Efficacy and Toxicity Profile of Brain Metastasis Treatment with Mask-based, Fractionated Stereotactic Radiosurgery

Abstract

Stereotactic radiosurgery (SRS) is increasingly being used to treat patients with a limited number of brain metastases given its overall low toxicity profile compared to whole brain radiation therapy (WBRT). However, many patients are ineligible for single-fraction SRS due to tumor size or unfavorable location. In this study, we sought to evaluate the efficacy and toxicity of frameless, fractionated SRS on the Gamma Knife for the treatment of brain metastases. Medical records from forty-eight patients with known brain metastases, obtained from two academic institutions were retrospectively reviewed. A total of 89 brain metastases treated between 2016-2019 were analyzed. Stereotactic MRI was obtained in all cases prior to treatment to assist with target volume delineation. A thermoplastic mask was used for immobilization. An on-board cone-beam CT (CBCT) was used as a stereotactic reference scan and was co-registered to the MRI for target localization. Daily CBCT was used for verification of positioning with infrared tracking used throughout the procedure for motion management. Patients were treated with 3-5 fractions to 25-30 Gy, with a median BED10 of 43.2 Gy (range: 28-51.3 Gy), to the 50% isodose line (range 45%-96% IDL). No planning target margin was utilized. Kaplan Meier method was used to estimate freedom from failure and overall survival. Cox regression analyses were performed to identify prognostic clinical or treatment characteristics. The primary cancer histologies included breast (37%), melanoma (11.2%), lung (18%), and other (33.7%). Median follow up time after SRS was 6.39 months overall. Among the 89 treated lesions, the median diameter was 3 centimeters (cm) (range 0.3-12.41 cm) with a target volume of 9.90 cc (range 1-25 cc). The local control at 6, 9, and 12 months was 83.3%, 66.1%, and 61.0%, respectively. Following the patient’s first treatment, the freedom from any intracranial failure at 6, 9, and 12 months was 49.9%, 41.6%, and 32.4%. Rates of freedom from leptomeningeal failure at 6, 9, and 12 months was 83.4%, 78.2%, and 72.6%. Overall, radiation necrosis occurred in 4 (4.5%) of the treated lesions. There was one case of radiation-related grade 3 toxicity, which occurred in a patient that had prior WBRT. Median overall survival was 8.1 months and was 68.3% at 6 months, 48.4% at 9 months, and 45.1% at 12 months. Higher KPS was associated with improved survival (p<0.05). Factors such as age, gender, histology, tumor location, resection status, and BED were not found to be prognostic for local control or survival (p>0.05). Given the reasonable toxicity ratio and steep dose fall off of SRS, efforts to optimize dose and fractionation for larger tumors or those in unfavorable locations is warranted to improve the local control. Stereotactic-guided frameless fractionated radiosurgery is a safe and effective option in patients who otherwise are not ideal candidates for single fraction SRS.