Fig. 7

ICI and TMZ sensitivity analysis of IEscore in LGGs. A-C The distributions of TIDE scores were compared between low- and high-IEscore groups in the TCGA (A), meta-CGGA (B) and meta-GEO (C) LGG cohorts. D Kaplan–Meier curves for low- (n = 15) and high-FRscore (n = 26) patient groups in Vanallen cohort (melanoma patients treated with anti-CTLA4 therapy) showed higher-IEscore melanoma patients had worse prognosis. E Comparison of the response proportions of melanoma patients with anti-CTLA4 therapy between low-(n = 15) and high-IEscore (n = 26) melanoma patients. F Kaplan–Meier curves for low- (n = 45) and high-IEscore (n = 201) patient groups in IMvigor 210 cohort (cancer patients treated with anti-PD-L1 therapy) showed higher-IEscore cancer patients had worse prognosis. G Comparison of the response proportions of cancer patients with anti-PD-L1 therapy between low- and high-IEscore cancer patients. H-J The TMZ AUC values, which represented the TMZ resistance level, were calculated using data from PRISM and CTRP datasets respectively, and were compared between low- and high-IEscore LGG subgroup in the TCGA (H), meta-CGGA (I) and meta-GEO (J) LGG cohort. K-N Kaplan–Meier curves represent the LGG subgroups of patients with different IEscore levels and TMZ therapy status to reflect the OS (K) and PFI (L) statuses in the TCGA cohort and OS status in the meta-CGGA cohort M. The labelled asterisk indicated the statistical p value (***p < 0.001)