基于 PANoptosis 相关亚型衍生的预后模型揭示了宫颈癌的免疫特征和治疗脆弱性

Cervical cancer progression is intricately linked to dysregulated cell death pathways, particularly PANoptosis, a coordinated form of programmed cell death. However, prognostic biomarkers and immunological implications of PANoptosis-related genes (PRGs) in cervical cancer remain underexplored.

Transcriptomic data from TCGA-CESC and GEO cohorts were analyzed. Molecular subtypes were defined via non-negative matrix factorization (NMF) clustering of 33 survival-associated PRGs. A 6-gene prognostic model was constructed using LASSO-Cox regression and validated for predictive performance. Immune infiltration, tumor mutation burden (TMB), and drug sensitivity were systematically evaluated.

Two PANoptosis-related subtypes exhibited distinct survival outcomes and immune landscapes. A 6-gene risk model (LRRN4, GNG8, CAVIN3, TNF, PLAAT4, ASPRV1) showed strong performance in the TCGA training cohort (AUC: 0.80-0.83) and maintained moderate prognostic value in the independent GEO validation cohort (AUC: 0.67-0.74). Despite the reduced AUCs in external validation, the model retained significant survival stratification, supporting its biological relevance. High-risk patients showed extracellular matrix (ECM) remodeling, PI3K-Akt activation, and immunosuppression, while low-risk patients exhibited enhanced immune infiltration (NK cells, DCs) and checkpoint expression (CCL19, BTLA). Drug sensitivity profiling identified Vinblastine and Trametinib as potential therapeutic candidates for high-risk patients.

This study establishes a PANoptosis-derived prognostic model with moderate but consistent prognostic utility across platforms, underscoring its potential research value. By linking PRGs to risk stratification and immune heterogeneity, the model provides insights into PANoptosis biology and supports future exploration of personalized immunotherapy in cervical cancer.