Background: A retrospective clinical analysis of astrocytomas which contained a significant proportion of gemistocytes was carried out in order to evaluate their effect on prognosis, and other factors influencing prognosis.
Method: From 253 consecutive cases of astrocytic tumours in adults, 25 were selected who had more than 20% gemistocytes in every high-power field examined. 9 of these had anaplasia, the remainder did not. They were divided into two groups according to the proportion of gemistocytes; group A, contained more than 60% gemistocytes, and group B, had between 20 and 60% gemistocytes. TUNEL and immunohistochemical staining for PCNA, p53, Ki-67, bcl-2 were performed in the 20 available cases.
Findings: The median follow-up period was 46 months. There were 14 recurrences, with a median time to recurrence of 15 months. Thirteen repeat operations were performed in nine cases, and two cases showed recurring malignant transformation. The overall median survival time following diagnosis was 73 months and the 5-year survival rate was 52%. There were no significant differences in median survival between groups A and B with different proportions of gemistocytes. On the other hand the median survival of the gemistocytic astrocytomas with anaplasia was 25 months, compared with 158 months for those without anaplasia (p=0.0005). The significant impact of anaplasia on survival persisted in both groups. There were no significant differences in immunohistochemical staining between the two groups, with the exception of staining for Ki-67 (means of the two groups: group A 1.40; group B 2.50).
Conclusions: It is suggested that the proportion of gemistocytes does not itself affect prognosis.