Neuroinflammation and glycosylation-related cerebrospinal fluid proteins for predicting functional decline in amyotrophic lateral sclerosis: a proteomic study.

Abstract

Listen

Translate article

The rate of disease progression varies widely among patients with amyotrophic lateral sclerosis (ALS). Prognostic assessment using biomarkers is highly anticipated to improve clinical trial design. We aimed to explore the cerebrospinal fluid (CSF) for prognostic biomarkers to predict future functional decline in patients with ALS. We collected CSF samples from 64 patients with ALS and 25 disease controls. The prospective progression rate was calculated using the Revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R) at CSF collection and in 6 months. The ALS patients were classified into slow, intermediate, and fast progression groups. We performed comprehensive proteomic analyses of the CSF samples. Factors with significant changes between slow and fast progression groups were investigated via receiver operating characteristic curve analyses. Moreover, the correlation of the CSF factors with progression rate was evaluated by multiple regression analyses. In total, 26 proteins changed significantly (p < 0.05 and q < 0.10), with levels varying within a large dynamic range (fold change of >1.5 or < 0.5). A receiver operating characteristic curve analyses showed that the following proteins showed high discrimination power between slow and fast progression groups: glycoprotein non-metastatic melanoma protein B (GPNMB; area under the curve [AUC], 0.88), glial fibrillary acidic protein (AUC, 0.81), glypican-1 (GPC1; AUC, 0.79), alpha-1,6-mannosyl-glycoprotein 2-beta-N-acetylglucosaminyltransferase (AUC, 0.74), and chitinase-3-like protein 2 (CHI3L2; AUC, 0.73). Of these, GPNMB, GPC1, and CHI3L2 were significantly correlated to prognostic progression rate. This study demonstrated that CSF levels of neuroinflammation and glycosylation-related proteins were significantly correlated with prospective progression rates in patients with ALS. These proteins could be useful prognostic biomarkers for ALS.