Comprehensive characterization of 412 muscle invasive urothelial carcinomas: Final analysis of The Cancer Genome Atlas (TCGA) project.

Abstract

405 Background: We reported the integrated molecular analysis of 131 tumors in 2014 (Nature 507:315, 2014) and now report on the entire cohort of 412 tumors from the TCGA project in chemotherapy-naïve muscle invasive urothelial bladder cancer. Methods: Following strict clinical and pathologic quality control, tumors were analyzed for DNA copy number variants, somatic mutations (WES), DNA methylation, mRNA, microRNA and (phospho-) protein expression, transcript splicing, gene fusions, viral integration, pathway perturbation, clinical correlates, and histopathology Results: There was a high overall somatic mutation rate (8.0/Mb), as previously reported, with a median of 245 and mean of 348 coding region mutations per sample. There were 54 significantly mutated genes (SMGs) (MutSig_2CV), increased from 32 in the original report. TP53 remained the most commonly mutated gene, and chromatin-modifying genes (MLL2, ARID1A, KDM6A) were also frequently mutated. KRAS, ERBB2, RB1, and ELF3 showed significant increased frequencies of mutation. High mutation burden was associated with improved outcome (p = 0.0004). APOBEC mutagenesis explained 70% of the mutation burden and also was associated with survival. 167 genes were silenced by promoter DNA hypermethylation in at least 5% of tumors. Several SMGs were also epigenetically silenced at various frequencies (ZNF773, CDKN2A, FAT1, CASP8). The previously identified four mRNA subtypes were predicted on the larger set identifying a similar proportion of samples in each subtype. We identified 5 expression subtypes based on reverse-phase protein arrays(344 tumors). Five miR expression subtypes were strongly associated with 4 mRNA-based subtypes (P = 3.8E-37) and associated with overall survival (P = 0.05). Conclusions: This integrated analysis of 412 TCGA patient samples validates and extends observations from the first 131 patients and significantly increases our power to detect additional low-frequency aberrations. The results provide a robust basis for functional studies to further our understanding of the biology of bladder cancer, and aid in the development of more precisely targeted therapies.