The apoptotic index (apoptotic cells/1000 tumor cells, AI) was evaluated in 71 ovarian carcinomas, all surgically resected. Apoptosis was examined by modified terminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate-biotin nick end-labeling (TUNEL) method in histologic sections. High AI (>/=2.8) significantly correlated with high mitotic index (P = 0.05), high histologic grade of the tumor (P = 0. 018), and short overall survival (P = 0.017). An inverse relationship between AI and bcl-2 protein expression was also observed (P = 0.007). In addition, AI was assessed in 5 ovarian epithelial tumors of borderline malignancy, and all were categorized as low AI (<2.8). No significant correlation was found between AI and other clinicopathologic factors, such as age, clinical stage, lymph node metastasis, tumor size, histology of the tumor, and expression of p53 protein. Multivariate survival analysis showed that only clinical stage (P = 0.0395) and mitotic index (P = 0.0387) had independent prognostic value, whereas AI did not. Our results suggest that counting apoptosis can be useful for predicting the patient survival in ovarian carcinoma, although AI is not an independent prognostic factor. It is also suggested that bcl-2 protein is an important regulator of apoptosis in ovarian carcinoma.
Copyright 1997 Academic Press.