Fig. 2: Comprehensive characteristics of MSS-PRs.
From: T-bet+CD8+ T cells govern anti-PD-1 responses in microsatellite-stable gastric cancers
a Violin plots of immune infiltration in MSI-H&TMB-H&TILs-H (n = 31), MSI-PR (n = 49), and MSS-NR (n = 167) patients in TCGA-STAD cohort, estimated by cibersortX. Two-tailed Mann-Whitney U test was used to evaluate statistical significance between every two subtypes. b Heatmap of somatic mutations of MSI-H&TMB-H&TILs-H, MSI-PR, and MSS-NR patients in TCGA-STAD cohort. Chi-square test was used to evaluate statistical significance between mutational rates of MSS-PRs and MSS-NRs. Patients were sorted in descending order of PUL scores. c Heatmap of somatic mutations in MSS responders and MSS non-responders in PRJEB25780 cohort (PIK3CA, P = 0.016, chi-square test). d Heatmap of chromosome number variations (CNV) in MSS-PRs of TCGA-STAD cohort. Q values were computed using GISTIC 2.0. e Differential expression genes between MSS-PRs and MSS-NRs in TCGA-STAD and ACRG cohorts. The y-axis denotes the fold changes in TCGA-STAD cohort and ACRG cohort combined. f The ROC curves of top 8 up-regulated genes in MSS-PRs for predicting ICI outcomes in MSS patients of PRJEB25780 cohort (mean AUC = 0.85 ± 0.08). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, ns is abbreviation for not significant.
