Characterization of prognosis and immune infiltration by a novel glutamine metabolism-related model in cutaneous melanoma

Glutamine metabolism (GM) plays an important role in tumor growth and proliferation. Skin cutaneous melanoma (SKCM) is a glutamine-dependent cancer. However, the molecular characteristics and action mechanism of GM on SKCM remain unclear. Therefore, we aimed to explore the effects of GM-related genes on survival, clinicopathological characteristics, and the tumor microenvironment in SKCM. In this study, 682 SKCM samples were obtained from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Consensus clustering was used to classify SKCM samples into distinct subtypes based on 41 GM-related genes. Differences in survival, immune infiltration, clinical characteristics, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways as well as differentially expressed genes (DEGs) between subgroups were evaluated. A prognostic model was constructed according to prognostic DEGs. Differential analyses in survival, immune infiltration, tumor microenvironment (TME), tumor mutation burden (TMB), stemness, and drug sensitivity between risk groups were conducted. We identified two distinct GM-related subtypes on SKCM and found that GM-related gene alterations were associated with survival probability, clinical features, biological function, and immune infiltration. Then a risk model based on six DEGs (IL18, SEMA6A, PAEP, TNFRSF17, AIM2, and CXCL10) was constructed and validated for predicting overall survival in SKCM patients. The results showed that the risk score was negatively correlated with CD8(+) T cells, activated CD4(+) memory T cells, M1 macrophages, and gamma delta T cells. The group with a low-risk score was accompanied by a better survival rate with higher TME scores and lower stemness index. Moreover, the group with high- and low-risk score had a significant difference with the sensitivity of 75 drugs (p < 0.001). Overall, distinct subtypes in SKCM patients based on GM-related genes were identified and the risk model was constructed, which might contribute to prognosis prediction, guide clinical therapy, and develop novel therapeutic strategies.