Objective: The objective of this study was to determine predictive factors for local control (LC) of brain metastases (BM) treated with Linac-based stereotactic radiosurgery (LB-SRS).
Methods: Between January 1994 and July 2001, 80 patients (126 BM) underwent LB-SRS. All patients had follow-up imaging with computed tomography (40%) or magnetic resonance imaging (60%). Most patients had either lung (41%) or renal cell (20%) cancer. The median SRS prescription dose was 18 Gy (range, 10-21 Gy). Most patients (86%) also received whole-brain radiotherapy (WBRT). LC was defined as the absence of enlargement of the BM on follow-up scans. Actuarial LC analyses were performed by the method of Kaplan-Meier and compared with the log-rank test. Factors analyzed included histology, volume, prescription dose, maximum and minimum tumor dose, target volume ratio, number of arcs and isocenters, total degrees, and WBRT. Multivariate analysis was accomplished.
Results: At a median follow up of 8.8 months, 11 BM failed (8.7%). The 1-and 2-year actuarial LC rates were 88.6% and 77.2%, respectively. The most significant factors correlated with LC were prescription (P = 0.0004) and minimum tumor (P = 0.002) doses, and tumor volume (P = 0.04). On multivariate analysis, the sole factor correlated with LC was minimum tumor dose (P = 0.03).
Conclusion: Our results confirm that LB-SRS is associated with excellent LC rates in the majority of patients treated. However, particular attention should be given to minimum target dose to ensure optimal outcome.