Impact of Metformin on Cancer Biomarkers in Non-Diabetic Cancer Patients: A Systematic Review and Meta-Analysis of Clinical Trials.

Introduction: Our aim was to investigate and evaluate the influence of metformin on cancer-related biomarkers in clinical trials.

Methods: This systematic study was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. Major databases, including Scopus, Web of Sciences, PubMed, Ovid-Medline, and Cochrane, were systematically reviewed by February 2020. Clinical trials investigating metformin effects on the evaluation of homeostatic models of insulin resistance (HOMA-IR), Ki-67, body mass index (BMI), fasting blood sugar (FBS), and insulin were selected for further analysis. Quality assessment was performed with version 2 of the Cochrane tool for determining the bias risk for randomized trials (RoB 2). Heterogeneity among the included studies was assessed using the Chi-square test. After quality assessment, a random effects model was performed to summarize the data related to insulin, HOMA-IR, Ki-67, and a fixed-effect model for FBS and BMI in a meta-analysis.

Results: Nine clinical trials with 716 patients with operable breast and endometrial cancer and 331 with primary breast cancer were involved in the current systematic and meta-analysis study. Systematic findings on the nine publications indicated metformin decreased insulin levels in four studies, FBS in one, BMI in two, Ki-67 in three studies, and HOMA-IR in two study. The pooled analysis indicated that metformin had no significant effect on the following values: insulin (standardized mean differences (SMD) = -0.87, 95% confidence intervals (CI) (-1.93, 0.19), = 0.11), FBS (SMD = -0.18, 95% CI (-0.30, -0.05), = 0.004), HOMA-IR (SMD = -0.17, 95% CI (-0.52, 0.19), = 0.36), and BMI (SMD = -0.13, 95% CI (-0.28, 0.02), = 0.09). Metformin could decrease Ki-67 in patients with operable endometrial cancer versus healthy subjects (SMD = 0.47, 95% CI (-1.82, 2.75), = 30.1). According to Egger's test, no publication bias was observed for insulin, FBS, BMI, HOMA-IR, and Ki-67.

Conclusions: Patients with operable breast and endometrial cancer under metformin therapy showed no significant changes in the investigated metabolic biomarkers in the most of included study. It was also found that metformin could decrease Ki-67 in patients with operable endometrial cancer. In comparison to the results obtained of our meta-analysis, due to the high heterogeneity and bias of the included clinical trials, the present findings could not confirm or reject the efficacy of metformin for patients with breast cancer and endometrial cancer.