Ferroptosis, Extracellular Vesicles and Circulating Tumor DNA as Potential Biomarkers for Uveal Melanoma Prognostication

Abstract

Purpose : Uveal melanoma (UM) is an ocular cancer, with propensity for fatal liver metastases. Efficacious treatments and better prognostication approaches for metastatic UM (MUM) are urgently needed. We previously identified putative biomarkers of UM progression and therapeutic response. Now, we are additionally analysing extracellular vesicles (EVs) and circulating tumor DNA (ctDNA) isolated from plasma of healthy donors (HD), primary UM (PUM) and MUM patients to discover potential therapeutic response or prognostic UM biomarkers.

Methods : Association between expression of candidate biomarkers and UM-patient survival was investigated by analysing the RNA data from surgically removed UM tumors and deposited in the Tissue Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). EVs were isolated from plasma of 6 HD, 8 PUM and 5 MUM subjects using sucrose cushion ultracentrifugation and their proteomes profiled by mass spectrometry. Cell-free DNA was isolated from 12 MUM patients and 4 HD plasma samples using the MagMax Cell-Free DNA Isolation kit and the relative amount of ctDNA for specific UM mutations quantified by qPCR.

Results : TCGA and GSE84976 analysis showed high levels of transcripts inhibiting ferroptosis to correlate with reduced overall and disease-free-survival in UM patients. In addition, by using the data included in these two datasets, we iteratively identified a gene signature which predicts with high accuracy (0.94) which patients will develop MUM. Preliminary data from the initial cohort of EVs identified ~180 differentially expressed proteins between HD, PUM and MUM subjects. We established assays profiling GNAQ Q209P and GNAQ Q209L ctDNA, with positivity near the expected frequencies in the Irish UM population (23% and 9%, respectively).

Conclusions : TCGA and GSE84976 data revealed that altered expression levels of ferroptosis-related genes correlates with UM prognosis and allowed us to generated a gene signature which can predict which patients will develop MUM, with higher accuracy compared to monosomy 3 alone. This could assist in selecting patients for more intensive surveillance protocols following primary therapy for UM. Currently, we are expanding the cohorts analysed for differentially expressed EV proteins between HD, PUM and MUM subjects. These and additional ctDNA profiling provide candidate prognostic or therapeutic response biomarkers for UM.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.