RRHGE: A Novel Approach to Classify the Estrogen Receptor Based Breast Cancer Subtypes

Table 1

Microarray datasets used in this study.

	*Desmedt et al. [20]; (GSE7390)**	van de Vijver et al. [11]**	*Loi et al. [21]; (GSE6532)	*Sabatier et al. [22]; (GSE21653)	*Schmidt et al. [23]; (GSE11121)
Platform	HG-U133A	Agilent human genome	HG-U133A, HG-U133B	HG-U133Plus2.0	HG-U133A
Samples	198	295	327	255	200
ER
ER+ (no. of samples)	134	226	263	150	156
ER− (no. of samples)	64	69	45	102	44
Tumour grade
Grade 1 (no. of samples)	30	—	52	44	29
Grade 2 (no. of samples)	83	—	158	88	136
Grade 3 (no. of samples)	83	—	57	116	35
Metastasis Free Survival
Yes (no. of samples)	62	101	70	81	46
No (no. of samples)	136	194	224	160	154
Age (in Years)
≤40 (no. of samples)	42	—	19	49	—
41–70 (no. of samples)	156	—	241	171	—
>70 (no. of samples)	0	—	55	34	—
Average (in Years)	46	—	59	54	—
Total Samples	1253 ( 929 (ER+) and 324 (ER−))
Total samples selected in our Study (on the basis of histologic grade and receptor status)	958 (703 (ER+) and 255 (ER−))

Patients with missing histologic grade and estrogen receptor status based information are excluded from the training sets. ∗The datasets used in our training sets; ∗∗The testing sets.