Unique Circulating MicroRNA Signatures Distinguish Severe Alopecia Areata from Other Inflammatory Skin Diseases

Background: Alopecia areata (AA) is a chronic immune-mediated skin disease that causes non-scarring hair loss. While most cases are mild, a subset of patients progress to severe forms such as alopecia totalis or universalis. Current severity assessments have limitations in predicting disease evolution and require invasive procedures. Objective: To identify circulating microRNAs (miRNAs) associated with severe AA and evaluate their potential as non-invasive biomarkers for diagnosis and stratification. Methods: We performed plasma miRNA profiling in a discovery cohort using TaqMan OpenArray and validated the findings by RT-qPCR in an independent cohort including patients with AA, psoriasis, atopic dermatitis, vitiligo, and healthy controls. Results: We identified 19 miRNAs significantly downregulated in severe AA. Ten miRNAs were technically and clinically validated, showing high specificity for AA compared to other skin conditions. A machine learning model based on the top four miRNAs achieved high classification accuracy (AUC = 0.94). Pathway enrichment and drug repurposing analyses revealed dysregulated immune, metabolic, and signal transduction pathways, identifying potential therapeutic targets including kinase inhibitors and antioxidants. Conclusions: Our study defines a circulating miRNA signature for severe AA, distinguishing it from other inflammatory skin diseases and offering a basis for non-invasive biomarkers and future therapeutic strategies.