Prognostic and clinicopathological significance of survivin expression in bladder cancer patients: a meta-analysis.

Survivin has been widely reported to play a role in diagnosis and prognosis of bladder cancer patients. However, published data on this subject are heterogeneous. Here, we conducted a meta-analysis to obtain a complete evaluation of the association between survivin and recurrence-free survival (RFS), disease-specific survival (DSS), overall survival (OS), and odds ratio (OR) in bladder cancer patients. Published studies on this subject were selected for further assessment by online articles in PubMed, MEDLINE, EMBASE, and OVID databases. Pooled hazard ratios (HR) with 95 % confidence interval (95 % CI) were estimated. Funnel plots were used to evaluate the publication bias. As well, heterogeneity and sensitivity were analyzed. In this meta-analysis, we included 13 studies with the total number of 1,963 patients. Positive survivin expression in bladder cancer was associated with a poor RFS (HR, 1.831; 95 % CI, 1.344-2.49), DSS (HR, 1.721; 95 % CI, 1.477-2.004), or OS (HR, 1.753; 95 % CI, 1.092-2.816) in patients. In addition, a significant association between expression of survivin and age (OR, 0.641; 95 % CI, 0.416-0.987) as well as stage (OR, 0.37; 95 % CI, 0.190-0.750) was revealed. Heterogeneity was observed among the included studies with RFS (x (2) =29.58, p = 0.009, I (2) = 52.7 %), OS (x (2) = 15.67, p = 0.008, I (2) = 68.1 %), and stage (x (2) = 11.97, p = 0.035, I (2) = 58.2 %). There was no publication bias according to Begg's and Egger's tests except for studies with gender. Furthermore, sensitivity analysis obtained the source of heterogeneity and confirmed opposite results of some studies. This study suggests that expression of survivin indicates poor prognosis in older patients and muscle invasive or advanced stage in bladder cancer. Survivin expression could be used in identifying a subgroup of patients with potential to benefit from a targeted therapy against survivin.