RESUMO
BACKGROUND: Despite sensitivity of MRI to diagnose multiple sclerosis (MS), prognostic biomarkers are still needed for optimized treatment. OBJECTIVE: The objective of this paper is to identify cerebrospinal fluid (CSF) diagnostic biomarkers of MS using quantitative proteomics and to analyze their expression at different disease stages. METHODS: We conducted differential analysis of the CSF proteome from control and relapsing-remitting MS (RRMS) patients followed by verification by ELISA of candidate biomarkers in CSF and serum in control, clinically isolated syndrome (CIS), RRMS and progressive MS (PMS) patients. RESULTS: Twenty-two of the 527 quantified proteins exhibited different abundances in control and RRMS CSF. These include chitinase 3-like protein 1 (CHI3L1) and 2 (CHI3L2), which showed a strong expression in brain of MS patients, especially in astrocytes and microglial cells from white matter plaques. CSF and serum CHI3L1 levels increased with the disease stage and CIS patients with high CSF (>189 ng/ml) and serum (>33 ng/ml) CHI3L1 converted more rapidly to RRMS (log rank test, p < 0.05 and p < 0.001, respectively). In contrast, CSF CHI3L2 levels were lower in PMS than in RRMS patients. Accordingly, CSF CHI3L1/CHI3L2 ratio accurately discriminated PMS from RRMS. CONCLUSIONS: CSF CHI3L1 and CHI3L2 and serum CHI3L1 might help to define MS disease stage and have a prognostic value in CIS.