Treatment of brain tumors is increasingly informed by biomarkers. One use is to appropriately group tumors with similar genetic/genomic characteristics and to design trials separately for the individual groups. The biomarker's use is to predict patient response so that the most effective treatment can be selected for patients based on their tumor characteristics. Trial designs that recruit unselected patients are poorly suited for identifying treatments effective only in subsets of patients given the relatively small numbers of patients available for trials. Investigators are beginning to use different designs that better account for tumor heterogeneity. In this article, an overview of the role of biomarkers in brain tumor trials is presented in the context of existing clinical trials as well as trials that may be launched within the next several years.