OBJECTIVE A pilot study was designed to evaluate the safety and efficacy of a novel regimen of hypofractionated intensity-modulated radiotherapy (RT) in the adjuvant treatment of primary glioblastoma multiforme (GBM). The rationale of the study was to combine the potential radiobiologic advantage of hypofractionation to GBM with a highly conformal radiotherapeutic technique. The study was designed to measure the acute and chronic morbidity of patients treated with this regimen, response of GBM to the treatment, overall survival, and time to disease progression after therapy completion. METHODS Twenty eligible patients were accrued between February 1999 and May 2000 for the study. All patients had Karnofsky performance scores of >/=70. All patients were treated with intensity-modulated RT using the NOMOS Peacock system. A dose of 50 Gy was delivered in 5-Gy daily fractions within 2 weeks to enhancing primary disease, residual tumor, or surgical cavity. Simultaneously, 30 Gy was prescribed in 3-Gy daily fractions to surrounding edema. The time to progression was measured with serial neurologic examinations and MRI or CT scans after RT completion. Acute and late toxicity was graded using Radiation Therapy Oncology Group neurotoxicity scores. RESULTS Of the 20 patients, 18 were evaluated for outcome. The median time to disease progression was 6 months after RT completion. The median overall survival was 7 months after treatment completion. All recurrences were within 2 cm of the operative bed. Neurotoxicity during therapy was minimal, with all patients experiencing Grade 0 or 1 toxicity. Late toxicity included 10 patients with Grade 0, 2 patients with Grade 2, and 3 patients with Grade 4 toxicity, manifesting as brain necrosis requiring surgical reexcision. The survival of the 3 patients with brain necrosis was 23, 20, and 9 months. Mortality in all cases was the result of tumor recurrence, with no mortality resulting from brain necrosis. CONCLUSIONS This regimen of hypofractionated intensity-modulated RT did not improve the time to disease progression or overall survival compared with historical experience using conventional fractionation. However, the treatment duration was reduced from 6 weeks to 2 weeks, which may be of palliative benefit in certain subsets of patients. This treatment regimen demonstrated a greater incidence of brain necrosis requiring surgical intervention; however, the 3 patients experiencing this toxicity had longer survival times. Future investigation may be useful to determine which fraction size may be optimal for GBM when highly conformal RT is used in the adjuvant setting.