Abstract 2377: Clonal evolution and mutational signatures in urothelial carcinoma revealed by paired exome analysis

Abstract

Abstract Background: Bladder cancer (BC) is the 5th most common cancer in the western world. 75% of patients are diagnosed with a single or multifocal non-muscle invasive bladder tumor. About 70% of patients have disease recurrence, and 10-30% will eventually have disease progression. Synchronous and metachronous tumors are mono- or oligo clonal of origin, pointing towards a common ancestry (field disease). The field disease is acquiring molecular alterations over time that may fuel disease progression and metastasis. Detailed knowledge of tumor heterogeneity, clonality and mutational dynamic of early clonal mutations with disease driving potential are needed in order to determine the possible effect of targeted treatment. Methods: In this study we characterized genomic alterations in two to five metachronous tumors from 29 patients initially diagnosed with stage Ta disease. Fourteen patients (32 tumors) had non progressive disease (NPD) and 15 patients (34 tumors) had progressive disease (PD). Whole exome sequencing (WES, ∼50x mean read depth), Ultra deep targeted sequencing (∼6,809x mean read depth) and whole transcriptome RNA-seq was performed for all samples. In addition multiregional WES was performed on 8 adjacent regions from a single tumor. Results: We observed more variation in the mutational spectrum of the tumors from PD patients compared to NPD patients (P = 0.0013). The frequency of APOBEC-related mutations was also higher in tumors from PD patients, and intra-patient shift in APOBEC classification was observed in only one NPD patient (1/14), while it was seen in 53% (8/15) of PD patients (P = 0.009). We also observed a higher proportion of clonal mutations in the ancestral branch compared to sample-specific branches (P = 0.002), based on analysis of phylogenetic trees of disease evolution and calculation of cancer cell fractions. Analysis of paired tumors revealed a high degree of intra-patient mutational heterogeneity, whereas multiregional sequencing of 8 regions from a single tumor biopsy showed little intra-tumor heterogeneity. A low number of subpopulations were present in the individual tumors (PyClone), and overall we found the presence of a subclone in all cancerous cells from all patient samples and the presence of one or two subclones characterizing individual sample. Finally, allele specific expression (ASE) in Ta tumors was found to be higher in tumors from PD patients compared to NPD patients (P = 1.18e-56), with tumor suppressor genes showing a larger ASE than oncogenes (P = 0.0164 vs P = 0.417). Conclusions:The higher intra patient variation of the tumor mutation spectrum in patients with progressive disease suggests that a new mutational profile develops during progression. Furthermore, we found high temporal and low spatial mutational heterogeneity, suggesting that targeted treatment decisions ideally should be based on analysis of biopsies from several tumors. Citation Format: Iver K. Nordentoft, Philippe Lamy, Karin Birkenkamp-Demtröder, Mathilde Borg Houlberg Thomsen, Søren Vang, Palle Villesen Fredsted, Jakob Hedegaard, Michael Borre, Jørgen Bjerggaard Jensen, Søren Høyer, Jakob Skou Pedersen, Torben Falck Ørntoft, Lars Dyrskjøt Andersen. Clonal evolution and mutational signatures in urothelial carcinoma revealed by paired exome analysis. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2377.