Parameters influencing local control of meningiomas treated with radiosurgery

To identify parameters that influence local control after stereotactic radiosurgery (SRS) for meningiomas we retrospectively analyzed all meningiomas treated with Gamma Knife SRS at our institution from 1991 to 2007. Endpoints were measured from the date of SRS and estimated using the Kaplan-Meier method; subgroups were compared with log-rank tests. Sex, performance status, age, SRS setting, radiation dose, grade, volume and location were evaluated with univariate and multivariate Cox proportional hazards analyses. Of 280 patients with 438 tumors, 264 patients with clinical follow-up and 406 tumors with imaging follow-up were analyzed (median follow-up: 75.9 months). Thirty-seven percent of the tumors had no tissue diagnosis, 32 % were benign (grade I), 12 % atypical (grade II), and 19 % malignant (grade III). Five-year freedom from progression (FFP) was 97 % for presumed meningiomas, 87 % for grade I tumors, 56 % for grade II tumors, and 47 % for grade III tumors (p < 0.0001). Five-year FFP probabilities for upfront SRS versus SRS at recurrence after surgery versus SRS at recurrence after RT were 97, 86, and 38 %, respectively (p < 0.0001). Univariate analysis revealed that higher grade, larger target volume (median diameter: 2.4 cm) and SRS setting were associated with poorer FFP. Only target volume and SRS setting remained significant on multivariate analysis. Local control of presumed and grade I meningiomas is excellent with Gamma Knife SRS, but is suboptimal with high-grade tumors as well as for those treated at recurrence after RT or of large volume.