Urachal (U) and non-urachal (NU) adenocarcinomas (adenoCA) of the bladder: A comparative comprehensive genomic profiling (CGP) study.

Abstract

535 Background: Although both U and NU bladder adenoCA share several histological similarities, they differ in site of origin and optimal treatment paradigms. They are both relatively resistant to conventional cisplatin-based chemotherapy and surgical resection is the only curative option for organ-confined stages. The purpose of this study is to investigate the differences of genomic alterations (GA) between these tumor types, with the aim of identifying potential therapy targets. Methods: A total of 133 U and 328 NU adenoCA were analyzed from a series of Formalin-Fixed Paraffin-Embedded tissues obtained from clinically advanced bladder tumors. Hybrid-capture-based CGP was performed to evaluate all classes of GA. Genomic ancestry and gene signatures, including the somatic-germline nature, were determined with algorithm-based analysis of sequencing data. Tumor mutational burden (TMB) was determined based on at least 0.8 Mbp of sequenced DNA, and microsatellite instability (MSI) was assessed on at least 1500 loci. All p-values were two-sided, and multiple hypothesis testing correction was performed using the Benjamini–Hochberg procedure to calculate the false discovery rate. Results: Sex distribution for U adenoCA was similar (M: 50.4%; F: 49.6%), whereas more men were diagnosed with NU adenoCA (M: 63.4%; F: 36.6%). Median ages were similar between the two groups (60 vs 62 years for U and NU respectively). The most frequent GA in both U and NU cohorts included TP53 (86.5% vs 81.1%) and KRAS (34.6% vs 27.7%). GAs characteristic of colorectal adenocarcinoma, such as SMAD4 and GNAS, were more common in U vs NU (28.6% vs 16.5% for SMAD4, p= 0,069; 18% vs 8.8% for GNAS, p= 0,071). Conversely, mutations typical of urothelial carcinoma, including TERT and RB1, were prevalent in NU adenoCA (14.7% vs 0.77% for TERT, p<0.01; 9.2% vs 2.3% for RB, p=0,071). Notably, both U and NU adenoCA exhibited targetable GA in PIK3CA (7.5% vs 7.9%) and ERBB2 (6.8% vs 7.6%). Biomarkers associated with potential benefit from anti-PD(L)1 were infrequent. These tumors were generally MSI stable, with a low TMB (2.61 vs 3.48 mut/Mb for U and NU respectively) and did not frequently show PD-L1 expression even at a low cut-off of > 1% (5 cases in U vs 4 in NU). Genomic ancestry distributions were similar, with EUR frequency 66% in U and 68% in NU patients. Genomic signatures were similar with both tumor types featuring a predominant mix of APOBEC (25/50%) and MMR signatures (75/36%). Conclusions: U and NU adenoCA revealed notable differences in GA, while PIK3CA and ERBB2 were identified as potential therapy targets. Putative biomarkers of response to anti-PD(L)1 were uncommon. Limitations include lack of clinical data, tumor heterogeneity and retrospective nature. This study highlights the potential of CGP to personalize the treatment of bladder adenoCA and may inform clinical trial designs for these tumors.