Clonal evolution and molecular stratification of urothelial carcinoma in situ of the urinary bladder
Reproductive Medicine, Urology
Final Report Abstract
In our project, the tumor evolution of 12 bladder carcinoma patients with extensive Carcinoma in situ (CIS) was reconstructed using entirely embedded and histologically analyzed cystectomy specimens, whole exome sequencing of multi-regional CIS, corresponding invasive tumor and normal urothelium samples (total n=84). The mutational landscape proved to be heterogeneous - albeit with the occurrence of classic driver mutations such as TP53, and ARID1A, which were already found in CIS. Some of these alterations, which often arise early in tumor development, could serve as targets for new therapeutic approaches in the future. In addition, the genomes were already characterized by a clear instability in CIS, which was reflected in a high number of copy number changes in chromosomal segments. Phylogenetic reconstruction indicated a monoclonal origin in 10 patients, while two carcinomas arose from at least two independent clones. In parallel, some of the invasive tumors showed no phylogenetically earlier precursor CIS lesions, which were subsequently named “early invasive”, while a second group (“late invasive”) displayed them in the phylogenetic tree. The identification of genetic alterations associated with invasive branches of tumor development was limited, also due to the total number of samples for each group. However, the genetic dimension is not necessarily the only one that can orchestrate an entire expression network, so it is likely that other regulatory mechanisms such as epigenetics may play a role. For this reason, we focused on mutated genes that encode for epigenetic factors and regulators and thus can cause a shift in epigenetic configuration. In a TCGA pan-cancer analysis, these mutations were found to be enriched in bladder carcinomas and were often clonal (=early in tumor development), e.g. KMT2A and KMT2D. Finally, to better understand the impact of KMTs on tumor development, especially KMT2A, two KMT2A knockout (KO) models were generated using CrispR/Cas9 - based on one model reflecting early tumor development (UROtsa) and one reflecting later stages of tumor development (J82). KMT2A KO clones of the UROtsa line showed a reduced clonogenic potential compared to corresponding controls, while no significant effects were observed in J82 cells. Similarly, the KMT2A KO in UROtsa led to a massive change in the transcriptome and DNA methylome. RNA-seq analyses suggested an KMT2A involvement in processes such as cell cycle and key pathways regulating immune and inflammatory response. A possible rationale for an early dysregulation of KMT2A in bladder cancer, as suggested by our phylogenies, could thus be based on the adaptation of the immune response, even if these mechanism does not confer a direct growth/proliferation advantage for the tumor cell itself.
