Proteomic Profiling of Cerebrospinal Fluid and Its Extracellular Vesicles from Extraventricular Drainage in Pediatric Pilocytic Astrocytoma, towards Precision Oncology

Simple Summary Childhood pilocytic astrocytoma (PA) is the most prevalent brain tumor in children, the pathogenesis of which remains incompletely understood. The identification of PA biomarkers is crucial for precise treatment and follow-up in affected patients. Cerebrospinal fluid (CSF), which is in close proximity to the brain and serves as a route for metastases, holds promise as a source for biomarker discovery. Our study entails a proteomic characterization of CSF collected from extraventricular drainage waste, as well as its extracellular vesicles, which express valuable disease targets. In a cohort of 19 PA patients, two proteins (inactive carboxypeptidase-like protein X2 and aquaporin-4) were found to be statistically significantly deregulated compared to controls of 18 children with congenital hydrocephalus (nontumoral controls) and 13 with medulloblastoma (unrelated tumoral controls). These proteins are considered promising potential biomarkers for validation in the blood through subsequent studies into the clinical application of minimally invasive translational screening.Abstract Pediatric pilocytic astrocytoma (PA) is the most common brain tumor in children. Complete resection provides a favorable prognosis, except for unresectable PA forms. There is an incomplete understanding of the molecular and cellular pathogenesis of PA. Potential biomarkers for PA patients, especially the non-BRAF-mutated ones are needed. Cerebrospinal fluid (CSF) is a valuable source of brain tumor biomarkers. Extracellular vesicles (EVs), circulating in CSF, express valuable disease targets. These can be isolated from CSF from waste extraventricular drainage (EVD). We analyzed the proteome of EVD CSF from PA, congenital hydrocephalus (CH, non-tumor control), or medulloblastoma (MB, unrelated tumoral control) patients. A total of 3072 proteins were identified, 47.1%, 65.6%, and 86.2% of which were expressed in the unprocessed total and in its large-EV (LEV), and small-EV (SEV) fractions. Bioinformatics identified 50 statistically significant proteins in the comparison between PA and HC, and PA and MB patients, in the same fractions. Kinase enrichment analysis predicted five enriched kinases involved in signaling. Among these, only Cyclin-dependent kinase 2 (CDK2) kinase was overexpressed in PA samples. PLS-DA highlighted the inactive carboxypeptidase-like protein X2 (CPXM2) and aquaporin-4 (AQP4) as statistically significant in all the comparisons, with CPXM2 being overexpressed (validated by ELISA and Western blot) and AQP4 downregulated in PA. These proteins were considered the most promising potential biomarkers for discriminating among pilocytic astrocytoma and unrelated tumoral (MB) or non-tumoral conditions in all the fractions examined, and are proposed to be prospectively validated in the plasma for translational medicine applications.