Single-cell RNA-seq combined with bulk RNA-seq explores shared gene signatures between thyroid and breast cancers

Objective: This study aims to identify key genes that are common to both breast cancer and thyroid cancer, as well as to determine shared therapeutic targets relevant to both conditions. Methods: We utilized transcriptome data from both breast and thyroid cancers, along with single-cell data, and applied cell deconvolution techniques to evaluate the extent of monocyte infiltration. Tumor-related gene modules were identified through weighted gene co-expression network analysis (WGCNA), followed by enrichment analysis to uncover significant signals shared within these gene modules. A machine learning approach was then employed to pinpoint hub genes. Additionally, RT-qPCR was performed to validate the expression levels of these hub genes in tumor and adjacent non-tumor tissues from patients with both cancer types. Results: Our analyses revealed that the transcriptional networks of breast cancer and thyroid cancer display significant similarities. WGCNA identified two consensus modules that are strongly associated with both cancers and monocyte infiltration. Enrichment analysis highlighted glycosaminoglycan synthesis pathways as critical signals that are common to both cancers. A total of seven hub genes were identified using the machine-learning approach. Results from RT-qPCR and immunohistochemistry in clinical samples showed that the expression levels of PILRA, Mki67, and UBE2C were markedly different between cancerous and adjacent tissues. Conclusion: PILRA, MKI67, and UBE2C, as potential diagnostic and prognostic biomarkers, are anticipated to serve as promising therapeutic targets for the clinical management of both breast cancer and thyroid cancer.