Table 2.
Variable | n | Univariate analysis | Multivariate analysis | ||
---|---|---|---|---|---|
HR (95 % CI) | p-value | HR (95 % CI) | p-value | ||
Tumor size | |||||
< 20 mm | 263 | 1.00 | <0.0005 | 1.00 | 0.006 |
> 20 mm | 589 | 2.05 (1.47–2.86) | 1.62 (1.15–2.28) | ||
Histologic grade | |||||
1-2 | 443 | 1.00 | <0.0005 | 1.00 | 0.070 |
3 | 409 | 1.86 (1.41–2.46) | 1.33 (0.98–1.80) | ||
Lymph node status | |||||
Negative | 453 | 1.00 | <0.0005 | 1.00 | <0.0005 |
Positive | 399 | 2.25 (1.70–2.98) | 1.83 (1.38–2.45) | ||
PAM50 subtype a | |||||
Luminal-likeb | 734 | 1.00 | <0.0005 | 1.00 | NS |
Basal-like | 118 | 1.91 (1.37–2.68) | 1.26 (0.87–1.82) | ||
33P hypoxia stromal signature | |||||
Q123 | 639 | 1.00 | <0.0005 | 1.00 | 0.001 |
Q4 | 213 | 2.120 (1.60–2.81) | 1.67 (1.23–2.27) |
Statistical test: Two-sided Wald test. Adjustment for multiple testing was not performed.
CI confidence interval, HR hazard ratio, n number of patients, NS not significant.
aOnly patients with luminal A, luminal B and basal-like breast cancers were included (n=852, METABRIC-Discovery cohort).
bLuminal-like: patients with luminal A and luminal B breast cancer subtypes.