Skip to main content

Table 4 Univariate and Multivariate Cox-regression model

From: An immune response gene expression module identifies a good prognosis subtype in estrogen receptor negative breast cancer

  ER- (n = 186) ER+ (n = 527)
  HR (95% CI) P HR (95% CI) P
LN 2.28 (1.36-3.85) 0.002 2.07 (1.47-2.90) <10-4
LI 1.06 (0.66-1.70) 0.9 1.50 (0.72-3.14) 0.58
IRM 2.02 (1.19-3.41) 0.009 1.25 (0.91-1.71) 0.19
LNa + IRM 2.16 (1.28-3.64) 0.004 2.10 (1.49-2.96) <10-4
LN + IRMa 1.93 (1.14-3.26) 0.015 1.29 (0.94-1.76) 0.11
LIa + IRM 0.86 (0.32-2.28) 0.76 1.75 (0.41-7.47) 0.45
LI + IRMa 2.05 (0.71-5.97) 0.19 0.57 (0.27-1.19) 0.13
LNa + LI + IRM 1.79 (0.70-4.62) 0.22 1.48 (0.68-3.19) 0.32
LN + LIa + IRM 0.84 (0.51-1.38) 0.72 1.65 (0.38-7.07) 0.5
LN + LI + IRMa 2.22 (0.76-6.50) 0.15 0.57 (0.27-1.20) 0.14
  1. The table summarizes the hazard ratio (HR), 95% confidence interval (CI), and log-rank test P values of univariate Cox proportional hazards regression models, with lymph node status (LN = 1/0 for LN +/-), level of lymphocytic infiltration (LI = 1 for low infiltration score, and LI = 0 for high infiltration score) and the classification based on the seven-gene immune response related module (IRM; 2 = down-regulation of module, 1 = upregulation of module) as predictors. aCorresponding values in the multivariate Cox models including LN, LI, and IR module as predictors. The table compares the values for estrogen receptor (ER)- and ER+ breast cancer.