Abstract
Purpose :
To determine the expression of DNA repair genes in uveal melanoma (UM) and to test whether certain DNA repair genes are differentially expressed between prognostically-good and bad tumors
Methods :
A genome-wide gene expression analysis was performed in 64 UMs that were primarily enucleated in the LUMC (Leiden, The Netherlands) between 1999 and 2008. Expression of 117 DNA repair genes was analyzed, and genes with variable expression were selected for further analysis. Expression was compared to histopathological data and survival for genes that showed a significantly differential expression between disomy 3 (D3) and monosomy 3 (M3) tumors. Bonferroni correction was applied for multiple comparisons.
Results were validated by analyzing the expression of these genes in relation to survival in a publicly available set of 110 UMs from the University of Genoa (Italy) and Curie Institute, Paris (France).
Results :
We identified 44 DNA repair genes that showed a sufficient level (cut-off standard deviation: ≥ 0.31) of variation in expression. A high expression of PRKDC (p=0.002) and STRA13 (p<0.001) was associated with poor survival in the LUMC set, as were a low expression of XPC (p<0.001), GTF2H4 (p=0.001), BAP1 (p=0.002), SHFM1 (p=0.006), and WDR48 (p<0.001). The associations with survival for PRKDC, XPC, BAP1, and WDR48 (Figure 1) were confirmed in the validation set. The expression values of these genes corresponded positively with the copy number of the respective chromosomes harboring these genes (PRKDC: chromosome 8q; XPC, BAP1 and WDR48: chromosome 3) (all p=0.001). A low expression of XPC and WDR48 was related to a large tumor diameter (p=0.004 and p=0.01, respectively) and a mixed/epithelioid cell type (p=0.03 and p=0.007, respectively).
Conclusions :
Certain DNA repair genes are differentially expressed between prognostically-good and bad UMs. An increased understanding of the role of DNA repair genes in UM may result in the identification of new targets of therapy.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.