(WHO) performance status (p < 0.0001), presence of neurological deficits (p = 0.0002), baseline administration of steroids (p < 0.0001), number of target lesions (p < 0.0001), tumour size (largest tumour diameter, p < 0.0001), and frontal tumour location (p = 0.02). 9 Considering these factors, patients who are eligible for radiosurgery at the time of their glioblastoma multiforme (GBM) recurrence generally have good performance status, single or few small tumours that do not require corticosteroids to manage symptoms. These are all recognized prognostic factors in patients with recurrent GBM. There have also been reports of prognostic factors specifically associated with improved outcomes after salvage radiation. These include younger age, higher Karnofsky performance status (KPS) and better recursive partitioning analysis (RPA) class and lack of steroid dependence, smaller and unifocal tumour targets, use of higher radiation prescription dose, the extent of pre-radiosurgery tumour resection, and use of concurrent chemotherapy. Increased time to tumour recurrence has also been associated with better outcomes. 10-16 Considering all these factors, the patients included in this study had particular favorable prognostic factors with a high average KPS of 85.2 prior to radiosurgery, small average tumour volume of only 4.4 cm 3 (range: 1.1-15.7 cm 3) and long duration between initial