During the past decade, several high-throughput analytical methods have been developed, and most of these are being explored as potential diagnostic tools. Gene expression profiling with DNA microarrays or with multiplex polymerase chain reaction are the methods closest to being of clinical use. Prediction of clinically meaningful response to particular chemotherapy regimens or drugs remains a persistent challenge. There are established clinical and histopathologic predictors of prognosis for breast cancer, but there is no test to assist in selecting the optimal chemotherapy regimen for patients. Here we review recent advances in the application of gene expression profiling to chemotherapy response prediction.