Spatial transcriptomic landscape and cellular neighborhood heterogeneity in cervical cancer: integrative single-cell and spatial RNA sequencing analysis.

BACKGROUND: Cervical cancer ranks as the fourth most common malignancy among women worldwide, with approximately 604,000 new cases and 342,000 deaths annually. Despite advances in HPV vaccination and screening programs, significant challenges remain in diagnostic accuracy, therapeutic efficacy, and prognostic assessment. Traditional bulk RNA sequencing methods average gene expression data across entire tissue specimens, masking cellular diversity and spatial architectural patterns that influence tumor progression. Telomere maintenance mechanisms play crucial roles in cervical cancer development, with over 90% of cervical cancers exhibiting telomerase reactivation. METHODS: This study integrated single-cell RNA sequencing and spatial transcriptomics technologies to construct the first comprehensive spatial molecular atlas of cervical cancer. Single-cell RNA sequencing analysis utilized public datasets from the GEO database with systematic quality control through a multidimensional assessment framework. Spatial transcriptomics analysis employed the 10x Genomics Visium platform, successfully identifying 38 distinct cellular neighborhoods (N1-N38). Spatial expression patterns of 10 key genes were analyzed, including epithelial markers (MUC1, CDH1, KRT16), stromal markers (COL1A1, COMP, DCN), and immune markers (CD3G, FCGR1A). In vitro validation was performed using HeLa and SiHa cell lines. RESULTS: Multiple distinct cell subpopulations and 38 cellular neighborhoods with unique molecular characteristics were successfully identified. MKI67 demonstrated spatial heterogeneity with proliferative "hotspots"; COMP was primarily expressed in stromal regions and participated in tumor-stroma interactions; KRT16 exhibited patterns reflecting epithelial differentiation gradients. Different neighborhoods showed enrichment for various cell types, including immune hotspots, stromal-rich regions, and epithelial-dominant areas. Immunoglobulin-related genes (IGLC2, IGHG1, IGHG2, etc.) displayed unique spatial expression characteristics. Quantitative PCR validation confirmed differential expression patterns between cell lines. CONCLUSIONS: This study established the first comprehensive spatial transcriptomic atlas of cervical cancer, revealing unprecedented insights into tumor microenvironment organization and cellular spatial relationships.