The biomarker potential of cell‐free microRNA from cerebrospinal fluid in Parkinsonian Syndromes

Abstract

BACKGROUND: MicroRNAs are small noncoding RNAs involved in the post-transcriptional regulation of protein synthesis. Extracellular microRNAs are accessible in a stable form in biofluids. OBJECTIVES: The aim was to identify individual microRNAs and/or subsets of microRNAs in CSF with biomarker potential and thus identify specific putative pathophysiological pathways. METHODS: In a two-step exploratory study design of PD, MSA, PSP, and controls, we initially profiled CSF microRNAs in a pilot cohort (n = 40) by screening for 372 microRNAs. Subsequently, we attempted to validate findings in an independent study cohort in CSF (n = 118) and ethylenediaminetetraacetic acid plasma (n = 114). This study cohort encompassed 46 microRNAs, of which 26 were singled out from the pilot cohort, and an additional 20 microRNAs were added based on previous publications. The most accurate diagnostic microRNA classifiers were identified in a multivariable logistic regression model adjusted for age and sex. RESULTS: A set of three microRNAs in CSF discriminated PD and MSA from controls with good diagnostic accuracy by receiver operating characteristics curve evaluation. The microRNAs were for PD versus controls: miR-7-5p, miR-331-5p, and miR-145-5p (area under the curve = 0.88) and MSA versus controls: miR-7-5p, miR-34c-3p, and miR-let-7b-5p (area under the curve = 0.87). The classifier that best distinguished MSA and PD consisted of two microRNAs: miR-9-3p and miR-106b-5p (area under the curve = 0.73). A single microRNA, miR-106b-5p, provided the best discrimination between PD and PSP (area under the curve = 0.85) in the CSF. CONCLUSIONS: Levels of specific trios of CSF-microRNAs discriminate well between α-synucleinopathies (PD and MSA) and controls. The results need to be validated in larger, independent cohorts. © 2018 International Parkinson and Movement Disorder Society.