Clinical prognostication and immunotherapy response prediction in esophageal squamous cell carcinoma using the DNA damage repair‐associated signature

Abstract

BackgroundThe relationship between DNA damage repair (DDR) and cancer is intricately intertwined; however, its specific role in esophageal squamous cell carcinoma (ESCC) remains enigmatic.MethodsEmploying single‐cell analysis, we delineated the functionality of DDR‐related genes within the tumor microenvironment (TME). A diverse array of scoring mechanisms, including AUCell, UCell, singscore, ssgsea, and AddModuleScore, were harnessed to scrutinize the activity of DDR‐related genes across different cell types. Differential pathway alterations between high‐and low‐DDR activity cell clusters were compared. Furthermore, leveraging multiple RNA‐seq datasets, we constructed a robust DDR‐associated signature (DAS), and through integrative multiomics analysis, we explored differences in prognosis, pathways, mutational landscapes, and immunotherapy predictions among distinct DAS groups.ResultsNotably, high‐DDR activity cell subpopulations exhibited markedly enhanced cellular communication. The DAS demonstrated uniformity across multiple datasets. The low‐DAS group exhibited improved prognoses, accompanied by heightened immune infiltration and elevated immune checkpoint expression. SubMap analysis of multiple immunotherapy datasets suggested that low‐DAS group may experience enhanced immunotherapy responses. The “oncopredict” R package analyzed and screened sensitive drugs for different DAS groups.ConclusionThrough the integration of single‐cell and bulk RNA‐seq data, we have developed a DAS associated with prognosis and immunotherapy response. This signature holds promise for the future stratification and personalized treatment of ESCC patients in clinical settings.

Advancing lung adenocarcinoma prognosis and immunotherapy prediction with a multi‐omics consensus machine learning approach

“…Gene expression is meticulously regulated through diverse genetic and epigenetic mechanisms, including methylation, mutations and histone modifications. 27 , 28 Thus, an extensive analysis of multi‐omics patient data is pivotal for delving into disease‐specific regulatory mechanisms. 29 , 30 However, thus far, research has largely been confined to individual omics studies, 31 and the selection of omics clustering methods is greatly influenced by personal preferences, which exacerbates the limitations of specific methodologies as their application broadens.…”

Long non-coding RNAs in ferroptosis and cuproptosis impact on prognosis and treatment in hepatocellular carcinoma

“…The immune score and matrix score of each LIHC sample were assessed using the R program "ESTIMATE" [ 33 ] to quantify the number of immune and matrix components present in vivo. ssGSEA [ 26 ] is a GSEA analysis performed on a single sample, and how the gene list is ordered and the ES is calculated is dependent on the expression values of the genes in the sample, and no longer on the correlation of the genes with the phenotype.…”