Introduction

DNA microarrays have proven to be a very powerful research tool. Since the term “DNA microarray” was first introduced in 1995 [1], there have been more than 12 000 publications on this new technology. The anticipated clinical utility has always been implied if not directly stated and it is notable that one of the first proof-of-principle studies to demonstrate clinical utility involved gene expression profiling of pediatric leukemias [2]. Acute leukemias have been ideal for exploring the potential of microarray technology because of the easy accessibility and purity of the neoplastic cells. This, and other early studies, investigated the differential gene expression of leukemia cells as a tool for predicting prognosis and therapy-related toxicity, and for identifying new mechanisms of pathogenesis and potential targets for therapy. Indeed, some investigators have suggested that gene expression microarray of acute leukemia will soon replace the standard diagnostic tools such as morphology, cytogenetics, and flow cytometry, in addition to establishing a molecular classification that would usher in the era of “personalized medicine.” However, despite these early successes and the consequent excitement they generated, some investigators have questioned accuracy and reproducibility of microarray data, arguing that the data are at best a starting point for investigations, and translation to clinical practice will require extensive validation using more established, reproducible technologies. Nevertheless, proponents of the gene expression microarray approach continue to press on with refinements in targets, techniques, and technology, and clinical trials have begun.