Fig. 3
From: Genome-wide methylation patterns predict clinical benefit of immunotherapy in lung cancer
Risk score calculated by Lasso Cox regression models and survival analysis in three different cohorts. a Training set (n = 60; our cohort). b IDIBELL set (n = 81). c–d TCGA high- (n = 151) and low-immune pressure cohorts (n = 259). Patients in the TCGA cohort were divided into high- and low-TIL cohorts according to mean value of tumor infiltrating lymphocyte (TIL) fraction. Kaplan-Meier survival analyses of the patients are displayed on the top. The patients in each cohort were divided into low- and high-risk groups based on mean of risk index produced by our model (i.e., mean score). P values were calculated using the log-rank test. The methylation levels of the eight genes included in our model are shown as a heatmap on the bottom. Methylation values were z score normalized per gene. Genes (x-axis) and samples (y-axis) are ordered in increasing order of coefficient and risk score, respectively. Methylation probe for CTSD gene selected by our model is not present in TCGA cohort