Long-Term Metformin Effect on Endometrial Cancer Development Depending on Glucose Environment <i>In Vitro</i>

Objectives: Hyperinsulinemia as well as prolonged and elevated estrogen exposure are considered as risk factors for endometrial cancer (EC) development. Metformin, an anti-hyperglycemic and insulin-sensitizing biguanide, displayed anti-proliferative effects in recent studies. In the present study, the effects of long-term exposure of endometrial cancer cells to low and moderate concentrations of metformin on cell viability, proliferation, clonogenicity and migration were investigated under different metabolic conditions. Study Design: EC cell lines HEC-1A and Ishikawa were cultured under normo- (NG, 5.5 mM) or hyperglycemic (HG, 17.0 mM) conditions and treated with metformin (0.01 - 5.0 mM) in the presence of β-estradiol (E2) for 7 d. Results: A concentration-dependent decrease of cellular viability was observed in the MTT and ATP assays after metformin treatment. IC50 values were between 0.7 - 3.7 mM (NG) and 3.0 - 18.3 mM (HG), respectively. A protective effect of glucose on cellular viability was detected only in the ATP assay. Furthermore, metformin (0.5 - 5.0 mM) led to a significant decrease in proliferation by 12% - 55% (NG). However, a decreased proliferation rate was only induced at 5.0 mM metformin (40%) in the presence of high glucose levels in HEC-1A cells, indicating a glucose-related resistance to anti-proliferative metformin effects, which—to a lesser extent—was also observed in Ishikawa cells. Metformin treatment also caused concentration-dependent effects on clonogenicity and decreased the number and size of colonies. In HEC-1A cells, metformin (0.5 - 5.0 mM) reduced the colony formation by 44% - 80% (NG) and 29% - 81% (HG), respectively. Slightly higher metformin concentrations (1.0 - 5.0 mM) were necessary in Ishikawa cells to reduce clonogenicity by 36% - 86% independent of glucose levels. An investigation of migration in the wound healing assay revealed that the % wound closure decreased with increasing metformin concentrations, but independent of glucose levels. After treatment with 5.0 mM metformin, migration was significantly reduced in both cell lines. Conclusion: Our in vitro findings support the hypothesis that metformin has a direct effect on endometrial cancer cell lines and reflects the importance of the local glucose environment, suggesting that metformin may be considered as a potential adjuvant agent in endometrial cancer therapy due to its direct and indirect effects on endometrial development. However, further studies are necessary that confirm the relevance of our data for clinical applications.