Objective High grade and non-endometrioid endometrial cancers carry a poor prognosis, and the lack of randomized prospective data has led to a wide range of practice regarding adjuvant therapy. The objective of this study was to evaluate the outcomes of different treatment strategies in patients with high-risk, early-stage endometrial cancer.

Methods Patients with high-grade endometrioid, serous endometrial cancer and carcinosarcoma diagnosed between 2000 and 2012 were identified from databases in three gynecologic oncology divisions, in Toronto and in Israel. Adjuvant treatment practices differed across the centers, creating a heterogeneous cohort. A comparison of stage I patients stratified by adjuvant treatment was undertaken. Log-rank tests and Cox proportional hazards models were employed to compare recurrence and survival across treatment groups.

Results 490patients with high risk endometrial cancer were identified, among them 213 patients with stage I disease. Israeli patients received more chemotherapy (41% vs 10% in stage I disease; P<0.001) than patients in Toronto. Chemotherapy was not associated with improved disease-free, disease-specific or overall survival, nor was it associated with fewer distant recurrences (50% vs 54%). Radiation was also not associated with improved recurrence or survival, nor did it affect the pattern of recurrence. On Cox multivariable analysis, neither radiation treatment nor chemotherapy were significantly associated with outcome (HR for recurrence, 0.72 for pelvic radiation (P=0.46) and 1.99 for chemotherapy (P=0.09); HR for death, 0.67 for pelvic radiation (P=0.29) and 1.03 for chemotherapy (P=0.94)).

Conclusions In this retrospective analysis, neither adjuvant radiation nor chemotherapy were associated with improved outcome in stage I, high risk endometrial cancer.