Abstract
Purpose :
To identify NKT cell infiltration-related gene biomarkers associated with survival and explore the rational development of tailored therapies targeting selected NKT subsets for metastatic uveal melanoma (UM).
Methods :
Single-cell transcriptomic datasets from three independent studies (GSE74597, PRJNA549112, & GSE128243) were combined to investigate the NKT cell-associated immune-related genes. Next, NKT cell gene biomarkers related to patient survival were shortlisted and validated across diverse repository UM cohorts: TCGA Ocular Melanomas (UVM) (n=80), primary UM dataset GSE22138 (n=63) and GSE84976 (n=28). A risk score developed based on the shortlisted gene biomarkers was assessed using multivariate Cox and Kaplan–Meier survival analyses. The epigenetic modulation of the discerned biomarkers was also explored.
Results :
From single-cell RNA sequencing, 598 genes were identified for univariate Cox regression analysis. Subsequent analysis in the TCGA UVM dataset revealed two gene clusters significantly associated with UM patients' survival outcomes, categorizing them as favorable and unfavorable prognostic genes (Figure 1A). Essential genes like HFE, CD83, and FCGRT were linked to Cd1d-associated proteins antigen presentation of glycolipids (Figure 1B). A risk score from favorable genes emerged as a robust prognostic marker for overall survival (Figure 1C&D). A negative correlation was found between gene expression and methylation within their respective promoter regions, as some example genes illustrated in Figure 2.
Conclusions :
Our study highlights the regulatory role of NKT-related genes in UM and their potential function in immunogenicity. This insight may guide the design of novel combinatorial treatments, improving UM therapeutic outcomes.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.