Table 4. Multivariate survival analysis using Cox proportional hazards regression models.
Model | Variable | Category | Hazard ratio | 95% CI | p value |
---|---|---|---|---|---|
1 | pT stage | T1 vs. T2-4 | 3.397 | 1.403–8.223 | 0.007 |
pN stage | N0 vs. N1-3 | 4.024 | 1.610–10.057 | 0.003 | |
Lymphovascular invasion | Absent vs. present | 1.097 | 0.462–2.608 | 0.834 | |
Estrogen receptor | ≥90% vs. <90% | 1.735 | 0.811–3.710 | 0.156 | |
Progesterone receptor | ≥40% vs. <40% | 1.900 | 0.886–4.075 | 0.099 | |
Ki-67 index by average method | <18% vs. ≥18% | 2.431 | 1.088–5.434 | 0.030 | |
2 | T stage | T1 vs. T2-4 | 3.212 | 1.319–7.823 | 0.010 |
N stage | N0 vs. N1-3 | 4.038 | 1.614–10.100 | 0.003 | |
Lymphovascular invasion | Absent vs. present | 1.056 | 0.445–2.505 | 0.902 | |
Estrogen receptor | ≥90% vs. <90% | 1.634 | 0.764–3.494 | 0.205 | |
Progesterone receptor | ≥40% vs. <40% | 1.971 | 0.919–4.226 | 0.081 | |
Ki-67 index by hot spot method | <22% vs. ≥22% | 3.058 | 1.287–7.267 | 0.011 |
P values were calculated from Cox proportional hazards regression models with backward stepwise selection.
Abbreviation: CI, confidence interval