Accelerated radiation therapy of 2 Gy b.i.d. to 70 Gy in 3.5 weeks in glioblastoma multiforme

Review of the literature suggests that repopulation during radiation therapy (RT) might be an important cause of failure in glioblastoma multiforme (GBM) and that shortening the overall treatment time might be beneficial. We undertook this study in order to evaluate the safety and effectiveness of accelerated RT (ART) in GBM. We treated 48 patients with histologically proven GBM by ART (patients with anaplastic astrocytoma were excluded). Median age was 60 years (range 31–74 years). Karnofsky performance status (KPS) prior to ART ranged from 40 to 90 (median 55). ART started 7–22 days (median 14 days) after resection/biopsy. Whole brain ART 40 Gy in 2 weeks was followed by cone‐down ART to 30 Gy in 1.5 weeks. The interfraction interval was 4–6 hr. No chemotherapy was given. The treatment was well tolerated and the acute toxicity was quite unremarkable. The observed survival rate at 12, 18, and 24 months was 30%, 11%, and 5%, respectively. The median survival time was 44 weeks. The overwhelming cause of death was persistent/recurrent cancer at the original tumor site. Serial computed tomographic (CT) scans typically revealed reduction in the size of the tumor over 2–3 months initially, followed by a period of stability and then rapid recurrence and death. Late toxicity was obvious only in one patient who developed bilateral blindness 1 year after treatment, having received a dose of 55 Gy in 3.5 weeks to the optic chiasm. In conclusion, 70 Gy in 3.5 weeks could be delivered to patients with GBM without excessive acute or late toxicity. The short treatment time is an advantage for these patients with a limited life expectancy. The survival rates are comparable to, but not notably better than, several contemporary series with conventional RT plus chemotherapy and/or halogenated pyrimidine radiosensitizers. While alternative dose/fractionation regimes might be more successful, the results of this study failed to support our hypothesis that repopulation during RT is an important reason for the failure of treatment in GBM. © 1995 Wiley‐Liss, Inc.