Abstract
Cervical cancer is a common malignancy in women globally. Its development links to immune evasion with T cell exhaustion as a key mechanism. This study used transcriptomic data to explore T cell exhaustion-related features and networks in cervical cancer. After datasets from TCGA and GEO were analyzed, WGCNA was used to identify gene modules related to T cell exhaustion, and machine-learning was applied to screen central regulatory genes. This comprehensive approach helped reveal the architecture of immune-related molecular networks and provided a reliable basis for identifying key regulatory factors. 1888 differentially expressed genes were found between CIN and controls. The magenta module from WGCNA correlated with T cell exhaustion. Five hub genes were selected (ADAMDEC1, MIAT, PGR, SLAMF8, SLC7A7). Immune infiltration analysis showed differences in immune cell distribution, suggesting a link between T cell dysfunction and immune microenvironment remodeling. Drug-gene prediction and docking suggested PGR as a potential therapeutic target. This study elucidates key molecular characteristics and immune regulatory mechanisms associated with T cell exhaustion during cervical cancer progression. The findings provide new insights into immune evasion in cervical cancer and suggest potential targets for immunotherapy, which may facilitate clinical prediction and personalized treatment strategies. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s12672-026-04423-4.