Exploring feasibility criteria for stereotactic radiosurgical treatment of multiple brain metastases using five linac machines.

Purpose: This study aimed to find descriptors that correlates with normal brain dose to determine the feasibility of performing fractionated stereotactic radiosurgery (SRS) for multiple brain metastases (BMs) using five linac machines. Methods: Thirty‐two patients with 1–30 BMs were enrolled. Treatment plans were created using TrueBeam, Novalis Tx, TrueBeam Edge, Halcyon, and Tomotherapy linacs. The sum of all planning target volumes (PTVs) was defined as PTVall, and the brain region excluding PTVall was defined as normal brain. The total surface area (TSA) of the PTV was calculated from the sum of the surface areas of the equivalent spheres for each PTV. Volumes receiving more than 5, 12, and 18 Gy (V5Gy, V12Gy, and V18Gy, respectively) were used for evaluation of normal brain dose. Correlations between normal brain dose and each tumor characteristic (number, PTVall, and TSA) were investigated using the Spearman rank correlation coefficient. Results: Correlations between each characteristic and normal brain dose were statistically significant (p < 0.05) across all machines. The correlation coefficients between each characteristic and V18Gy for the five machines were as follows: tumor number, 0.39–0.60; PTVall, 0.79–0.93; TSA, 0.93–0.99. The fit equations between TSA and V18Gy exhibited high coefficients of determination, ranging from 0.92 to 0.99 across five machines. Conclusion: This study devised fractionated SRS plans using for 1–30 BMs across five linac machines to find descriptors for determining SRS feasibility based on normal brain dose. TSA proved to be a promising descriptor of SRS feasibility for treating multiple BMs. [ABSTRACT FROM AUTHOR]