Gene Expression Profiles for Predicting Metastasis in Breast Cancer: A Cross-Study Comparison of Classification Methods

Table 1

Overview of datasets used.

Dataset	Chip	Probes (K)	Patients	Outcome	Treatment	Define genes	Internal CV	External validation train	External validation test	Reference
Amsterdam	Agilent/Rosetta	25	295, N^+, N^−	DM	None, et, ct	√				[31]
Amsterdam (AM); (subset of the above)	Agilent/Rosetta	25	151, N^−	DM	None	√	√	√		[31]
Rotterdam (RO)	Affymetrix; HG-133A	22	286, N^−	DM	None	√	√	√		[28]
HUMAC	Spotted oligonucleotides	29	60, N^−	ME	None	√				[4]
Huang	Affymetrix; 95av2	12	52, N^+	RE	ct	√				[27]
Sotiriou 2003	Spotted cDNA	7.6	99, N^+/N^−	RE	et, ct	√				[24]
Sotiriou 2006	Affymetrix; HG-133A	22	179, N^+/N^−	DM	et	√				[25]
Uppsala	Affymetrix; HG-133A+B	44	236, N^+/N^−	DF	None, ct, et	√				[6]
Stockholm	Affymetrix; HG-133A+B	44	159, N^+/N^−	RE	None, ct, et	√				[23]
TRANSBIG (TR)	Affymetrix; HG-133A	22	147, N^−	DM	None				√	[32]
Mainz (MA)	Affymetrix; HG-133A	22	200, N^−	DM	None				√	[33]

The columns show the following: “dataset”: the individual names for the eight datasets; “chip”: microarray chip used; “probes”: number of probes on the chip measured in thousands (K = 1000); patients': number of patients in the study and their nodal status (N⁺ and N⁻ indicates number of node-positive and -negative patients; “outcome” covers the clinical outcome being DM: distant metastasis, ME: metastasis, RE: relapse, and DF: death from breast cancer; “treatment” shows patient treatments abbreviated by et: endocrine therapy, ct: chemo therapy, and none: no adjuvant therapy.