Construction of prognostic features associated with the tumor microenvironment and endothelial cells in cervical cancer by combining single-cell and transcriptomic data and PCR validation

Cervical cancer (CC) ranks among the most common gynecological malignancies. This study aimed to investigate the prognostic value of genes associated with tumor ecosystem (TES) and endothelial cells in CC. Key cells were identified through the analysis of single-cell RNA sequencing (scRNA-seq) data. Differentially expressed genes (DEGs) associated with TES were intersected with marker genes associated with key cells to identify candidate genes. Prognostic genes were selected through analyses such as univariate regression and machine learning. A nomogram was constructed and validated. Immune cell infiltration profiles were assessed using the CIBERSORT algorithm. scRNA-seq analysis identified endothelial cells as key cells. The intersection of marker genes associated with key cells and DEGs associated with TES yielded 17 candidate genes. Five prognostic genes (ALKBH2, CXCL1, TFPI2, PLAU, and CXCL8) were identified through analyses such as machine learning algorithms. A nomogram model exhibited high sensitivity and specificity. Immune infiltration analysis identified 7 types of differential immune cells, including plasma B cells, Macrophage M0, activated mast cells, resting myeloid dendritic cells, and others. These findings collectively provided novel insights into the molecular mechanisms and immune microenvironment characteristics underlying TES, laying a foundation for the development of potential therapeutic targets and prognostic biomarkers.