Objectives Diabetes mellitus is a risk factor for the development of endometrial hyperplasia and endometrial carcinoma (EC). We aimed to evaluate the association between metformin use and oncologic outcome in diabetic women with EC.

Methods A retrospective multi-center cohort study of diabetic women with EC treated in nine gynecologic oncology centers between 2000–2014. Univariate, Kaplan-Meier survival and Cox proportional hazard model analyses were performed to compare survival outcomes between women treated with metformin and those who were not.

Results A total of 577 diabetic women with EC were included, 330 (57.2%) were treated with metformin and 247 were not. There was no difference between the groups in terms of age, hypertension, statin use, hormonal replacement therapy use, disease stage, grade, lymphovascular space invasion (LVSI) or median follow up time. Women treated with metformin were more likely to have positive abdominal fluid cytology and be operated by minimally invasive route (Odds Ratio [95% Confidence Interval]: 2.8 [1.1–7.2] vs.1.6 [1.1–2.3], respectively). Median follow up was 53 months (interquartile range 20–91). Recurrence rate did not differ between study groups (p=0.267). Cox proportional hazards model adjusted for age, disease stage, grade, LVSI, radiation therapy and chemotherapy, demonstrated comparable progression free survival and overall survival between diabetics who used metformin versus those who did not (p=0.486, p=0.194, respectively).

Conclusions Metformin use did not influence prognosis in diabetic women with EC. Large prospective studies to elucidate the association of metformin and oncological outcomes in diabetic subgroups of women with EC are of need.