Premium
Limitations of the World Health Organization classification of childhood supratentorial astrocytic tumors
Author(s) -
Gilles Floyd H.,
Brown William D.,
Leviton Alan,
Tavaré C. Jane,
Adelman Lester,
Rorke Lucy B.,
Davis Richard L.,
HedleyWhyte Tessa E.
Publication year - 2000
Publication title -
cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.052
H-Index - 304
eISSN - 1097-0142
pISSN - 0008-543X
DOI - 10.1002/(sici)1097-0142(20000315)88:6<1477::aid-cncr28>3.0.co;2-8
Subject(s) - anaplastic astrocytoma , medicine , astrocytoma , context (archaeology) , pathology , brain tumor , homogeneous , classification scheme , glioblastoma , oncology , cancer research , paleontology , physics , information retrieval , computer science , biology , thermodynamics
BACKGROUND In the context of many implied but not rigorously stated histologic feature combinations, the World Health Organization (WHO) classification of astrocytic tumors specifies only the presence or absence of endothelial proliferation, necrosis, and mitosis to distinguish astrocytoma, anaplastic astrocytoma, and glioblastoma multiforme. METHODS The authors examined the effects of these and other reliably recognized histologic features on survival in the Childhood Brain Tumor Consortium (CBTC) sample of 340 children with supratentorial astrocytic tumors. RESULTS Overall, the WHO criteria distinguished only two prognostically distinct classes of astrocytomas. When the specific combinations of the three features were unambiguously designated, three diagnostic categories resulted. These revised diagnostic categories are consistent with WHO guidelines and have significantly different survival distributions. However, neither the original WHO diagnoses nor the revised categories adequately separated these tumors prognostically, because histologic features other than those specified by WHO were significantly associated with improved or worsened survival. CONCLUSIONS Classifications based on small numbers of specified histologic features may not be feasible because they inadequately separate childhood astrocytic tumors into prognostically homogeneous groups. Preferable classification techniques are those that simultaneously account for all reliably recognized histologic features. Cancer 2000;88:1477–83. © 2000 American Cancer Society.