A Systematic Review and Meta-Analysis of Radiation Necrosis Incidence in Brain Metastasis Treated by Gamma Knife and CyberKnife Stereotactic Radiosurgery.

Abstract

Brain metastases from non-central nervous system malignancies are increasingly common due to enhanced screening and treatment. Stereotactic radiosurgery (SRS), including Gamma Knife (GK) and CyberKnife (CK), is preferred over whole brain radiation therapy for solitary or oligometastatic brain metastases. Radiation necrosis (RN) is a significant complication. This systematic review compares RN incidence after GK-SRS and CK-SRS. Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines, we searched PubMed, MEDLINE, EMBASE, and Web of Science using terms like "Gamma Knife," "CyberKnife," and "radiation necrosis." Included studies reported RN in adults treated with GK-SRS or CK-SRS. Data on patient numbers, lesion size, SRS dosage, RN incidence, and follow-up were extracted. A random-effects meta-analysis was conducted, with heterogeneity assessed via Higgins' I² test. Subgroup and sensitivity analyses examined publication bias, study size, and radiation dose. From 5245 articles (3562 GK, 1683 CK), 52 studies (37 GK, 15 CK) with 9186 patients were analyzed. Pooled RN incidence was 7.5% (95% confidence interval [CI]: 5.8%-9.7%) for GK-SRS and 6.8% (95% CI: 4.5%-10.1%) for CK-SRS (P = 0.6659). High heterogeneity (I² = 81.2% for GK, 71.2% for CK) persisted across sensitivity analyses. Egger's test indicated publication bias (P < 0.0001 for GK, P = 0.0004 for CK). Meta-regression found no significant RN predictors. GK-SRS and CK-SRS show comparable RN incidence, but heterogeneity and bias limit conclusions. Standardized diagnostics and prospective studies are needed.