Gliomas are the most common primary brain tumors in adults and, despite advances in the understandings of glioma pathogenesis in the genetic era, they are still ineradicable, justifying the need to develop more reliable diagnostic and prognostic biomarkers for this malignancy. Because changes in cerebrospinal fluid (CSF) are suggested to be capable of sensitively reflecting pathological processes, e.g., neoplastic conditions, in the central nervous system, CSF has been deemed a valuable source for potential biomarkers screening in this era of proteomics. This systematic review focused on the proteomic analysis of glioma CSF that has been published to date and identified a total of 19 differentially expressed proteins. Further functional and protein-protein interaction assessments were performed by using Protein Analysis Through Evolutionary Relationships (PANTHER) website and Ingenuity Pathway Analysis (IPA) software, which revealed several important protein networks (e.g., IL-6/STAT-3) and four novel focus proteins (IL-6, galanin (GAL), HSPA5, and WNT4) that might be involved in glioma pathogenesis. The concentrations of these focus proteins were subsequently determined by enzyme-linked immunosorbent assay (ELISA) in an independent set of CSF and tumor cyst fluid (CF) samples. Specifically, glioblastoma (GBM) CF had significantly lower GAL, HSPA5, and WNT4 levels than CSF from different grades of glioma. In contrast, IL-6 level was significantly higher in GBM CF when compared with CSF and, among different CSF groups, was highest in GBM CSF. Therefore, these candidate protein biomarkers, identified from both the literatures and in silico analysis, may have potentials in clinical diagnosis, prognosis evaluation, treatment response monitoring, and novel therapeutic targets identification for patients with glioma.