Abstract Book

S677

ESTRO 37

and presented in the form of parametric maps. GTV-PET and GTV ADClow were defined for each patient by two independent investigators. Results In 30 of 41 patients, DWI-MRI showed areas with restricted diffusion (mean ADC value of GTVs ADClow = 0.74 ± 0.22 mm 2 /s, range 0.25-1.44 mm 2 /s). The volumetric comparison to the GTVs-PET revealed that the majority of the GTVs ADClow is located outside the areas of increased amino acid metabolism (non-overlapping volume = 67%). The evaluation of dose-volume histograms showed that the retrospectively defined GTVs ADClow were not completely included in the irradiated high dose volume. A mean dose of 82% of the reference dose was located within the restricted diffusion areas (i.e. V95% received 72% and V80% received 78% of the reference dose). An adjusted volume taking into account both PET- positive and diffusion restricted areas would imply an increase of 26% of the GTV. Conclusion The relevance of the aforementioned target volume differences between amino acid PET and DWI-MRI is yet to be fully understood. One can, however, suspect that planning a re-irradiation according to contrast enhanced MRI and amino acid PET alone may not cover the entire biologically active tumor. The clinical consequence of significant underdosage in areas of increased cellularity in DWI-MRI has still to be explored in prospective studies, such as the ongoing multicentric GLIAA trial. EP-1219 Dosimetric Comparison of Proton RT with Standard of Care Photon RT Techniques in CNS Tumors S. Adeberg 1 , S. Harrabi 1 , N. Bougatf 1 , V. Verma 2 , D. Bernhardt 1 , S. Combs 3 , T. Haberer 1 , K. Herfarth 1 , J. Debus 1 , S. Rieken 1 1 University Hospital Heidelberg, Department of Radiation Oncology, Heidelberg, Germany 2 University of Nebraska Medical Center, Department of Radiation Oncology, Omaha, USA 3 Technische Universität München, Department of Radiation Oncology, München, Germany Purpose or Objective In intracranial neoplasms dosimetric advantages of proton radiotherapy (PRT) over photon modalities are likely related to tumor location. This study helps to identify patients with cerebral tumors who benefit most from PRT. Material and Methods Patients were stratified into four groups (n≥10 each) according to the tumor localization (posterior cranial fossa, frontal, temporal and parietal). Each patient was re-planned using original dose constraints for organs-at- risk for three-dimensional conformal radiotherapy (3DCRT) and volumetric-modulated arc therapy (VMAT). Dmean, Dmax and integral dose (ID) were evaluated within and between each location group. *Statistical significance (p<0.05). Results The target volume coverage was excellent for all three modalities. PRT showed essential potential to spare organs at risk in all localizations. Especially parietal tumors showed a Dmean reduction using PRT, compared to VMAT and 3DCRT in the infratentorial- (-38%*; -98%*), supratentorial- (-87%*; -54%*), brain stem (-87%*; -90%*) and whole brain (-40%*; -55%*). In tumors of the posterior cranial fossa, reduction of the Dmean to the optic nerves (-99%*, -100%*), optic chiasm (-84%*; -82%*) and pituitary gland (-95%*; -86%*) was achieved best using PRT. Similar results could be achieved for the Dmax and ID in all groups. Conclusion In particular patients with parietal tumors profited from a PRT by sparing non-target brain tissue effectively compared to photon techniques. These results help to

improve the patient selection for PRT, toxicity analyses and cost-effectiveness. EP-1220 Use of VMAT in Linac-SRS Vestibular Schwannomas. Dosimetric report of 34 cases in our Institution. M.J. Perez-Calatayud 1 , B. Ibanez-Rosello 1 , D. Farga Albiol 1 , M. Rodriguez Pla 1 , E. Martin Garcia 1 , V. Carmona 1 , J. Gimeno-Olmos 1 , F. Celada Álvarez 1 , M.Y. Bernisz Diaz 1 , J. Perez-Calatayud 1 , A. Tormo Micó 1 , A. Menéndez-López 2 1 La Fe University and Polytechnic Hospital, Radiotherapy, Valencia, Spain 2 La Fe University and Polytechnic Hospital, Neurosurgery, Valencia, Spain Purpose or Objective Vestibular Schwannoma represents a very specific scenario for SRS considering its close proximity to OARs (brainstem, trigeminal nerve and cochlea). High modulation techniques as VMAT could be more beneficial than the dynamic conformal arc technique in order to increase conformality and OAR sparing. The aim of this study is to report the experience with the first 34 cases of VMAT in our institution. A specific template has been created to collect the dosimetric characteristics to be correlated, in a second phase, with the clinical follow-up (local control and complications). Material and Methods 34 patients have been treated with VMAT (15 men, 19 women, with age range 27-78 y) from Apr 2014- Sept 2017. Patients were immobilized using a headframe (Brainlab), MR-CT registration and PTV-OAR delineation were done with iPlan TPS (Brainlab). VMAT treatments planning were done with Eclipse v13.0 (Varian). Typically 4-6 non-coplanar arcs were used. Patients set-up were performed with CBCT and 6D couch on the Truebeam 2.0 with HDMLC 2.5 mm width (Varian). Dose prescription was selected in relation to the tumor volume in the range 11-12.5 Gy at D95-D98. PTV volume ranges 0.11-8.92 cc. The collected set of dosimetric parameters includes for OAR: D1% (assumed as Dmax) and D10%; for PTV: D2, D10, D90, D95, D98 and Dmean; for Normal tissue: V12 and V10. Indexes: Conformity Index (CI), Homogeneity Index (HI), Paddick Conformity Index (PCI) and Gradient Index (GI). The intended objectives of the plan are: a) to administer the prescription dose at least to the 95% of the PTV, b) to keep the brainstem up maximum dose of 12 Gy (D1%), c) D10 of the trigeminal nerve lower than 8 Gy and d) cochlea with Dmax (D1%) lower than 4.5 Gy if hearing preservation was mandatory. In cases of a non-useful hearing it was secondary to the PTV conformity. In some few cases of higher PTV, an internally defined overdose volume was used at the VMAT optimization, with the purpose of facilitate tumor shrinkage. Results For the 34 patients the dosimetric parameters summary in mean and range (standard deviation) for PTV, OAR, Normal Tissue and Indexes is as follow; Brainstem: D1% (6.3±3.2) Gy, D10% (2.6±1.7) Gy. Trigeminal nerve: D1% (6.8±3.3) Gy, D10% (5.7±3) Gy. Cochlea: D1% (8.1±2.8) Gy, D10% (7.1±2.7) Gy. PTV: D2% (13.4±0.5) Gy, D10% (13.2±0.4) Gy, D90% (12.1±0.4) Gy, D95% (11.9±0.4) Gy, D98% (11.7±0.7) Gy and Dmean (12.7±0.35) Gy. Normal tissue: V10Gy (1.53±1.16) cc, V12Gy (0.16±0.11) cc. Indexes: CI (1.28±0.31), HI (1.12±0.05), PCI (0.77±0.13), GI (5.74±1.67). Conclusion VMAT application in SRS for Vestibular Schwannomas is a feasible technique improving PTV conformation and OAR sparing during the planning procedure. In a second phase of this study, the clinical aspects will be analyzed at a larger follow up (control of tumor volume, adverse radiation effects, neuropathies). The preliminary results