Abstract
The prognosis of skin cutaneous melanoma (SKCM) remains poor, and patients with SKCM show a poor response to immunotherapy. Thus, we aimed to identify necroptosis-related biomarkers, which can help predict the prognosis of SKCM and improve the effectiveness of precision medicine. Data of SKCM were obtained from The Cancer Genome Atlas (TCGA) and GEO databases. TCGA samples were classified into two clusters by consensus clustering of necroptosis-related genes. Univariate Cox and least absolute shrinkage and selection operator regression analyses led to the identification of 11 genes, which were used to construct a prognostic model. GSE65904 was used as the test set. Principal component, t-distributed stochastic neighbor embedding, and Kaplan-Meier survival analyses indicated that samples in the train and test sets could be divided into two groups, with the high-risk group showing a worse prognosis. Univariate and multivariate Cox regression analyses were performed, and a nomogram, calibration curve, and time-dependent receiver operating characteristic curve were constructed to verify the efficacy of our model. The 1-, 3-, and 5-year areas under the receiver operating characteristic curves for the train set were 0.702, 0.663, and 0.701 and for the test set were 0.613, 0.627, and 0.637, respectively. Moreover, we performed Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses between the high- and low-risk groups. Single sample gene set enrichment analysis, immune cell infiltration analysis, tumor microenvironment scores, immune checkpoint analysis, and half-maximal inhibitory concentration prediction indicated that the high-risk group showed weaker antitumor immunity; further, the response to immune checkpoint inhibitors was worse, and the high-risk group was sensitive to fewer antitumor drugs. Tumor mutational burden analysis, Kaplan-Meier survival analysis, and correlation analysis between risk score and RNA stemness score revealed that the high-risk group with low tumor mutational burden and high RNA stemness score was potentially associated with poor prognosis. To conclude, our model, which was based on 11 necroptosis-related genes, could predict the prognosis of SKCM; in addition, it has guiding significance for the selection of clinical treatment and provides new research directions to enhance necroptosis against SKCM.