Gene-expression microarrays provide new prognostic and predictive tests for breast cancer

Wide availability of systemic therapy agents has led to a considerable decline in mortality from breast cancer. However, the biology of breast cancer remains poorly understood. Currently, highly accurate markers to predict prognosis and probability of response to a given systemic therapy on an individual basis are lacking, and routinely used clinicopathologic variables fail to fully capture the heterogeneity of breast cancer. As a result, many patients are overtreated, whereas others may not receive the necessary therapy. It has been hypothesized that molecular differences in breast cancers might account for the heterogeneous potential in growth, invasion and metastasis in each individual tumor. Gene-expression microarrays have been extensively applied in breast cancer research in the hope that a combination of multiple genes (i.e., gene signatures) will more informatively predict disease outcome and response to a specific systemic therapy. This technique holds substantial promise for optimizing clinical decision making and tailoring therapeutic regimens to individual patients in the near future. This review focuses on the recent progression in feasibility and reliability of gene-expression microarrays in identifying new prognostic and predictive indicators of breast cancer.