Integrative analysis of cell death- and senescence-associated transcriptional programs identifies molecular subtypes and key genes in clear cell renal cell carcinoma
Year: 2026
Type: preprint
Abstract: Clear cell renal cell carcinoma (ccRCC) exhibits substantial molecular and immune heterogeneity, yet the role of integrated cell death- and senescence-associated (CDS) programs remains incompletely understood. We characterized 21 CDS transcriptional signatures using multicohort transcriptomic, clinical, pharmacogenomic, and single-cell data. Consensus clustering of 538 TCGA-KIRC tumors identified three subtypes characterized by metabolic–proliferative, immune–inflammatory, and hypoxia–angiogenesis-associated profiles. The immune-enriched C2 subtype showed the poorest survival with elevated exhausted and regulatory T-cell signatures. Focusing on C2, edge perturbation analysis identified 246 network-associated genes, which were narrowed to 23 CDS-related candidates and 12 survival-associated genes. Among 101 machine-learning combinations, CoxBoost followed by random survival forest achieved the best performance, generating a ten-gene prognostic signature with a mean C-index of 0.733 across five cohorts. The resulting risk score was associated with survival outcomes and adverse clinicopathological characteristics. Among the ten genes, LMNB1, LY96, and HSPA8 were further investigated based on their differential expression and high tumor–normal discrimination AUCs. Single-cell profiling and virtual knockout analyses linked LMNB1 to T-cell cytotoxicity-associated networks, LY96 to macrophage antigen-presentation programs, and HSPA8 to T-cell cytokine signaling. Pharmacogenomic analysis identified nine compounds with differential predicted drug responses in C2. Across three independent treatment cohorts, C2 consistently exhibited the lowest response rates. Collectively, these findings demonstrate that integrated CDS transcriptional profiling delineates prognostically distinct molecular states in ccRCC, revealing an immune-enriched yet unfavorable subtype and identifying candidate genes that link cellular stress-associated programs to tumor immune heterogeneity.
Source: Research Square
Authors: Xiangyi Wang, Hao Hu, Kaihong Wang
Institutions: Nanchang University, Chongqing Medical University
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Keywords: clear cell renal cell carcinoma, cell death, cellular senescence, gene expression signatures, molecular subtypes, prognostic gene signature, immune heterogeneity, T cell exhaustion, regulatory T cells, lamin B1, MD-2, Hsc70, drug response prediction, tumor immune microenvironment, CoxBoost, random survival forest, single-cell profiling, virtual knockout
Open Access status: green
OpenAlex ID: W7236386792
