Experience of Next-Generation Sequencing in urothelial carcinoma specimens with panel for 523 genes

Background. Genomic alterations in urothelial carcinoma (UC) cells range from point DNA mutations to complex chromosomal rearrangements and changes in the number of chromosomes in the tumor cell. The UC genetic profile is highly heterogeneous, leading to significant variability in the natural history of the disease, prognosis, and responses to treatment. To evaluate the genetic alterations of Russian patients with bladder cancer is of great interest. Aim. To evaluate the mutation profile in UC specimens with the next-generation sequencing (NGS) panel for 523 genes. Materials and methods. Thirty-six patients' UC samples fixed in formalin and embedded in paraffin were studied. Carcinoma in situ without papillary tumor was verified in 1 (2.9%), Ta in 14 (38.8%), T1 in 19 (52.7%), TT1 in 2 (5.6%) patients. High-grade UC was verified in 14 (38.9%) specimens. DNA and RNA were isolated from the paraffin blocks, libraries were prepared with the Illumina TruSight Oncology 500 panel, and then NGS was performed, followed by bioinformatics data processing. Results. The median tumor mutation burden (TMB) was 14.1 (1.6-102.9) mutations/Mb: TMB≥20 mutations/Mb – 6 (16.7%). In all cases, the level of microsatellite instability was low. In 36 specimens, 181 therapeutically significant and oncogenic mutations were identified in 62 genes; the median was 5 (1–16) mutations per specimen. Single nucleotide variants prevailed in the mutation structure: 123 (68%); GA had the highest frequency 36 (29.3%). There were 47 (26.0%) indel mutations, 10 (5.5%) amplifications, and 1 (0.6%) translocation. Clinically significant mutations were detected in all specimens. The highest frequency of clinically significant mutations was observed in the FGFR3 genes – 22 (61.1%) specimens with mutations in this gene, KDM6A – 22 (61.1%), STAG2 – 13 (36.1%), PIK3CA – 9 (25.0%), and ARID1A – 9 (25.0%). Pathogenic level 1-2 mutations providing potential therapeutic targets were detected in 29 (80.6%) of 36 specimens and included alterations of 13 genes (AKT1, ATM, BRAF, CHEK2, ERBB2, FGFR3, IDH1, MLH1, NF1, NRAs, PIK3CA, PTEN, and TSC1). Frequent mutations of level 3-4 therapeutic significance were in KDM6A (61.6%), ARID1A (25.0%), and CDKN2A (11.4%) genes. Conclusion. A 523-gene NGS panel study confirmed the high TMB and low rate of microsatellite instability in UC tumor cells. The most common pathogenic mutations associated with potential therapeutic targets in UC were FGFR3, PIK3CA, and ERBB2 alterations.