Several treatment options, including endocrine therapy, chemotherapy, and targeted therapy, have been shown to improve survival of breast cancer patients. Currently, clinical tests for predicting cancer response are not available, and individual markers have shown little predictive value. Several gene expression profiling studies have been carried out in the attempt to identify predictive signatures. The neoadjuvant setting revealed to be ideal for this purpose because it allows the direct assessment of response to treatment, and tumor is readily available for multiple time point biopsies. Although the results are promising, at the moment, none of these signatures has been proven to be of sufficient discriminatory power to be used in clinical setting. More effective therapies targeted to specific subsets of patients, accurate and standardized definition of therapeutic response, and properly designed clinical trials are required before microarrays can reliably be used as tools for clinical decision making.