Genomic landscape of uveal melanoma

Abstract

Uveal melanoma (UM) is the most common cancer of the adult eye which can manifest as a highly aggressive form approximately half of the time. Here a comprehensive landscape of genetic alterations in UMs is described. It was identified by integrating copy number alterations (CNAs), and transcriptomic and whole exome sequencing data from 207 primary UMs. Focal copy number analysis with the GISTIC algorithm refined the boundaries of chromosomal segments with chromosomal gains or losses and candidate cancer genes within these segments were identified. Chromosome 8q24.3 was the region most frequently amplified in UMs, being detected in 72% of tumours. A comparison of focal copy gains and losses with that described by a pan-cancer study revealed Plectin 1 as a candidate gene within the 8q24.3 amplicon. Integration of copy number and transcriptomic data also revealed enrichment of genes within pathways leading to activation of NF-kappa B, WNT signaling and RNA splicing. Using a complementary bioinformatics approach, additional novel mutations in known dominant UM driver genes (GNAQ, GNA11, BAP1, SF3B1, EIFIAX and CYSLTR2) were identified and an accurate estimate of the frequencies of mutations in each gene were obtained. Finally, integration of data obtained from CNAs with mutational and transcriptome data reveled homozygous deletions, protein damaging mutations and gene fusions that targeted chromatin modifiers, and specifically genes encoding components of the human SWItch/Sucrose NonFermentable (SWI/SNF) chromatin remodeling complex. Genes from the BAF complex (ARID1A and ARID1B) and the PBAF complex (PHF10) were subjected to functional loss through CNAs, gene fusions and mutations. Two of these chromatin modifiers (ARID1B and PHF10) map to chromosome 6q whose loss is associated with metastasis in a subset of UMs, and an ARID1B fusion is found in a tumour with a BAP1 mutation that subsequently underwent metastasis. In conclusion, this study provides a comprehensive overview of the landscape of genomic alterations in UM, identifying candidate genes in regions of CNAs and providing further insights into the altered pathways of tumour development and progression.