many developing countries is likely to be due to under-diagnosis.