TPS102 Background: Genomic assays have been shown to provide the potential for personalized approaches to breast cancer therapy. Full-transcriptome assays using microarrays are challenging, but have the advantage of providing multiple prognostic and predictive signatures in one assay. We have developed genomic predictive signatures for various clinically relevant cancer characteristics, including sensitivity to cytotoxic chemotherapies. We have investigated the analytic performance of these signatures in biologic replicates, and begun applying them in this prospective randomized clinical trial. Methods: Eligible patients have intact HER2 negative primary breast cancers with tumor size over 1.5 cm and any clinical lymph node status. A fresh frozen sample of the primary tumor must be obtained. After pathology assessment, RNA is analyzed on an Affymetrix H133 Plus 2.0 microarray. The expression data is used to predict doxorubicin (A) and docetaxel (T) sensitivity using previously defined signatures and analy...