The "ideal" classification of pediatric central nervous system neoplasms.

The "ideal" classification of pediatric brain tumors will depend on a full characterization of pathologic variables and their influence on prognosis. Currently available information about these variables remains incomplete, even for traditional histologic and cytologic features, and certainly for those studied by the newer methods that will continue to appear. In light of the multiplicity and specialized nature of some of the techniques that are or will become available, it is suggested that specific pediatric central nervous system neoplasms be targeted for study by groups of pathologists according to a protocol that delegates specific studies to specific individuals. In light of the large number of cases required, as well as the need to control for effects of different treatments, such studies are most readily done within the framework of large cooperative clinical trials. Such organizations provide the mechanism to test the biologic significance of pathologic findings that can then be used in the ideal classification.