Because of the many genetic and epigenetic alterations that define a given tumor cell, and given the heterogeneity of these changes, between, as well as within a tumor class, progress in the understanding and treatment of cancer has been slow. The sequencing of the human genome, in combination with advances in robotics, computing, and imaging technologies, has resulted in rapid advances in the development of microarray methodology. This technology now places us in a position to simultaneously consider the consequence of all of these genetic changes through measurement of a large proportion of the complement of genes expressed in a given tissue at a given time. The power of this methodology for (1) the classification and identification of tumor classes, (2) gene discovery, (3) determining mechanisms of drug action, and (4) predicting drug response has now clearly been demonstrated: however, several challenges remain. First, there is a growing need for a standardization of the methodology, such that different datasets may be compared directly and meaningfully. The complete sequencing and annotation of the human genome may be the first step toward this attainable goal. Also, the databases generated should be made publicly available to facilitate further analysis by other researchers. To this end, the results of our studies are available at http://sequence.aecom.yu.edu/bioinf/Augenlicht/default.html. Second, a number of reported findings on tumor classification require further validation in independent patient data sets. Third, extensive clinical studies with appropriate patient follow-up are required to determine the validity of this method for the prediction of patient response to chemotherapy. Finally, the possibility needs to be considered that gene expression profiling may need to be combined with other global approaches such as proteomics and mutation screening analyses, for optimization of its potential. The advent of methodologies that enable gene expression profiling provides an opportunity to gain insights into the genetic makeup of a cancer cell on a global scale. Given the heterogeneity of this disease, such a global approach is likely to enhance significantly our understanding and management of this disease.