Abstract
Neutrophils are crucial immune components within the tumor microenvironment, significantly impacting tumor progression and anti-tumor immunity. To systematically characterize the heterogeneity of neutrophils in bladder cancer (BLCA), we integrated large-scale single-cell RNA sequencing (scRNA-seq) data of BLCA to define the transcriptomic landscape of neutrophil subtypes. Functional enrichment, pseudotime analysis, cell-cell communication, and deconvolution of bulk RNA sequencing (RNA-seq) samples from BLCA were conducted to comprehensively characterize the biological profiles and functions, as well as the prognostic relevance of neutrophil subtypes. A machine learning-based predictive model was developed based on the balance of prognosis-related neutrophil subtypes. We identified five distinct subtypes of neutrophils in BLCA and focused on two subtypes that were prognostically antagonistic. VEGFA+ neutrophils (Neu_0), characterized by pro-angiogenic, immunosuppressive, and extracellular matrix remodeling signatures, showed a significant correlation with poorer survival. GBP1 + neutrophils (Neu_4), characterized by response to interferon, exhibited increased innate immune activities and the production of cytokines that activate anti-tumor immunity, significantly correlated with improved survival. Pseudotime analysis positioned both Neu_0 and Neu_4 as terminal states. Cell-cell communication further identified Neu_0 as a hub orchestrating multiple pro-tumorigenic interactions. The predictive model based on the balance of Neu_0 and Neu_4 effectively stratified BLCA patients into distinct risk groups with significant differences in clinical outcomes, immune landscapes, and response profiles to antibody-drug conjugate (ADC) treatment. The investigation provided novel insights into the functional profiles of neutrophils in BLCA and offered a novel tool for guiding therapeutic strategies in BLCA. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s12672-026-04559-3.