The impact of survivin on prognosis and clinicopathology of glioma patients: a systematic meta-analysis.

Up to now, survivin has been recommended as a prognostic and diagnostic indicator in glioma patients. However, there are still many controversies. Here, a meta-analysis was conducted to draw a more definitive conclusion on the correlation of survivin with overall survival (OS), age, gender, and WHO grade. Eligible studies were available through careful assessment, and then pooled hazard ratios (HRs) or odds ratios (ORs) with 95% confidence intervals (95% CIs) were estimated. Funnel plots were introduced to evaluate the publication bias. Additionally, heterogeneity and sensitivity were also evaluated. In the present meta-analysis, 15 eligible studies with a total of 1,089 patients were incorporated. Survivin expression in gliomas correlated with 2-year OS (n = 8; HR 0.17, 95% CI 0.11-0.26) and 5-year OS (n = 7; HR 0.12, 95% CI 0.07-0.22) in patients. In addition, a fixed-effect model revealed a significant association between survivin and age (male/+; OR 2.10, 95% CI 1.44-3.05) and survivin and WHO grade (I+II/+; OR 0.27, 95% CI 0.19-0.38). No heterogeneity was observed across all studies. According to Begg's and Egger's test and funnel plot, no publication bias was reported. Taken together, our meta-analysis suggests that survivin expression is associated with poor survival, older age, and higher WHO grade and could be suggested as a useful prognostic and diagnostic biomarker, or an effective therapy target.