Prognostic Value And Clinicopathological Differences Of Bmi1 In Gastric Cancer:A Meta-analysis

Background: B cell-specific Moloney murine leukemia virus integration site 1(Bmi1) was identified as a biomarker of cancer stem cells, and over-expression of Bmi1 might enhance tumor aggressive clinical behavior in gastric cancer(GC). Our aim of this meta-analysis is to investigate the prognostic role and clinicopathological differences of Bmi1 in GC patients.Methods: Up to September 2014, we performed a comprehensive literature search of the Pub Med, Embase, Springer and Science Direct databases. A meta-analysis method was conducted with eligible studies to evaluate the relationship of Bmi1 expression with clinicopathological features and prognosis of patients with GC. OR with 95%CI was used to evaluate the relationships between Bmi1 expression and its potential related clinicopathological factors. RR with 95%CI was used to estimate the associations between Bmi1 expression and overall survival(OS). The Q-test and P-value were used to evaluate the heterogeneity among studies. Publication bias was assessed using Begg’s funnel plot and homologous P-values. Sensitivity analysis was used to evaluate whether any one study had an obvious impact on the overall results.Results: A total of 6 studies were included in our study by searching relevant databases. Our results showed that there were no relationships between Bmi1 expression and the gender(pooled OR=0.87, 95%CI=0.66-1.14, P=0.319, fixed effect), age(pooled OR=1.22, 95%CI=0.95-1.59, P=0.126, fixed effect) and differentiation(pooled OR=1.15, 95%CI=0.71-1.86, P=0.582, random effect) in GC patients. But high Bmi1 expression was significantly correlated with the clinical stage(pooled OR=3.04, 95%CI=1.31-7.07, P=0.010, random effect), tumor size(pooled OR=2.01, 95%CI=1.14-3.55, P=0.016, random effect), T classification(pooled OR=2.79, 95%CI=1.94-4.03, P<0.001, fixed effect), lymph node metastasis(pooled OR=2.24, 95%CI=1.47-3.39, P<0.001, random effect) and distant metastasis(pooled OR=5.05, 95%CI=1.29-19.70, P=0.020, random effect), and led to a poor OS in GC patients(RR=3.38, 95%CI=2.43-4.69, P<0.001, fixed effect).Conclusion: Bmi1 could affect the clinical stage, tumor size, T classification, lymph node metastasis, distant metastasis and OS in GC patients. These findings suggested that Bmi1 might serve as a novel and effective prognostic biomarker in GC, and could be a promising emerging molecular target in GC therapy.