Breast cancer is one of the most frequent and deadly cancers in industrialized countries. The identification of accurate biomarkers that improve the screening, diagnosis, prognostication and prediction of therapeutic response or toxicity, and the identification of novel molecular therapeutic targets are crucial. Today, high-throughput molecular techniques permit investigators to systematically interrogate the genome, transcriptome, and proteome of cancer cells. During the past decade, mRNA expression profiling has been successfully applied to the molecular characterization of breast cancers. Application of proteomics-based techniques is also considered crucial for detecting new biomarkers. In this review, we present the proteomics-based methods that have been applied to date to breast cancer samples for diagnostic and prognostic purposes. Despite their current limits, these pioneering techniques are promising. The most important results as well as the current limitations and perspectives are summarized and discussed.