Utilizes microsatellite instability characteristics for predicting the prognosis and immunotherapy sensitivity of melanoma patients

Abstract

Great progress has been made for Immunotherapy in various tumor diseases due to microsatellite instability (MSI), but research on MSI melanoma is still limited and its potential mechanism still unclear. In this research, we developed a framework derived from MSI melanoma features to forecast the prognosis of melanoma patients and their susceptibility to immunotherapy. At first, we downloaded gene expression data, protein data, somatic mutation data, and copy number variation data from a cancer genome map (TCGA) for patients with melanoma. Chip data and single cell data were also downloaded from Gene Expression Profiling (GEO). And then, based on the 18 differential genes (DEGs) selected from the differential genes in TCGA, GSE62254, and GSE122401, we can divide the patients into three categories. These three clusters have obvious pathway enrichment characteristics. The main enrichment pathways in Cluster A were mismatch repair-related pathways, while the main enrichment pathways in Cluster C were tumor-related pathways and angiogenesis. In addition, Cluster A and Cluster B exhibited a broader tumor mutation burden than Cluster C, suggesting that Cluster A might be more sensitive to immunotherapy. We also found that group A had a significantly better survival rate than group C. These significantly different results confirm the reliability of the classification. Subsequently, by applying the minimum absolute contraction and selection operator (LASSO) cox regression method, we developed a two-gene marker prediction model. Low-risk scoring, characterized by increased mutation burden and immune activation, but low survival with matrix activation and cancer-related pathways observed in the high-risk scoring group. Patients with low risk scores diagnosed in the immunotherapy cohort showed significant therapeutic advantages and clinical benefits. In a word, we constructed a new model for predicting the prognosis of patients with melanoma and their response to immunotherapy. Thus guide the choice of treatment methods and the identification of novel biomarkers for melanoma.