A MULTIVARIATE-ANALYSIS OF CLINICOPATHOLOGICAL FACTORS FOR PREDICTING OUTCOME IN UTERINE SARCOMAS

This study involved a comprehensive review of the histologic slides of 62 patients who were diagnosed with uterine sarcoma from 1978 through 1988 at a single institution. In addition, DNA content (ploidy level) could be determined from the H & E slides of these tumors using image analysis. Also, 42 of these cases had retrievable cell blocks on which DNA analysis was performed by means of flow cytometry. A linear regression analysis found a high degree of correlation (r = 0.8) between the measurement of the DNA index of these tumors by these two techniques. All cases were retrospectively restaged using the newly adopted FIGO surgical staging criteria which found the following distribution: 22 (35.5%) Stage I, 10 (16.1%) Stage II, 12 (19.4%) Stage III, and 18 (29%) Stage IV. A multivariate analysis of 60 evaluable patients using the Cox proportional hazard model found that surgical staging was the most significant prognostic factor with respect to the endpoint of overall survival (P = 0.00004). Both patient age at diagnosis and mitotic index were independent from surgical staging in predicting outcome. Furthermore, there was a trend suggesting that DNA index also had prognostic value. Of particular interest was that patients with diploid tumors (DNA index, 0.9-1.1) had a 5-year overall survival of 72% and did not approach median survival; however, hyperdiploid tumors (DNA index >1.1) and hypodiploid tumors (DNA index <0.9) were associated with median survivals of 18 and 12 months, respectively. In conclusion, this study supports the use of surgical staging of patients with uterine sarcomas in order to optimally determine their chance for survival. Further biologic investigations which may result in identifying those patients who could benefit from adjunctive treatment are recommended. (C) 1994 Academic Press, Inc.