Abstract
INTRODUCTION: Early detection of Parkinson's disease (PD) remains challenging due to the lack of reliable blood-based biomarkers. While microRNAs (miRNAs) show promise as circulating biomarkers, translating preclinical discoveries to clinically applicable panels requires rigorous validation across platforms and populations. METHODS: We performed temporal miRNA profiling in an acute MPTP mouse model (day 0 vs. day 5, n = 4 per group) using limma differential expression analysis with FDR correction. To address high-dimensional small-sample challenges, we employed global permutation testing and stability selection with elastic net regularization over 2,000 iterations. A compact miRNA panel was derived and validated in three independent human cohorts: GSE16658 (PBMC, n = 32), GSE269776 (serum exosomes 2021, n = 76), and GSE269775 (serum exosomes 2020, n = 100). Performance was assessed using ROC analysis with permutation-based p-values. RESULTS: Seventeen miRNAs showed significant time-dependent changes in MPTP-treated mice (FDR <0.05), with 15 down-regulated and 2 up-regulated at day 5. Stability selection identified a 6-miRNA panel comprising miR-92b, miR-133a, miR-326, miR-125b, miR-148a, and miR-30b. External validation demonstrated consistent discriminative performance across platforms: GSE16658 AUC = 0.696 (p = 0.060), GSE269776 AUC = 0.791 (p < 0.001), and GSE269775 AUC = 0.725 (p < 0.001). DISCUSSION: The signature showed platform-agnostic stability, performing comparably in PBMC and serum exosomes despite biological and technical differences. A 6-miRNA signature derived from acute MPTP response translates effectively to human blood samples, demonstrating reproducible PD discrimination across multiple platforms. The compact panel size and cross-platform compatibility support its potential for clinical biomarker development. By integrating AI-enhanced feature selection and permutation-based validation, this study provides a reproducible framework for biomarker discovery and a foundation for future early detection and precision medicine in Parkinson's disease.