Identification and immunological characterization of cuproptosis-related molecular clusters in retinoblastoma

Abstract

Abstract Background: To proposed a novel strategy for retinoblastoma (RB) treatment and study by positing a connection between cuproptosis and immunological changes and the development and incidence of RB. Methods: Using "Retinoblastoma" as the search phrase, two microarray datasets of Retinoblastoma (GSE208143 and GSE97508) were obtained from the GEO database. 42 samples of retinal tissue were collected, comprising 33 samples from Retinoblastoma patients and 9 samples from healthy individuals in the GEO database. Results: We carefully examined the immunological characteristics and differential expression of CRGs in normal and retinoblastoma people for the first time in this study and developed a unique machine learning model based on the selected genes that has the ability to forecast patients with accuracy. Conclusion: Our bioinformatic analysis uncovered the relationship between CRGs and immune cells that have been infiltrated, revealed the significant immune heterogeneity among RB patients with distinct cuproptosis clusters, and created a signature machine learning model based on the chosen genes that could accurately predict the patients.