Primary central nervous system lymphomas (PCNSLs) are mature B-cell lymphomas confined to the central nervous system (CNS). Blood-brain barrier (BBB) dysfunction drastically alters the cerebrospinal fluid (CSF) proteome in PCNSL patients. To reveal the interaction of PCNSL tumors with CNS structures and the vasculature, we conducted a whole-proteome analysis of CSF from PCNSL patients (n = 17 at initial diagnosis) and tumor-free controls (n = 10) using label-free quantitative mass spectrometry. We identified 601 proteins in the CSF proteome using a one-step approach without further prefractionation, and quantified 438 proteins in detail using the Hi-N method. An immunoassay revealed that 70% of the patients in our unselected PCNSL patient cohort had BBB dysfunction. Correlation analysis indicated that 127 (30%) of the quantified proteins were likely increased in PCSNL patients due to BBB dysfunction. After the exclusion of these proteins, 66 were found to differ in abundance (fold-change > 2.0, p < 0.05) between PCNSL and control CSF proteomes, and most of those were associated with the CNS. These data also provide the first evidence that proteomic changes in CSF from PCNSL patients are mainly associated with protein ectodomain shedding, and that shedding of human leukocyte antigen class 2 proteins is a mechanism of tumor-cell immune evasion.

Primary central nervous system lymphomas (PCNSL) account for approximately 2% to 3% of all primary brain tumors. Until now, neuropathological tumor tissue analysis, most frequently gained by stereotactic biopsy, is still the diagnostic gold standard. Here, we rigorously analyzed two independent patient cohorts comprising the clinical entities PCNSL (n = 47), secondary central nervous system lymphomas (SCNSL; n = 13), multiple sclerosis (MS, n = 23), glioma (n = 10), other tumors (n = 17) and tumor-free controls (n = 21) by proteomic approaches. In total, we identified more than 1220 proteins in the cerebrospinal fluid (CSF) and validated eight candidate biomarkers by a peptide-centric approach in an independent patient cohort (n = 63). Thus, we obtained excellent diagnostic accuracy for the stratification between PCNSL, MS and glioma patients as well as tumor-free controls for three peptides originating from the three proteins VSIG4, GPNMB4 and APOC2. The combination of all three biomarker candidates resulted in diagnostic accuracy with an area under the curve (AUC) of 0.901 (PCNSL vs. MS), AUC of 0.953 (PCNSL vs. glioma) and AUC 0.850 (PCNSL vs. tumor-free control). In summary, the determination of VSIG4, GPNMB4 and APOC2 in CSF as novel biomarkers for supporting the diagnosis of PCNSL is suggested.