19 Is dose escalation in intracranial pediatric ependymoma feasible with advanced radiation techniques?

Abstract

Introduction For pediatric ependymoma, a frequent pediatric brain tumor, radiotherapy is systematically delivered after surgery and the standard dose is 59, 4 Gy with still a 40% relapse rate. A majority of relapses occur within the high dose regions (Tensaouti et al., 2017) and a survival benefit was described with an additional boost with SBRT on residue (Massimino et al., 2016). MacDonald et al., found that proton beams and IMRT have similar target coverage, but normal tissue sparing was better in proton (MacDonald et al., 2008). This prompted us to perform a dose escalation approach with an in silico dosimetric comparison between Volumetric Modulated Arc Therapy (VMAT) photon therapy and Intensity Modulated Proton Therapy (IMPT). Methods The study included 101 patients (60.4 % with posterior fossa), from the national multicentric PEPPI database (Ducassou et al., 2015). The dose to PTV59.4 was 59.4/1.8 Gy and the dose to simultaneous integrated boost volume, PTV67.6 was 67.6/2.05 Gy (this dose being chosen as an equivalent to sequential SBRT boost in the European trial NCT02265770). The Gross Total tumor (GTV) was defined as the tumor bed plus residual tumor, CTV59.4 was GTV + 5 mm, PTV59.4 = CTV59.4 + 3 mm. PTV67.6 was GTV + 3 mm. Ballistic choices were made regarding the risk on brainstem. All treatment plans were optimized with the physical dose-volume objectives/constraints (for targets as well as OARs). All plans were created by the same physicist using RaySearch System planning (Raysearch laboratories, Stockholm, Sweden, v5.0) and validated by an experienced radiation oncologist. Doses to target volume and 11 OARs were extracted from the HDV and treatment plans quality indices were derived. Additionally doses to OARs obtained in silico with SIB were compared to those really delivered to the patients in initial treatment (IT) Results The mean integral dose was much lower with SIB IMPT compared with SIB VMAT (p 0.001). Planned target volume coverage (V95%) was similar. Conformity (CI) and homogeneity indices (HI) were better with protons for most cases (p 0.001). Mean and maximum doses to organs at risk were significantly lower with protons (p 0.001) and not significantly different between SIB VMAT and initial treatment. In cases of posterior fossa tumor, with OARs in close vicinity to the PTV, a significant gain with protons was observed on dose for the brainstem, on V59Gy, mean and D2% near maximum doses. Conclusions Dose escalation in pediatric ependymoma is feasible with both techniques, but there is a potential significant gain on dose to OARs, including brainstem, with protons. One perspective for this study is LET based optimization. The other one is to stratify the population that could undergo such dose increase based on residue presence, advanced imaging factors (Tensaouti et al., 2016, Tensaouti et al. 2018) and molecular biology (Pajtler et al., 2015).