The correlation between ARGscore and clinical characteristics was investigated. Using univariate and multivariate analyses, we assessed whether the ARGscore is separate from other clinical factors as a prognostic indicator. We quantified 22 infiltrating immune cells using CIBERSORT and compared their levels between subgroups. Moreover, immune checkpoint expression levels were compared between distinct groups. Additionally, we examined the correlation between the ARGscore, microsatellite instability (MSI), and cancer stem cell (CSC). The waterfall function of the “mafTools” package was used to present mutations in the TCGA-COAD/READ cohort in order to differentiate somatic mutations from ARGscore-low and ARGscore-high subgroups. We calculated the IC50 of the used chemotherapeutic agents for CRC using the pRophetic package to explore the differences in efficacy between the two subgroups.
Free full text: Click here