Fig. 7: Correlation between human immune NAMPT, malignancy, inflammation, and outcome.

a qPCR of Nampt from the indicated human myeloid cell lines that were stimulated for 6 h with IFNγ, with Hs CXCL10 induction as a control for induction. Data are the average of N = 3 biological replicates, and are representative of at least three independent experiments. p values were determined by a two-tailed unpaired t-test: *p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.00005. Error bars represent SEM. b Kaplan–Meier plots of TCGA full 30-year (left) or first 10-year (right) human melanoma patient survival data showing curves from median-divided NAMPT expression, that is, from patients with tumors expressing the highest vs. lowest 50% of NAMPT expression. The right plot is from data that were further stratified with respect to both the highest vs. lowest 10% of NAMPT expression levels and by classifying tumor samples cytologically as containing high vs. low immune infiltration. p values are reported within the figure panels for the difference between high and low NAMPT expression plots for either the keratin signature or immune signature-rich samples, and are from the log-rank (Mantel–Cox) test. c GSEA analysis for indicated pathways of gene expression data from TCGA melanoma transcriptomes in b. Gene rank order is determined by comparison of each gene’s expression between samples with high vs. low NAMPT levels. NES normalized enrichment score, p_adj adjusted p value: two-tailed, corrected for multiple comparisons using Benjamini–Hochberg method. b, c are based on data from a total of 463 patients. d Metagene signature analysis of NAMPT expression levels and macrophage subtypes in TCGA SKCM melanoma cohort, showing in immune rich tumors significant positive correlation of NAMPT abundance with inflammatory macrophage gene expression signatures (M1, middle). e Metagene signature analysis of NAMPT expression levels correlating with IFNγ induced responses from TCGA datasets for multiple tumor types in comparison to SKCM, where the y-axis shows Pearson correlation values. For d, e, p values are two-tailed, from tests for positive slopes of correlations between the two variables.