Fig. 3: Balance between the INF-γ signature and IMS as a biomarker for cancer. | npj Genomic Medicine

Fig. 3: Balance between the INF-γ signature and IMS as a biomarker for cancer.

From: Ratio of the interferon-γ signature to the immunosuppression signature predicts anti-PD-1 therapy response in melanoma

Fig. 3

A Pearson correlations of IMS score with the IFN-γ signature (r = 0.54, R2 = 0.28, p < 0.0001) for all TCGA patients (n = 11,043). B Pearson correlation of logTMB with IFN-γ/IMS ratio (r = 0.20, R2 = 0.04, p < 0.0001) for all TCGA patients (n = 11,043). C, Boxplots showing a summary of the distribution of IMS scores for all TCGA patients, with tumor types ordered by their median IMS score. D, Log hazard ratio estimates and 95% confidence intervals, with adjustment for sex (Female versus Male), age (≥ 60 versus < 60), and TMB with a binary cutoff (top 20% of each cancer type). Cancers in which the IFN-γ/IMS ratio was statistically significantly (p < 0.05) associated with good prognosis are highlighted in blue; significant associations with poor prognosis are in red. E–G Associations of the ORR to immunotherapy for different cancer types with their median IFN-γ signature (linear regression goodness-of-fit R2 = 0.27, p = 0.381; in E), IMS (R2 = 0.01, p = 0.715; in F), and IFN-γ/IMS (R2 = 0.52, p = 0.017; in G) values in the TCGA datasets.

Back to article page