The optimal management of patients with malignant gliomas begins with the accurate determination of the pathologic diagnosis based on adequate sampling of the tumor. Clear differences in prognosis and therapeutic options have been established for the various tumor grades and cellular classification. Current recommendations, on the basis of the results of a recent phase III randomized trial, are that patients with glioblastoma should have maximal surgical resection followed by concurrent radiation and chemotherapy with temozolomide. It is further recommended that patients then be treated with 6 to 12 months of adjuvant temozolomide. However, despite the shown improvement in survival with this chemoradiation regimen, the impact on outcome is modest. It is increasingly evident that a greater understanding of the molecular mechanisms of gliomagenesis is needed to improve treatments for these patients. Recent and ongoing investigations strongly indicate that specific molecular markers tremendously impact prognosis and often can predict response to treatment. For example, allelic loss of the 1p and 19q chromosome arms predicts a dramatic improvement in response to treatment and survival for tumors histologically classified as anaplastic oligodendroglioma. Future advances for treating primary brain tumors likely will be directly related to our ability to molecularly subcategorize tumors and customize therapy based on the molecular profile within each histologic type and grade of tumor. This is evident in preliminary data indicating that inactivation of the methyl guanine methyltransferase gene by hypermethylation of the promoter region specifically predicts a better tumor response rate to chemotherapies that alkylate DNA as their mechanism of action. Similarly, elucidation of overly active signal transduction pathways within tumor cells may provide an opportunity to select the optimal therapeutic regimen composed of modulators of these pathways, analogous to restricting the use of trastuzumab to breast cancers expressing the Her-2 receptor. Advances in treating primary malignant brain tumors will likely depend on collaborative clinical trials that are designed to select patients on the basis of histologic and molecular characteristics and to determine the optimal biologic dose of the best agent that can treat each specific tumor type.