The aim of this study is to evaluate the efficacy of stereotactic radiotherapy boost (SRB) in patients with glioblastoma multiforme (GBM) by comparing two different regimens, single dose or fractionated treatment. Between December 1994 and January 2000, 24 patients with GBM were treated with SRB in conjunction with external beam radiotherapy (EBRT). Fourteen patients (58%) were treated with stereotactic radiosurgery (SRS) and 10 patients (42%) with fractionated stereotactic radiotherapy (FSRT). Median interval between EBRT and SRS or FSRT was 1.4 months (range -0.4-3.9 months). Actuarial survival rates of the entire 24 patients at one and two years following SRB were 63% and 34% respectively, with median survival time of 16 months. Variables predicting survival were age, extent of surgery, re-operation and the RTOG (Radiation Therapy Oncology Group) classes based on recursive partitioning analysis (RPA). In comparison to historical controls, improved survival benefit after SRB was observed. The median survival times for the RTOG classes 4, 5, and 6 were 28.3, 10.3, and 6.0 months following EBRT+SRB, respectively. Expected values for these classes after EBRT are 11.1, 8.9, and 4.6 months, respectively. This improvement in survival was seen predominantly for the RTOG class 4. There was no difference in survival between SRS and FSRT treated groups. Late complications developed in 4 patients in the SRS group and 1 patients in the FSRT group. Our retrospective data suggest that SRB in conjunction with EBRT may improve survival in patients with GBM with median survival time of 16 months, when compared to historical controls of the RTOG data following EBRT. The addition of SRB appeared to improve the median survival most demonstrably in RTOG RPA class 4 patients. SRS and FSRT are equally effective with similar median survival, but potentially less late complications associated with FSRT. Since this is a nonrandomized study, further investigation is needed to confirm this and to determine an optimal dose/fractionation scheme.