Deciphering the molecular landscape of Sjögren's disease, mucosa-associated lymphoid tissue lymphoma, and thyroid cancer: unraveling the complexities of disease mechanisms and diagnostic biomarkers.

[BACKGROUND] Sjögren's disease (SjD), mucosa-associated lymphoid tissue lymphoma (MALT lymphoma), and thyroid cancer (THCA) are clinically distinct yet immunologically intertwined diseases characterized by chronic inflammation and aberrant immune activation. However, the shared molecular basis linking autoimmunity and oncogenesis remains poorly defined.

[METHODS] We performed an integrative transcriptomic analysis across three datasets (TCGA-THCA, GSE40611/GSE127952 for SjD, and GSE25638 for MALT lymphoma), totaling 749 samples. Analyses included differential expression, Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Set Enrichment Analysis (GSEA) enrichment, Weighted correlation network analysis (WGCNA), and immune infiltration profiling. Machine learning algorithms (support vector machine (SVM), random forest (RF), and least absolute shrinkage and selection operator (LASSO)) were applied to construct a multi-disease diagnostic model. Immune infiltration patterns were further characterized using single-sample gene-set enrichment analysis (ssGSEA) and cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT).

[RESULTS] A total of 33 differentially expressed genes (DEGs) were shared among the three diseases, including immune-related genes-diagnosis common differentially expressed genes (DCDEGs). Functionally, these genes were enriched for leukocyte chemotaxis, cytokine signaling, MAPK cascade regulation, and NF-κB signaling (GO:BP), extracellular matrix components (GO:CC), cytokine/chemokine/receptor activity (GO:MF), and KEGG pathways including cytokine-cytokine receptor interaction, chemokine signaling, and T-cell receptor signaling. WGCNA identified multiple immune-associated modules across datasets, with CCL20, NXN, PLA2G7, and TGFB1I1 consistently emerging as hub genes in THCA, SjD, and MALT lymphoma. An integrative diagnostic model based on eight overlapping genes (CCL21, FGL2, FYN, OGN, PLA2G7, POSTN, TGFB1I1, and TMEM155) demonstrated strong predictive accuracy (AUC = 0.992 in THCA, 0.973 in MALT lymphoma, and 0.755 in SjD). Immune infiltration analyses revealed disease-specific immune landscapes: dendritic cells and T cell subsets were positively correlated with most DCDEGs.

[CONCLUSION] This integrative multi-cohort study uncovered common transcriptional and immune signatures underlying SjD, MALT lymphoma, and thyroid cancer. The identification of shared hub genes, particularly PLA2G7 and TGFB1I1, provides novel insight into the immune-driven transition from chronic inflammation to malignancy and offers promising biomarkers for cross-disease diagnosis and immunotherapeutic stratification. Key Points • Key genes (PLA2G7 and TGFB1I1) affecting the occurrence of Sjögren's syndrome, mucosa-associated lymphoid tissue lymphoma, and thyroid cancer are identified for the first time. • Bioinformatics methods were employed to simultaneously study three diseases for the first time.