Relationship between Ki-67 positive cells, growth rate and histological type of human intracranial tumors.

The proliferation rate of 40 intracranial neoplasms (30 gliomas, 1 hemangioblastoma, 3 meningiomas, 1 neurinoma and 5 brain metastases) was investigated using the monoclonal antibody Ki-67. In eleven of the gliomas recurrences could be observed, and two of them recurred for second time. In total the Ki-67 labelling indices of 53 specimens were investigated. The Ki-67 nuclear antigen was demonstrated in frozen sections by application of the appropriate monoclonal antibodies according to a modified alkaline phosphatase-antialkaline phosphatase (APAAP) technique. The proliferation rate was evaluated by cell count calculation of the staining index. Ki-67-labelled glioma cells varied from 0.2 percent in one meningioma (WHO-grade I) to 9.1 percent in one glioblastoma. In ten glioma recurrences, higher Ki-67 staining indices could be observed than in their primaries, even when the histological grading did not change substantially. In a cerebellar hemangioblastoma, a trigeminal neurinoma and two endotheliomatous meningiomas the fraction of stained nuclei was less than one percent; however, one recurrent transitional meningioma without any histological sign of malignancy showed a staining index of 2.4 percent. The staining indices of five brain metastases of different malignancies ranged from 1.5 percent in a malignant melanoma to 6.1 percent in bronchial carcinoma. In the majority of the cases examined, the percentage of Ki-67 labelled cells was in accordance with the histologic grade of the neoplasm. In general, there was a direct relationship between the number of stained nuclei and the frequency of mitoses (mitotic index) evaluated in hematoxylin-eosin stained frozen sections. Interestingly, the frequency of mitosis and stained nuclei were higher in tumor recurrences than in the primaries. The results of this study imply that immunohistological labelling of the proliferating cell fraction should become an important additional criterion to predict the biological behaviour of human nervous system neoplasms.