Table 4.

Association relationship between hub genes with survival within breast cancer molecular subtypes

	Training data set (n = 87)				Validating dataset (n = 449)
	DMFS				DMFS				RFS
	luminal A		luminal B		luminal A		luminal B		luminal A		Luminal B
Gene	HR	p-value	HR	p-value	HR	p-value	HR	p-value	HR	p-value	HR	p-value
CDK1	0.90	8.6 × 10–1	1.83	2.8 × 10–1	1.50	3.6 × 10–1	1.84	5.9 × 10–2	1.52	1.0 × 10–1	1.76	1.6 × 10–2
DLGAP5	1.74	3.5 × 10–1	2.80	8.2 × 10–2	1.44	4.0 × 10–1	0.91	7.8 × 10–1	1.61	6.7 × 10–2	1.08	7.3 × 10–1
MELK	1.88	2.8 × 10–1	2.60	9.0 × 10–2	2.70	2.8 × 10–2	2.04	2.8 × 10–2	2.13	5.4 × 10–3	1.91	6.4 × 10–3
NUSAP1	1.25	7.0 × 10–1	2.88	7.5 × 10–2	2.23	7.0 × 10–2	1.12	7.2 × 10–1	1.50	1.2 × 10–1	0.92	7.0 × 10–1
RRM2	1.09	8.9 × 10–1	1.16	7.9 × 10–2	1.41	4.4 × 10–1	1.08	8.1 × 10–1	1.91	1.4 × 10–2	1.22	3.8 × 10–2

Recurrence free survival (RFS), Distant metastasis free survival (DMFS). Hazard ratios (HRs), 95% confidence intervals (CI), and p-values were calculated using Cox proportional hazards regression analysis after grouped the breast cancer patients by the median of gene level.