LAUSR

Background: Bladder cancer is a highly heterogeneous disease - both clinically, and at the genomic and the transcriptomic levels. Despite complete tumor resections, recurrent tumors share a high number of mutations with the initial/earlier tumors, indicating that new tumors arise from fields of transformed areas of cells in the urothelium (i.e. field disease). Furthermore, the high mutational load found in bladder cancer and the assumed high degree of heterogeneity may have huge implications on therapy response. Thus, detailed knowledge on tumor heterogeneity and mutational dynamics in adjacent normal urothelial cells is needed in order to understand field disease, disease development and disease progression. To address this, we performed comprehensive genomic and transcriptomic analysis of multiple samples of tumor and normal tissue procured from four cystectomy specimens and associated metastases. Methods: Multiple tumors, lymph node metastasis, relapse biopsies and adjacent normal tissue from four patients with muscle invasive disease were analyzed. A targeted panel (illumina TruSeq Custom Amplicon v1.5) was designed from whole exome sequencing of bulk tumor DNA and germline DNA. Targeted sequencing of DNA from small cellular regions procured by laser-microdissection from tumor biopsies and from adjacent normal tissue was used for generating a comprehensive genomic map of the bladder. Transcriptomic profiling (Fluidigm GE) was carried out using RNA from LMD procured tumor regions matching the regions used for targeted sequencing. Results: In total, whole exome sequencing was performed on DNA from 15 bulk tumors with an average coverage of 59.2X. We identified a total of 331, 862, 685 and 1344 mutations, respectively, in the four patients with 15-85% of the mutations being ancestral. Of all the mutations, 95, 380, 117 and 335, respectively, were estimated to have a functional impact. Targeted sequencing of 207, 595, 538 and 483 mutations in the four patients was performed on DNA from 137 small cellular regions, including DNA from 17 adjacent normal tissue areas. We observed low local heterogeneity while comparing more distant regions revealed significantly higher levels of heterogeneity. This observation was in agreement with the heterogeneity observed at the transcriptomic level, where some regions within the same bladder appeared to have basal characteristics while others appeared to have luminal characteristics. Conclusion: Large intra-patient genomic and transcriptomic heterogeneity can be observed in bladder cancer. However, even in multifocal tumors, all the regions shared a substantial number of mutations, indicating a common origin. Ongoing investigations indicate the presence of low frequency tumor associated mutations in normal appearing tissue. Citation Format: Mathilde B. Thomsen, Philippe Lamy, Iver Nordentoft, Soren Vang, Line Reinert, Soren Hoyer, Torben F. Orntoft, Jorgen B. Jensen, Lars Dyrskjot. Comprehensive characterization of genomic and transcriptomic heterogeneity in advanced bladder cancer by multiregional analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3948. doi:10.1158/1538-7445.AM2017-3948