Retrospective analysis of prognostic variables and clinical outcomes in surgically staged intermediate risk endometrial carcinoma.

OBJECTIVE: To identify independent prognostic variables for surgically staged intermediate risk endometrial carcinoma as defined by the Gynecologic Oncology Group 99 (GOG99) criteria. STUDY DESIGN: Retrospective study of 239 patients with FIGO stage IB-occult IIB endometrioid type endometrial cancer, who were primarily treated with comprehensive staging surgery. Data were collected on clinicopathological variables, extent of primary surgery, postoperative adjuvant treatment, and patterns of recurrences. Kaplan-Meier survival curves were used to estimate disease free survival (DFS) and overall survival (OS), and multivariate Cox regression models were used to identify independent prognostic variables. The median follow-up time was 67 months (range, 12-183 months). RESULTS: The 5-year DFS and OS were 91.0% and 93.0%, respectively. On univariate Kaplan-Meier analysis, age>60 years, deep myometrial invasion (MI), presence of lymph vascular invasion (LVSI), and negative progesterone receptor (PR) status were significantly associated with diminished 5-year DFS and OS. The univariate analysis on patterns of failures demonstrated that patients with older age or positive LVSI were more inclined to develop locoregional recurrence, while PR status and the depth of MI had a statistically significant impact on distant failure. On multivariate analysis, PR status, age, and the depth of MI were independent prognostic variables for 5-year DFS, and age was the only independent prognostic variable for 5-year OS. LVSI and age were independent prognostic variables for locoregional recurrence, while PR status and depth of MI were independent prognostic variables for distant recurrence. CONCLUSIONS: Age, depth of MI, PR status and presence of LVSI are of independent prognostic value for intermediate risk endometrial cancer. The presence of these variables warrants consideration when deciding upon treatment strategies.