Untargeted Metabolomics of Cerebrospinal Fluid in Patients with Primary Central Nervous System Lymphoma

Primary central nervous system lymphoma (PCNSL) is characterized by a high degree of malignancy and poor prognosis. The current diagnostic techniques include cerebrospinal fluid (CSF) metabolomics examination, imaging examination, needle biopsy, etc., but each examination has certain limitations, and it is difficult for the early identification and diagnosis of the disease. In terms of treatment, relapse and refractory, neurotoxicity and drug resistance are still the main problems currently faced. Metabolomics can measure the composition of metabolites and the dynamic changes of metabolites in cell models, tissues, organs or the whole organism, and has been widely used in disease diagnosis, efficacy evaluation, nutrition and other fields. Ultra-high performance liquid chromatography coupled with high-resolution mass spectrometry (UPLC-HRMS/ technology) has been be used to observe the metabolites changes in CSF. Here, we aimed to explore the application value of metabolomics in the diagnosis and prognosis evaluation of PCNSL. From January 2021 to December 2021, CSF samples from 37 DLBCL patients were collected in Shengli Oilfield Central Hospital and Shandong Cancer Hospital, including 20 male patients and 17 female patients. These patients fall into 4 groups based on clinical presentations: 12 patients in PCNSL group (CNS group), 5 patients in primary testicular diffuse large B-cell lymphoma (DLBCL) group (TE group), 10 patients in Germinal center B (GCB) cell-like DLBCL without central nervous system infiltration (GCB group), and 10 patients in non-GCB DLBCL group without central nervous system infiltration (non-GCB group). The CSF samples were collected before clinical treatment, and the samples were analyzed with UPLC-HRMS/technology. For data analysis, principal component analysis (PCA) was carried out on the obtained components, and the partial least squares discriminant analysis model (PLS-DA) was used for diversification analysis. Analytes with P<0.05 and VIP>1 were considered differential metabolites. The Venn diagram was used to narrow the range of metabolites. By using MetaboAnalyst 5.0 software, pathway enrichment analysis was performed on the identified differential metabolites in CSF, and the corresponding P value was obtained. The metabolic pathways that have a significant effect in CNS group vs TE group, CNS group vs GCB group and GCB group vs non-GCB group were glycerophospholipid metabolism and sphingolipid metabolism. The metabolic pathway that has a significant effect in CNS group vs non-GCB group and TE group vs non-GCB group was glycerophospholipid metabolism. Fatty acid degradation and fatty acid elongation were the significant effects metabolic pathway in TE group vs non-GCB group. We applied Venn diagrams to narrow the range of differential metabolites. There were 47 differential metabolites in the CSF of patients in the PCNSL group vs the non-PCNSL group (Figure 1). Compared with the non-PCNSL group, 33 metabolites such as 3-Hydroxyisovalerylcarnitine, Hesperetin 3'-O-sulfate, and Arteether were up-regulated and 14 metabolites such as Pentaglutamyl folate, Calcipotriol and N-Hexadecanoylpyrrolidine were down-regulated in the CSF of patients in the PCNSL group. In the CNS infiltration high-risk group and the CNS infiltration low-risk group, there were 16 differential metabolites (Figure 2). Compared with the patients in the CNS infiltration low-risk group, 8 metabolites such as 3-Allylphenol sulfate, p-Aminobenzoic acid and 2,3,4-Trichlorophenol were up-regulated and 8 metabolites such as Palmitaldehyde, Carboxamide and Rioprostil were down-regulated in the CNS infiltration high-risk group. In conclusion, glycerophospholipid, sphingolipid, and fatty acid metabolites were the three most important abnormal lipid metabolites existed in PCNSL. This finding may have some value in the development of diagnosis and prognostic tools for evaluating CSF in patients with PCNSL. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal