Predicted Risk of Post-Irradiation Cerebral Necrosis in Pediatric Brain Cancer Patients: A Treatment Planning Comparison of Proton Therapy vs. Photon Therapy

Abstract

Purpose: To predict the risk of radiation necrosis in a cohort of pediatric patients with glioma and ependymoma and compare the predicted risk between volumetric modulated arc photon therapy (VMAT), passively scattered proton therapy (PSPT) and intensity modulated proton therapy (IMPT). Methods: Thirteen pediatric patients with varying age and sex were selected for this study. A radiation oncologist contoured a clinical treatment volume (CTV) on 8 patients selected for glioma in the cerebral hemisphere and 5 with ependymoma located in the posterior fossa. A 1 cm margin was added to the CTV to define the planning treatment volume (PTV). VMAT plans were constructed using Phillips Pinnacle treatment planning system. PSPT and IMPT plans were constructed using Varian Eclipse. Plans were compared using several dose metrics to ensure consistency between plan coverage. Normal tissue complication probability (NTCP) with radiation necrosis as an endpoint was calculated using the Lyman Kutcher Burman probit model. The ratio of risk was calculated between protons and photons and compared to a value of 1 using the student t-test and Wilcoxon signed rank test. Sensitivity tests were performed to determine if the predicted risk of necrosis was sensitive to positional errors, proton range errors and selection of risk models. Results: PSPT plans resulted in an average ratio of risk of 0.44 (p<0.00001) and 0.62 (p<0.02) for glioma and ependymoma patients compared to VMAT respectively. IMPT plans resulted in an average ratio of risk of 0.33 (p<0.00001) and 0.32 (p<0.00001) for glioma and ependymoma plans compared to VMAT respectively. Conclusion: Both PSPT and IMPT plans statistically significantly reduced the predicted risk of radiation necrosis using the LKB NTCP risk model. Sensitivity analysis upheld these qualitative findings.