Depression and breast cancer (BC) demonstrate significant clinical comorbidity, yet their shared molecular mechanisms remain unclear, particularly regarding immune pathway regulation. This study systematically analyzed Depression-associated gene expression profiles (Gene Expression Omnibus (GEO) database) and BC transcriptomic data (The Cancer Genome Atlas (TCGA) database), identifying overlapping differentially expressed genes (DEGs). Functional enrichment (Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG)) and protein–protein interaction (PPI) network analyses (STRING/Cytoscape) were employed to elucidate biological processes, followed by least absolute shrinkage and selection operator (LASSO) regression and receiver operating characteristic (ROC) curve validation to prioritize key genes. Immune infiltration patterns were assessed via the xCell algorithm, with Spearman correlation linking genes to immune subsets, and single-gene Gene Set Enrichment Analysis (GSEA) evaluating pathway activity. In total, 93 overlapping genes were identified, with predominant involvement in immune-related pathways being revealed by functional enrichment analysis. BHLHE41, EpCAM, and GSTM2 were prioritized as mechanism-associated genes through integrated LASSO regression and ROC analyses. Significant correlations were observed between these genes and specific immune cell populations. GSEA further linked these genes to immune response pathways, suggesting their regulatory roles. These findings highlight immune dysregulation as a shared mechanism underlying Depression-BC comorbidity, providing a foundation for developing early diagnostic strategies and therapeutic strategies targeting both conditions.

Primary sclerosing cholangitis (PSC) is a chronic inflammatory precursor associated with an increased risk of intrahepatic cholangiocarcinoma (ICC), yet identifying malignant features within the persistent inflammatory background remains challenging. In this study, a background-deviation framework was applied to explore malignant-associated determinants during PSC-associated cholangiocarcinogenesis. Single-cell RNA sequencing data from PSC, ICC tumor tissues, and adjacent non-tumor tissues were integrated, followed by functional enrichment, CellChat analysis, Monocle 2 pseudotime reconstruction, Non-negative Matrix Factorization (NMF), STRING/Cytoscape network analysis, and diagnostic signature construction using LASSO regression and exhaustive best subset selection. Single-cell profiling suggested disease-associated cellular remodeling, including cholangiocyte expansion in ICC samples. Functional and intercellular communication analyses indicated a putative transition from an immune-dominant PSC state toward a hyper-biosynthetic ICC-associated phenotype, accompanied by a possible MIF receptor-usage shift from CXCR4 to CD44. Monocle 2 and NMF further identified candidate malignant-associated trajectories and meta-programs, with MYC/TP63-related regulatory signals emerging as potential contributors. Based on these exploratory findings, best subset selection identified a two-gene transcriptomic candidate signature comprising PMAIP1 and GADD45A, which showed promising discriminative performance in internal cross-validation and an external tumor-versus-adjacent validation cohort. These findings provide a transcriptomic basis for further validation of PSC-associated cholangiocarcinogenesis and potential ICC surveillance markers.