摘要：

Breast cancer is a heterogeneous disease, encompassing several histological types and clinical behaviors.Current histopathological classification systems arebased on descriptive entities with prognostic significance.Few prognostic and predictive markers beyondthose offered by histopathological analysis are available.High-throughput molecular technologies are reshapingour understanding of breast cancer, of which microarray-basedgene expression has received most attention.This method has been used to derive a moleculartaxonomy for breast cancer, which has provided interestinginsights into the biology of the disease. Classprediction studies have generated a multitude ofprognostic/predictive signatures, which herald thepromise for an improvement in treatment decisionmaking. However, most of the signatures developed todate seem to have discriminatory power almostrestricted to estrogen receptor-positive disease. Thisreview addresses the contribution of gene expressionprofiling to our understanding of breast cancer and itsclinical management.

展开