Abstract
BACKGROUND: The optimal fractionation schedule of radiotherapy (RT) for Glioblastoma multiforme (GBM) is yet to be determined. We aim to compare different fractionation regimens and identify prognostic factors to better tailor RT for newly diagnosed GBM patients. METHODS: All data for patients who underwent surgery for GBM between January 2005 and December 2012 were compiled. Clinical information was collected using patient charts and government registry. Cox analysis was used to identify variables affecting survival and treatment outcome. RESULTS: The median follow-up time was 13.2 months. Two hundred and seventy-six patients met the inclusion criteria, including 147 patients in the 60 Gy in 30 fractions (ConvRT) group, 86 patients in the 60 Gy in 20 fractions (HF60) group, and 43 patients in the 40 Gy in 15 fractions (HF40) group. Median survival (MS) was 16.0 months with a median progression-free survival (PFS) of 9.23 months in the ConvRT group. This was comparable to outcome in the HF60 group with MS 15.0 months and a median PFS of 9.1 months. Patients in the HF40 group had MS of 8 months, with a median PFS 5.4 months. Cox analysis showed no significant difference in OS between the ConvRT and HF60 groups but worse outcome in the HF40 group (HR 2.22, P=0.04). MGMT methylation, extent of resection, use of chemotherapy, and repeat surgery were found to be significant independent prognostic factors for survival. CONCLUSIONS: HF60 constitutes a safe RT approach that shows survival comparable to standard RT while allowing for a shorter treatment time.