Multivariate independent prognostic factors in endometrial carcinoma: A clinicopathologic study in 181 patients

Abstract

The aim of this study was to evaluate the biologic outcome of endometrial carcinomas as compared to clinical and pathologic parameters and to identify multivariate independent prognostic factors. Charts were abstracted from patients with endometrial carcinoma from 1985 to 1995. Data on clinicopathologic variables, adjuvant treatment, site of recurrence, and survival were collected. χ2 test was used to test association between variables. Kaplan-Maier method was used for survival analysis and Cox proportional hazards model for multiple regression analysis. Univariate analysis revealed that FIGO stage, tumor grade, depth of myometrial invasion, biochemical analysis of progesterone receptor status, age, additional diabetes mellitus, lymph node metastasis, and type of tumor were significantly associated with the overall-survival. For disease-free interval, FIGO stage, tumor grade, depth of myometrial invasion, biochemical analysis of progesterone receptor status, lymph node metastasis, and type of tumor were also significantly associated. Multivariate analysis revealed that FIGO stage, tumor grading, tumor type, depth of myometrial invasion, and biochemically measured progesterone receptor status were associated significantly with overall survival. A significant correlation as independent prognostic factors were also seen for recurrence free interval for FIGO stage, tumor grade, and biochemical progesterone receptor status. In multivariate statistical analysis we identified FIGO stage, tumor type, tumor grade, biochemical analysis of progesterone receptor status, and depth of myometrial invasion as independent prognostic factors for overall survival, and FIGO stage, biochemical analysis of progesterone receptor status, and tumor grade as independent prognostic factors for recurrence-free interval.