Optimal Scheme in Hypofractionated Stereotactic Radiation Therapy for Brain Metastases

Abstract

Stereotactic radiosurgery (SRS) is common technique used in the treatment of brain metastases. In lieu of single-fraction SRS, hypofractionated stereotactic radiation therapy (HF-SRT) is employed for large metastases and/or for those in sensitive areas of the brain in order minimize toxicity while maintaining effectual local control (LC) rates. There is currently no consensus on an optimal dose-fractionation, with most of the surrounding evidence being in the form of retrospective institutional studies resulting in a variety of utilized fractionation schemes. Some of these studies have suggested that BED10 may predict for improved LC. Our hypothesis is that HF-SRT delivered to BED10 ≥50 Gy would confer improved LC without a significant detriment in radionecrosis (RN) rates.A search of medical literature in PubMed/MEDLINE, Embase, and Cochrane databases was conducted using Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Observational Studies in Epidemiology (PRISMA) guidelines. Search terms included 'brain metastases and 'radiosurgery' yielding studies which were individually reviewed for inclusion. To meet inclusion, studies must have 1) utilized hypofractionated stereotactic radiosurgery techniques defined as ≥2.5 Gy per fraction in the treatment of brain metastases, 2) described LC and RN rate data, and 3) administered radiosurgery as definitive or post-operative treatment. Linear regression of the logarithmically transformed 1-year LC rate weighted by the number of lesions in each study was performed. In addition, a non-parametric, exact Mann-Whitney U-test was used to test a potential difference in 1-year LC or RN rate according to whether studies achieved a median BED10 ≥50 Gy.A total of 27 studies spanning 2000 to 2019 comprising of 2399 lesions met our inclusion criteria and were analyzed. Median BED10 across all studies was 51.3 Gy (range 37.5 to 69.3), with 836 lesions receiving BED10 < 50 Gy and 1563 lesions BED10 ≥50 Gy. Median total dose and number of treatments were 30 Gy (range 23.1 to 40) and 5 fractions (range 3 to 10), respectively. There was no statistically significant difference in LC with BED10 ≥50 Gy versus BED10 < 50 Gy (2-tailed, P = 0.104), nor was there a difference in reported necrosis rate (P = 0.85). On multivariate analysis of 1-year LC, the only significant factor was volume of tumor treated, where increased tumor volume was associated with increased LC (P = 0.0011).This analysis showed there was no statistically significant difference between dose-fractionation regimens delivering BED10 ≥50 Gy versus BED10 < 50 Gy with regards to 1-year LC and the incidence of RN. Studies appeared heterogeneous and this may explain the unexpected association between increasing tumor volume and LC. Future randomized controlled trials should be conducted to identify the optimal dose-fractionation for patients with large brain metastases.O.M. Siddiqui: None. C. Haskins: None. S.M. Bentzen: Travel Expenses; University of Copenhagen. M.V. Mishra: Employee; Orthofix. Research Grant; ASTRO, Keep Punching. Advisory Board; Patient Centers Outcomes Research Institute (PCORI. Travel Expenses; Patient Centers Outcomes Research Institute (PCORI.