Application of a novel combination of Next Generation Sequencing and DNA chromosomal microarray analysis in the molecular diagnosis of uveal melanoma

Abstract

Purpose : Uveal melanoma (UM) is the most common primary malignancy in the adult eye. UMs comprise 5% of all melanomas, and tend to metastasize to the liver, with poor survival prognosis. In this study, we performed a novel combination of comprehensive molecular screening of a collection of UMs using Next Generation Sequencing (NGS) and OncoScan DNA chromosomal microarray analysis to identify pathogenic mutations as well as chromosomal alterations of potential functional significance.

Methods : Clinical data and molecular data from formalin-fixed and paraffin-embedded biopsy specimens of enucleated uveal melanomas were collected from 2012 to 2014 (n = 12). DNA was extracted from unstained tissue sections. NGS was used to interrogate all exons of a comprehensive panel of 397 genes known to be frequently mutated in solid tissue tumors. DNA samples were also subjected to OncoScan DNA microarray analysis, a genome-wide SNP-based analysis to detect the genomic content of copy number abnormalities and gene amplification. All data were submitted for bioinformatics analysis.

Results : All 12 UM specimens in this study had a Q209 mutation in either GNAQ or GNA11. In addition, mutations in SF3B1 (p.R625H) were seen in 33% of cases. BAP1 premature termination mutations, associated with poor prognosis, were seen in two specimens, with a third demonstrating loss of heterozygosity at the BAP1 locus by OncoScan microarray analysis. EIF1AX exon 1 and exon 2 mutations were identified in two cases. Finally, novel mutations in CDKN2A/B, KAT6B, SMAD4, ESR1, and PBRM1 cancer genes were identified in one to two cases each.

Conclusions : This study utilized a novel combination of NGS-based and DNA chromosomal microarray analysis to perform successful molecular diagnosis of a series of UM cases. Many of the mutations found, including GNAQ, GNA11, SF3B1, BAP1, and EIF1AX, are known UM pathogenic mutations. We also identified mutations in other frequently mutated cancer genes, including CDKN2A/B, KAT6B, SMAD4, ESR1, and PBRM1 that carry possible functional significance and deserve additional investigation. A comprehensive capture of the mutation profile holds great promises in the prognostication and development of potential treatment of UM.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.