RADT-03. GOLDILOCKS AND THE THREE DOSES OF RADIOSURGERY FOR SMALL BRAIN METASTASES

Abstract

Abstract INTRODUCTION The optimal dose for single-fraction radiosurgery (SRS) for small brain metastases (BM) (< 2 cm) has not been established considerable variability in dose selection. METHODS Consecutive patients diagnosed with intact BM < 2 cm in maximum dimension treated with SRS to prescription doses of 24, 22, or 20 Gy on the Gamma Knife® between 3/2019-1/2021 were reviewed. Institutional dosing prescriptions were: 24 Gy for < 10 lesions, 22 Gy for 10-20, and 20 Gy for >20). Endpoints included freedom-from-local-failure (FFLF) and radionecrosis on a per-lesion basis calculated from date of SRS to event. Univariable and multivariable analyses were performed using the Kaplan-Meier method and Cox proportional hazards regression. RESULTS 133 patients with 863 BM met inclusion criteria. Prescription dose was 24 Gy for 416 (48.2%) lesions, 22 Gy for 266 (30.8%), and 20 Gy for 181 (21.0%). 20 Gy was associated (all p< 0.001) with younger age, presence of extracranial-disease, progressive extracranial-disease, lung primary, higher number of BM per SRS course (median 17 for 20 Gy, 14 for 22 Gy, 6 for 24 Gy), and smaller per-lesion tumor volume. With a median follow-up of 15 months, 76 (8.8%) LFs occurred. Actuarial 1-year FFLF was 91.2% (95%CI: 88.6-94.8%) for 24 Gy, 92.5% (89.1-95.9%) for 22 Gy, and 77.9% (71.2-84.6%) for 20 Gy (p< 0.001). On multivariable analysis, factors independently associated with adverse FFLF (all p< 0.05) were male sex, presence of extracranial-disease, breast or lung primary, and 20 Gy prescription dose (HR: 2.52, 95%CI: 1.34-4.76). The cumulative 1-year radionecrosis rate was 7.0% for 24 Gy, 4.0% for 22 Gy, and 3.8% for 20 Gy (p=0.035). CONCLUSION For BM < 2 cm, we propose a testable hypothesis that 22 Gy prescription dose may represent the “Goldilocks zone”- superior to 20 Gy regarding LF and superior to 24 Gy concerning radionecrosis.