Skip to main content


Table 2 Univariate Cox regression modeling for the overall survival using each of the three gene models (A) and bivariate Cox regression modeling by combining two of the three gene models (B) in validation set I

From: A robust blood gene expression-based prognostic model for castration-resistant prostate cancer

A. Univariate analysis (individual gene model)    
  Concordance index P value (Likelihood ratio test) P value (Logrank test)
Wang_4genescore 0.81 0.0006 7×10−05
Olmos_9genescore 0.72 0.004 0.005
Ross_6genescore 0.68 0.028 0.026
B. Bivariate analysis (Combining two gene models)
  P value (Wald’s test)   
Wang_4genescore 0.042   
Olmos_9genescore 0.35   
Wang_4genescore 0.010   
Ross_6genescore 0.99   
Olmos_9genescore 0.054   
Ross_6genescore 0.40