Advancing personalized, predictive, and preventive medicine in bladder cancer: a multi-omics and machine learning approach for novel prognostic modeling, immune profiling, and therapeutic target discovery.

OBJECTIVE

This study aimed to identify and analyze immunogenic cell death (ICD)-related multi-omics features in bladder cancer (BLCA) using single-cell RNA sequencing (scRNA-seq) and bulk RNA-seq data. By integrating these datasets, we sought to construct a prognostic signature (ICDRS) and explore its clinical and biological implications, including its association with immune cell infiltration, tumor microenvironment (TME), and drug sensitivity.

METHODS

Publicly available datasets from TCGA and GEO, including scRNA-seq (GSE222315, 9 samples) and bulk RNA-seq (TCGA-BLCA, 403 samples; GSE13507, 160 samples), were analyzed. Single-cell data were processed using Seurat, and ICD scores were calculated using single-sample gene set enrichment analysis (ssGSEA). Weighted gene co-expression network analysis (WGCNA) identified ICD-related modules, and machine learning algorithms (Lasso, Ridge, CoxBoost) were employed to construct the ICDRS. Survival analysis, immune infiltration, pathway enrichment, and drug sensitivity were evaluated to validate the model.

RESULTS

The ICDRS, based on eight key genes (IL32, AHNAK, ANXA5, FN1, GSN, CNN3, FXYD3, CTSS), effectively stratified BLCA patients into high- and low-risk groups with significant differences in overall survival (OS, P < 0.001). High ICDRS scores were associated with immune-suppressive TME, including increased infiltration of T cells CD4 memory resting (P = 0.02) and macrophages M0/M1/M2 (P = 0.01). Pathway enrichment revealed correlations with cholesterol homeostasis, epithelial-mesenchymal transition (EMT), and KRAS signaling. Drug sensitivity analysis showed high-risk groups were resistant to Cisplatin (P = 0.003), Mitomycin C (P = 0.01), and Paclitaxel (P = 0.004), with IC50 values significantly higher than low-risk groups.

CONCLUSION

The ICDRS serves as a robust prognostic biomarker for BLCA, offering insights into tumor immune evasion mechanisms and potential therapeutic targets. Its integration with clinical features enhances personalized treatment strategies, highlighting the importance of ICD in BLCA immunotherapy and precision medicine. The model's predictive accuracy and biological relevance were validated across multiple datasets, underscoring its potential for clinical application.