Mitosis identification in diffuse gliomas: implications for tumor grading.

Abstract

BACKGROUND: The histologic differentiation of mitotic figures is a cornerstone of several highly predictive grading systems for gliomas. In some systems, the presence of even a single mitotic figure is sufficient to classify an astrocytoma as high grade. However, the extent of microscopic examination necessary to exclude the presence of significant mitotic activity has not been determined. METHODS: Hematoxylin and eosin-stained slides from 410 astrocytomas and 107 oligodendrogliomas/oligoastrocytomas were reviewed until the first mitosis was identified or 100 400x fields had been reviewed without identification of a mitosis. The number of the field in which the first mitosis was found was correlated with diagnosis, grade, and survival. RESULTS: A review of 50 400x fields was necessary to achieve a >90% sensitivity in identifying a mitosis in a Grade 3 astrocytoma specimen, compared with 20 400x fields in anaplastic (Grade 3 and 4) oligodendroglioma specimens. For Grade 3 astrocytomas, there was a significant independent correlation between survival and the field in which the first mitosis was found (P = 0.02). For the oligodendroglial tumors, there was a strong correlation between the number of fields counted until the first mitosis was found and grade (P < 0.0001). CONCLUSIONS: The evaluation of mitotic activity offers more prognostic information than can be obtained by the simple approach of noting only their presence or absence. Data were acquired regarding the diligence of the microscopic examination necessary to evaluate the presence or absence of mitotic activity. A 1 cm x 0.1 cm needle biopsy contains adequate tissue to evaluate mitotic activity for the purpose of histologic grading; however, this adequacy is dependent on the sample's being representative and composed entirely of cellular tumor.