Stereotactic radiosurgery for glioblastoma multiforme.

From February 1989 to August 1992, 26 patients who presented with an initial pathological diagnosis of glioblastoma multiforme underwent tumor debulking (17) or biopsy (9), stereotactic radiosurgery (SR) and standard radiation therapy (dose range 50-66 Gy) as part of their primary tumor therapy. Presenting characteristics included median age of 55 years (range 20-79), Karnofsky Performance Score (KPS) median 82.5 (20-100), and median tumor volume 18.6 cm3 (2.2-59.7). SR collimator size ranged from 2.25 to 4 cm with a central dose of 15-35 Gy. Isocenter location, collimator size and beam paths were individualized using three-dimensional software such that the maximum possible solid angle was subtended without exceeding a 20% tumor dose gradient. The mean follow-up was 10.9 months (6-19.5) with a median of 9.5 months. Statistical analysis was performed using Kaplan-Meier actuarial analysis developing predicted 12-month survival rates. There were no significant differences noted in patient survival for the parameters of biopsy versus debulking, single versus multiple isocenters, age, initial KPS, and patterns of steroid requirement. Radiographic recurrences were divided by location into central (within central SR dose) = 0, peripheral (within 1 cm of central dose) = 16, and distant (< 1 cm) = 4. Predicted 12-month survival was 24%, with a median survival of 9.5 months. These values are similar to previous results for surgery and standard radiotherapy alone [1]. The results suggest that radiosurgery, when used as a mode of primary therapy, offers little or no benefit in quality of life or survival as recurrences occur immediately outside or distant to the central SR field.