Circulating tumor cells (CTCs) and disseminated tumor cells (DTCs) displayed their roles in prognosis prediction in prostate cancer. The objective of the present study was to conduct a systematic review and meta-analysis of published literature while investigating the correlation between survival outcome and CTCs or DTCs counts in patients with prostate cancer. Relevant literature was searched in Pubmed and Embase. Survival data of included study were extracted. Forrest plots were used to estimate the effect of CTCs/DTCs on the survival of patients. Publication bias was evaluated using Begg's test. The estimated HRs and 95 % confidence interval for the effect of CTCs/DTCs on overall survival (OS) and biochemical relapse-free survival (bRFS) or disease-free survival (DFS) were 2.43 [2.07, 2.86] (p<0.00001) and 2.15 [1.69, 2.73] (p<0.00001), respectively. Subgroup analysis revealed that CTCs were also relevant to poor prognosis (hazard ratio (HR) 2.43 [2.05, 2.89] for OS, HR 2.46 [2.08, 2.90] for bRFS/DFS). A similar result was yielded in DTCs (1.47 [1.21, 1.80] for DFS). CTCs/DTCs could also predict poor OS in metastatic prostate cancer (2.37 [1.99, 2.82], p<0.00001) and in localized stage (HR 1.84 [1.47, 2.28], p<0.00001). In addition, CTCs/DTCs detected by different methods, especially by CellSearch system (HR for OS 2.36 [1.95, 2.85] and HR for bRFS/DFS 2.53 [1.66, 3.85]), were relevant to poor prognosis. Available evidence supported the notion of the strong prognostic value of CTCs. CTCs are promising biomarkers that are clinically implemented in the therapeutic decision-making process in patients with prostate cancer.