No abstract

Background Current epidemiological data show that diabetes, especially type 2 diabetes, is considered to be a complicated factor in the prevalence and prognosis of prostate cancer. Methods The computer retrieves four databases to obtain controlled trials at home and abroad on the efficacy and prognostic effect of metformin in patients with diabetes and prostate cancer. After a rigorous literature quality evaluation, we utilized the RevMan 5.3 software for analyzing the data. Results We included the 8 studies for this comprehensive meta-analysis. These studies were considered to be controlled or array trials. For comparing the test group and the control group, we collected the data recurrence rate from the 5 studies. We revealed that the 5-year recurrence rate was not statistically different (OR: 0.92; 95% Cl: 0.67,1.28; P = 0.631). Similarly, we found the 10-year recurrence rate (OR:0.81; 95% Cl: 0.25,2.58; P = 0.715), 5-year mortality rate (OR:0.66; 95% Cl: 0.14,3.11; P = 0.597), and 10-year mortality rate (OR: 0.44; 95% Cl: 0.26,0.74; P < 0.01). Although we found the asymmetrically distributed effective rate in the funnel plot analysis, publish bias was not potentially found in Egger’s test (P = 0.958). Conclusion This meta-analysis demonstrated that metformin is not potentially effective in patients with diabetes and prostate cancer, as evidenced by the 5-year recurrence rate, 10-year recurrence rate, 5-year mortality, and 10-year mortality, and the above conclusions need to be verified by more high-quality research.