Search

This article reports on a study of whether breast cancer prognosis can be refined through combining genomic information with clinical and pathological risk factors. The article describes the retroactive study of early-stage breast carcinoma patients and the predictors that were applied to characterize relevant information. The study concluded that gene expression signatures can be used for refining prognosis, but that more studies will be needed to determine their value for individualizing therapeutic strategies. The article also discusses the prognosis of cancer in general, the effect of genomic technology on clinical decisions, and the role of chemotherapy and radiotherapy in cancer treatment.

Gene expression signatures have been developed in an effort to dissect the biologic phenotypes of malignancies. These signatures have tremendous power to identify new cancer subtypes and to predict clinical outcomes based on patterns of gene expression. Expression profiles specific to a phenotype can be derived from in vitro data, as well as from patient cohorts with clinically relevant outcomes. In addition to predicting outcomes in non-small-cell lung cancer (NSCLC), similar techniques have been used to develop gene expression signatures that predict sensitivity or resistance to specific chemotherapeutic agents. Additionally, expression data have been used to identify oncogenic pathway deregulation to help direct the use of targeted agents. Used in combination, it is likely that gene expression signatures will help assess prognosis and may also be of value in guiding the use of cytotoxic and targeted therapy in NSCLC. Clinical trials are ongoing to validate these predictive gene expression signatures in a prospective manner. [ABSTRACT FROM AUTHOR]