FIGURE 4 from An Early Neoplasia Index (ENI10), Based on Molecular Identity of CD10 Cells and Associated Stemness Biomarkers, is a Predictor of Patient Outcome in Many Cancers

ENI10 predict pan-cancer survival independently of the cell cycle and more efficiently than other SC–derived signatures. A, ssGSEA score of the ENI10 signature in all tumors and normal samples from TCGA database. B, Correlation of CD10 signature score and survival in TCGA's Pan-Cancer Atlas. All tumor samples were pooled and the effect of the CD10-signature score discretized by deciles on survival outcome was evaluated from Cox models stratified on cancer types, using unadjusted (black marks), adjusted on age alone (blue marks, modeled with a 3-degree polynomial spline) or with a supplemental stratification term for stage (I/II/III/IV; green marks) or grade (1/2/3/4; red marks) pathologic scoring systems. Dots show the HR for PFI and adjacent bars the 95% confidence interval. C, Left: Overall Pan-Cancer analysis of the correlation between the ENI10 score and survival in TCGA's Pan-Cancer Atlas. All tumor samples were pooled and the effect of the ENI10 score discretized by deciles on survival outcome was evaluated from Cox models stratified on cancer types. Right: Same analysis with a reduced ENI10 signature were genes known to be regulated during the cell cycle were removed. D and E, Overall Pan-Cancer analysis of the effect of CD10 score on survival in TCGA's Pan-Cancer atlas. All tumor samples were pooled and the effect of the CD10 enrichment score discretized by deciles on survival outcome was evaluated using a multivariable Cox model including as covariables both the CD10 score and the Smith and colleagues (D) or Pece and colleagues (E) signatures. Dots show the hazard ratio for PFI and adjacent bars the 95% confidence interval.