2667 - Photon-Based Stereotactic Radiosurgery-Style Planning for the Treatment of Primary Brain Tumors to Reduce Radiation Dose to the Uninvolved Brain

Purpose/Objective(s): Conventional photon-based planning techniques for primary brain tumors often use coplanar arcs, mostly focusing on conformality of prescribed dose with the result of delivering intermediate and low doses to large volumes of healthy brain tissue. This study evaluates the feasibility and effectiveness of LINAC-based stereotactic radiosurgery (SRS) style planning to reduce dose to uninvolved brain tissue.

Materials/Methods: Twenty patients with primary brain tumors treated with 59.4-60 Gy/30-33 fractions were planned with SRS-style technique with 3 non-coplanar arcs, steep dose gradients, and an indexed Encompass mask for immobilization. Emphasis was placed on reducing intermediate and low dose spread to the brain minus PTV and to the memory circuit substructures (hippocampi, fornix, corpus callosum, amygdalae). All patients were subsequently replanned with the institution’s established planning method using 2–3 coplanar arcs, prioritizing target conformity and hippocampal sparing when possible. All plans were reviewed for clinical acceptability by physicians and physicists. Metrics from both techniques such as brain minus PTV dose, 50% and 30% isodose clouds, linear dose gradients at 1cm, and dose to midline structures were compared using the Shapiro-Wilk test for assessing normality, followed by a paired t-test or Wilcoxon signed-rank test, with significance set at P < 0.05.

Results: Compared to the established planning technique, SRS-style plans showed significant reduction in mean and median brain minus PTV doses (-12% and -18.7% respectively), as well as steeper dose gradients (1.67 Gy/mm +/- 0.16 Gy/mm vs 1.08 Gy/mm ± 0.14 Gy/mm at 1cm from PTV for SRS and established techniques respectfully, P=5.71E-5). All plans passed physics second check and showed similar conformity index (CI) and PTV coverage. There was also a statistically significant reduction of the 50% and 30% isodose volumes on the order of -19%. In addition to the brain minus PTV, there was an observed reduction in dose to the memory circuit as a whole and specifically to the fornix and corpus callosum as seen in the table.

Conclusion: SRS-style planning and beam arrangements allowed for a statistically significant reduction in dose to uninvolved brain tissue without compromising plan quality. Additionally, there was improvement in planning gradient metrics and reduction in dose to memory circuit substructures. Future studies are warranted to measure the impact of the technique on patients’ quality of life and neurocognitive function.