Demographic and clinico-pathological characteristics in patients with invasive breast cancer receiving metformin.

There was a contradictory data with metformin use on breast cancer risk, but there is growing evidence that the use of metformin in diabetic patients was associated with lower risks of breast cancer mortality and incidence. The effect of metformin on clinical and pathological properties of breast cancer was not known exactly, and we aimed to investigate the demographic and clinico-pathological characteristics of patients with metformin users at the time of breast cancer diagnosis. Patients with breast cancer diagnosed from 2000 to 2012 in our clinic were retrospectively analyzed. Patient's demographics, including survival data and tumor characteristics were obtained from medical charts. Breast cancer patients who were taking metformin at the time of breast cancer diagnosis were enrolled as metformin users (n = 148), where the patients matched with the same age who were not taking metformin were included as a control group (n = 636). A total of 784 patients were included in this study. Median age of both metformin users and nonusers was 57 (23-87). There were no significant differences in baseline tumor size (P = 0.60), tumor stage (P = 0.76), and node positivity (P = 0.13) between the two groups. Metformin user patients compared to nonusers had significantly lower incidence of histological grade III tumor (P = 0.03). A similar significant trend for lower incidence of triple negative (P = 0.01) and higher incidence ER positivity (P = 0.008) and PR positivity (P = 0.01) was also seen in metformin users. In survival analysis, the estimated median disease-free survival was 118 months in metformin users, whereas 69 months in nonusers (P = 0.09). Median overall survival (OS) could not be obtained due to low events. In patients with metformin users, OS rate was 98.4, 97.1, and 93.8 %, and in nonusers was 99.6, 94.4, and 90.5 %, the first, third, and fifth years, respectively. The use of metformin at the time of breast cancer diagnosis was associated with better clinico-pathological properties and nonsignificantly improved disease-free survival in breast cancer patients.