Transcriptomic meta-analysis and deconvolution approaches unveil cellular dynamics in scarring and non-scarring primary lymphocytic alopecias

Inflammatory alopecia encompasses various hair loss disorders with immune-mediated inflammation and scarring. Methods: Through a systematic review of publicly available Gene Expression Omnibus (GEO) studies, we investigated studies involving mircroarrays gene expression analysis on scalp samples in patients with Alopecia Areata (AA), Frontal Fibrosing Alopecia (FFA), Lichen Planopilaris (LPP), and Central Centrifugal Cicatricial Alopecia (CCCA). Through transcriptomic analysis and deconvolution of bulk gene expression data sets, we estimated the proportions of immune cells in each sample. Results: A meta-analytic model was used to integrate gene expression profiles from eight independent GEO datasets (GSE68801, GSE45512, GSE80342, GSE58573, GSE74761, GSE59131, GSE58934, and GSE113052) involving patients with non-scarring alopecia (AA, n=82) and scarring alopecia (FFA, n=10; (LPP, n=11; CCCA, n=8), and healthy controls (n=80). Distinct cellular changes were observed in each subtype. AA showed increased class-switched memory B cells, macrophages, mast cells, and certain T cell subsets, with decreased B cell plasma and T cell CD8+ naïve cells. FFA displayed increased class-switched memory B cells, endothelial cells, granulocyte-monocyte progenitors, and T cell CD4+ central memory, with decreased B cell plasma and T cell CD4+ Th1. LPP demonstrated increased B cells, eosinophils, plasmacytoid dendritic cells, and specific T cell subsets, with decreased common lymphoid progenitors and T cell CD4+ Th1. CCCA exhibited increased B cells, cancer-associated fibroblasts, mast cells, and T cell subsets, with decreased granulocyte-monocyte progenitors, T cell CD4+ memory, and T cell CD8+ naïve cells. Conclusions: These findings deepen our understanding of cellular involvement in inflammatory alopecia and offer potential targets for future therapeutic interventions.