Contrasting epidemiology of childhood osteosarcoma, Ewing's tumor, and rhabdomyosarcoma.

Marked dissimilarities in the epidemiology of osteosarcoma, Ewing's tumor, and rhabdomyosarcoma indicate differences in their origins. A major clue to the genesis of Ewing's tumor comes not from defining persons at high risk but from the observation that blacks are at unusually low risk. The neoplasm does not aggregate in families and is not part of any known syndrome. No environmental causes have been identified. By contrast, osteosarcoma may be caused by external or internal ionizing radiation, and it aggregated in families with the same tumor or with dissimilar tumors and in certain genetic disorders of bone. In man and in dogs, the frequency of the neoplasm is related to bone mass and growth. Rhabdomyosarcoma of the upper versus the lower limbs seems related to muscle mass. Age peaks in the occurrence of the tumor elsewhere vary with the anatomic site; head and neck tumors develop in early childhood and urogenital tumors both in early years and in adolescence. The sex ratio (male to female) also varies with the site affected. Rhabdomyosarcoma aggregates with certain other tumors in families and overlaps with osteosarcoma in some of these relationships but is distinguished from that tumor by its excessive occurrence in neurofibromatosis.