Objective:

To identify multiple system atrophy (MSA)-specific biomarker (BM).

Background:

The clinical diagnostic criteria for MSA (Gilman, MDS criteria) require the involvement of multiple systems for diagnosis. In the early stages of MSA, the disease presents as a single-system disorder, making it difficult to distinguish from Parkinson’s disease (PD) and spinocerebellar degeneration (SCD). 

Design/Methods:

Cerebrospinal fluid (CSF) samples were obtained from 56 MSA patients, 33 SCD, 71 PD, and 68 healthy controls. Comprehensive proteomic analysis was carried out employing SomaScan^Ⓡ Assay for more than 7,000 proteins. BM candidates were obtained by volcano plots, and their diagnostic performance was assessed using the area under the curve (AUC) of receiver operating characteristics curve. We performed immunostaining for calretinin (CR), a top priority BM, using autopsy brains from controls and patients with MSA and those with other neurodegenerative diseases, including SCD. 

Results:

Volcano plot comparing MSA-cerebellar type (MSA-C) and SCD revealed the highest fold change (log₂FC 1.060) and lowest p-value (p = .001) in CR. CR level enabled accurate discrimination between MSA-C and SCD (AUC 0.970), and MSA-parkinsonian type (MSA-P) and PD (AUC 0.860). The AUC values of patients within 3 years of onset were further improved (MSA-C and SCD, 0.993; MSA-P and PD, 0.921). CR levels showed significant negative correlations with the severity (r = -0.425, p = 0.049) and duration (r = -0.578, p < 0.001) in MSA-C. Immunostaining showed a specific reduction of CR immunoreactivity in the pontine nuclei, pontine transverse fibers, and cerebellar white matter of MSA patients. 

Conclusions:

We identified a novel MSA-specific BM, CSF CR, useful for early diagnosis. The enhanced values in the early stage with subsequent decrease might be explained by the extent of initial cellular damage followed by gradual neuronal loss. These results provide a novel clue to elucidate the pathogenesis of MSA.