Integrated single-cell and bulk transcriptomic analysis reveals shared pathogenesis and prognostic biomarkers in breast and thyroid cancers.

Background: Breast cancer (BC) and thyroid cancer (TC) are two hormonally regulated malignancies with increasing evidence of significant comorbidity. However, the underlying molecular mechanisms contributing to their co-occurrence remain unclear. Purpose: This study aimed to elucidate the shared pathogenesis of BC and TC and to identify common prognostic biomarkers and therapeutic targets. Study design: An integrative bioinformatics study combining single-cell and bulk RNA sequencing data was conducted to investigate shared molecular features between BC and TC. Methods: Differentially expressed genes (DEGs) were identified and subjected to functional enrichment analysis. Single-cell transcriptome analysis was performed to characterize tumor microenvironment composition and malignant cell heterogeneity. Copy number variation (CNV) and non-negative matrix factorization (NMF) analyses were used to identify key gene expression modules. Weighted gene co-expression network analysis (WGCNA) was applied to bulk transcriptomic data to determine critical cell populations. A prognostic signature was constructed using 101 machine learning algorithms, and functional assays were conducted to validate gene function. Results: Enrichment analyses indicated that the JAK-STAT signaling pathway and cytokine-cytokine receptor interaction are shared pathogenic mechanisms. Single-cell analysis revealed immune cell involvement and malignant cell heterogeneity. Modules MP2, MP4, and MP5 were identified as critical in both cancers. WGCNA highlighted SFRP2+ fibroblasts and HLA_DPB1+ myeloid cells as key players in tumorigenesis. A prognostic model was developed, and SMR3B was validated as a shared prognostic gene that influenced proliferation, migration, and invasion in both BC and TC. Conclusion: This study provides comprehensive insights into the shared molecular mechanisms of BC and TC and identifies SMR3B as a promising prognostic biomarker and therapeutic target, offering new avenues for managing patients at dual risk.