Frameless Cobalt60-Based Hypofractionated Stereotactic Radiosurgery (HSRS) for Brain Metastases: Impact of Dose and Volume.

Frameless, gated, image-guided, Cobalt60-based hypofractionated stereotactic radiosurgery (Co60-HSRS) is a novel technical paradigm in the treatment of brain metastases that allows for both the dosimetric benefits of the Co60 based stereotactic radiosurgery (SRS) platform as well as the biologic benefits of fractionation. We report mature local control (LC) and adverse radiation effects (ARE) outcomes following 5 fraction Co60-HSRS for intact brain metastases. All patients with intact brain metastases treated with 5-fraction Co60-HSRS between 2017-2020 were retrospectively reviewed. Patients were typically selected for HSRS as opposed to single fraction SRS if the metastases were larger (>2cm diameter), were in eloquent areas, or were in proximity to another lesion receiving HSRS. Survival estimates were determined per patient using Kaplan Meier methods, and LC as well as symptomatic ARE rates were determined per lesion using competing risk methods. Univariable competing risk regression using Fine and Gray's methods were performed, and subsequent multivariable (MVA) regression using a backwards step-wise selection technique generated the final adjusted models. In total 299 metastases in 146 patients were identified. The median clinical and radiologic follow-up was 10.6 (range 0.5-49.9) and 10.7 months (range 0.5-47.6), respectively. The median maximum tumor diameter and volume were 1.7 cm (range, 0.2-3.9 cm) and 2.38 cc (range, 0.004-24.92 cc), respectively. The median total dose and prescription isodose was 27.5 Gy (range, 20-27.5 Gy) in 5 daily fractions and 52% (range, 45%-93%), respectively. The median overall survival (OS) was 12.7 months and the 1-year LC rate was 85%. MVA identified a total dose of 27.5 Gy vs. ≤25 Gy (hazard ratio [HR] 0.59, p = 0.042), and prior chemotherapy exposure (HR 1.99, p = 0.015), as significant predictors of LC. The 1-year ARE rate was 10.8% and the symptomatic ARE rate was 1.8%. MVA identified a gross tumor volume of ≥4.5cc (HR 7.29, p < 0.001) and a mean intra-tumoral dose of ≥39 Gy (HR 3.17, p = 0.034) as significant predictors of symptomatic ARE. Moderate total doses in 5 daily fractions of Co60-HSRS were associated with high rates of LC and a low incidence of symptomatic ARE. A prescription dose of 27.5 Gy was superior with respect to local control versus ≤25 Gy in 5 daily fractions. Target volumes of 4.5cc or larger as well as mean dose >39 Gy, were associated with higher rates of symptomatic necrosis. Further study will help refine the optimal dosimetric constraints for Co60-HSRS. [ABSTRACT FROM AUTHOR]