Breast cancer is a major health problem in developed countries. Pathological and clinical heterogeneity, partly responsible of therapeutic failures, reflects its poorly documented complex and combinatory molecular basis. Thorough molecular typing could allow not only better tackling that diversity and improving the current prognostic classifications, but also could help in the identification of new molecular therapeutic targets. The recently developed DNA microarray technology allows the analysis of the RNA expression of several thousands of genes simultaneously in a sample. Recent studies have shown the promising prognostic impact of gene expression profiling in breast cancer by identifying new prognostic subclasses unidentifiable by conventional parameters.