Abstract
An analysis which includes the majority of the cases of childhood cancer occurring in Britain over a period of about 20 years suggests that there is a small familial element in the aetiology of these diseases; aggregations within sibships were observed more frequently than would be expected by chance. Possible explanations of these findings are considered. Some, perhaps many, of the cases within such sibships may be due to associations between malignant disease and various genetically determined conditions at a suclinical level or in the heterozygous state. Alternatively, the observed familial aggregations may be attributable to the fact that sibs share a common environment. Childhood cancer in twins is discussed and findings compared with those from the United States. Attention is drawn to a number of interesting combinations of tumours in sibs, particularly brain tumours and bone cancers. The implications of the findings for genetic counselling are discussed; it is emphasized that, though there appears to be an increased risk that sibs of children with malignant disease will also be affected by such diseases, this amounts overall only to a doubling of the general population risk. Whether or not the explanation is a genetic one, the actual magnitude of the risk for such sibs is only about 1 in 300.