Although Breast Cancer (BC) has been considered for many years as a single entity with a common management and treatment, it is actually a extremely heterogeneous disease which includes at least 4 or 5 very different subtypes. The first step in the recognition of the heterogeneity of BC was the demonstration of the presence of functional hormonal receptors (HR) in nearly two thirds of breast cancer specimens. This finding, which established a first classification of BC in two clear subtypes (HR-positive and HR-negative) was followed by the demonstration of many other differential features. The her2/neu gene alteration, present in nearly 20% of BC tumors, is probably the most relevant of them, but certainly not the only one. The development of new technologies and, in particular, the use of complementary DNA (cDNA) microarrays will allow us now the simultaneous analysis of thousands of genes and the establishment of new, more refined BC subtypes based on gene expression profiles/genetic fingerprints. This review discusses the practical applications of molecular analysis of BC, which can be classified in four categories: 1. Establishment of a new molecular taxonomy of breast cancer. 2. Definition of prognostic factors/prognostic indexes based on molecular/genetic peculiarities. 3. Prediction of response to diverse antitumoral treatments. 4. Identification of molecular targets that allows the development of new tailored antitumor treatments.