Background: The current "gold-standard" for Parkinson's disease (PD) diagnosis is based primarily on subjective clinical rating scales related with motor features. Molecular biomarkers that are objective and quantifiable remain attractive as clinical tools to detect PD prior to its motor onsets.
Objective: Here, we aimed to identify, develop, and validate plasma-based circulating microRNA (miRNAs) as biomarkers for PD.
Methods: Global miRNA expressions were acquired from a discovery set of 32 PD/32 controls using microarrays. k-Top Scoring Pairs (k-TSP) algorithm and significance analysis of microarrays (SAM) were applied to obtain comprehensive panels of PD-predictive biomarkers. TaqMan miRNA-specific real-time PCR assays were performed to validate the microarray data and to evaluate the biomarker performance using a new replication set of 42 PD/30 controls. Data was analyzed in a paired PD-control fashion. The validation set was composed of 30 PD, 5 progressive supranuclear palsy, and 4 multiple system atrophy samples from a new clinical site.
Results: We identified 9 pairs of PD-predictive classifiers using k-TSP analysis and 13 most differentially-expressed miRNAs by SAM. A combination of both data sets produced a panel of PD-predictive biomarkers: k-TSP1 (miR-1826/miR-450b-3p), miR-626, and miR-505, and achieved the highest predictive power of 91% sensitivity, 100% specificity, 100% positive predicted value, and 88% negative predicted value in the replication set. However, low predictive values were shown in the validation set.
Conclusions: This proof-of-concept study demonstrates the feasibility of using plasma-based circulating miRNAs as biomarkers for neurodegenerative disorders such as PD and shows the challenges of molecular biomarker research using samples from multiple clinical sites.