Extended Data Fig. 2: Cox PH models with straight-line assumption to assess the association between biomarkers and DFS. | Nature Medicine

Extended Data Fig. 2: Cox PH models with straight-line assumption to assess the association between biomarkers and DFS.

From: Adjuvant nivolumab in muscle-invasive urothelial carcinoma: exploratory biomarker analysis of the randomized phase 3 CheckMate 274 trial

Extended Data Fig. 2

a. Cox PH model with straight-line assumption for CD8 digital IHC. Overall effect: P < 0.001. Interaction effect: P = 0.590. Shaded areas indicate 95% CI. b. Cox PH model with straight-line assumption for CD4 gene expression. Overall effect: P < 0.001. Interaction effect: P < 0.001. Shaded areas indicate 95% CI. c. Cox PH model with straight-line assumption for the 25-IFNγ GES. Overall effect: P < 0.001. Interaction effect: P = 0.002. Shaded areas indicate 95% CI. d. Cox PH model with straight-line assumption for TMB score. Overall effect: P = 0.001. Interaction effect: P = 0.061. Shaded areas indicate 95% CI. e. Cox PH model with straight-line assumption for the 19-FTBRS GES. Overall effect: P = 0.059. Interaction effect: P = 0.973. Shaded areas indicate 95% CI. f. Cox PH model with straight-line assumption for the 8-EMT/Stroma Core GES. Overall effect: P = 0.640. Interaction effect: P = 0.597. Shaded areas indicate 95% CI. g. Cox PH model with straight-line assumption for the 9-TGFβ Activation GES. Overall effect: P = 0.740. Interaction effect: P = 0.943. Shaded areas indicate 95% CI. h. Cox PH model with straight-line assumption for the 13-High risk TGFβ GES. Overall effect: P = 0.005. Interaction effect: P = 0.204. Shaded areas indicate 95% CI. i. Cox PH model with straight-line assumption for the 4-TLS gene signature. Overall effect: P = 0.006. Interaction effect: P = 0.025. Shaded areas indicate 95% CI.

Back to article page