Molecular Subtypes of Urothelial Bladder Cancer: Results from a Meta-cohort Analysis of 2411 Tumors

Previous molecular subtyping for bladder carcinoma (BLCA) involved <450 samples, with diverse classifications.To identify molecular subtypes by curating a large BLCA dataset.Gene expression publicly available were combined and reanalyzed. The dataset contained 2411 unique tumors encompassing non-muscle-invasive (NMIBC) and muscle-invasive BLCA (MIBC). Subtypes were reproduced on The Cancer Genome Atlas, UROMOL, and IMvigor210.Subtypes were assigned by gene expression.Kaplan-Meier analyses were performed for subtype-clinical outcome correlations; Chi-square/Fisher exact tests were used for subtype-clinicopathological parameters associations.We identified six molecular subtypes with different overall survival (OS) and molecular features. Subtype Neural-like (median OS, 87 mo) is prevalent in MIBC and characterized by high WNT/β-catenin signaling. HER2-like (107.7 mo) is distributed evenly across NMIBC and MIBC, with higher ERBB2 amplification and signaling. Papillary-like (>135 mo), an NMIBC subtype enriched in urothelial differentiation genes, shows a high frequency of actionable FGFR3 mutations, amplifications, and FGFR3-TACC3 fusion. Luminal-like (91.7 mo), predominantly NMIBC, has higher MAPK signaling and more KRAS and KMT2C/D mutations than other subtypes. Mesenchymal-like (MES; 86.6 mo) and Squamous-cell carcinoma-like (SCC; 20.6 mo) are predominant in MIBC. MES is high in AXL signaling, whereas SCC has elevated PD1, CTLA4 signaling, and macrophage M2 infiltration. About 20% of NMIBCs show MIBC subtype traits and a lower 5-yr OS rate than Papillary-like NMIBC (81% vs 96%). The main limitations of our study are the incomplete clinical annotation, and the analyses were based on transcriptome subset due to comparisons across gene expression quantification technologies.BLCA can be stratified into six molecular subtypes. NMIBC, with a high risk of progression, displays the molecular features of MIBC.Biomarkers are urgently needed to guide patient treatment selection and avoid unnecessary toxicities in those who fail to respond. We believe molecular subtyping is a promising way to tailor disease management for those who will benefit most.