Abstract Book

S1024

ESTRO 37

Purpose or Objective As proton pencil beams propagate through media, multiple Coulomb scattering (MCS) causes lateral spread as a function of depth, thus increasing the spot size at the depth of the Bragg peak ( R 100 ). The purpose of this work is to evaluate computed spot profiles and ellipticity at R 100 for four commercially available proton treatment planning systems (TPS), benchmarked against measured in-air data and a known model. The four TPS (Eclipse TM , XiO ® , Pinnacle 3 , RayStation ® ) were commissioned using pencil beam scanning data from the University of Pennsylvania (UPenn) facility. Material and Methods Individual pencil beam lateral profiles were calculated at depth z = R 100 in each TPS for 27 nominal energies, ranging from 100 to 226.7 MeV. Profiles at five source-to- surface distances (SSD) relative to isocentre (+20, +10, 0, -10, -17cm) were calculated in both x and y directions for all energies. A calculation grid size of 1mm was used. To benchmark these, spots sizes at depth z = R 100 were also calculated for measured in-air data ( σ 0 ) using equation 1:

EP-1891 Implementation of the optimization algorithm Photon Optimizer in VMAT for prostate cancer treatment I. Birba 1 , M. Robilliard 1 1 Institut Curie, Service de Physique Medicale, paris, France Purpose or Objective The aim of the current study is to evaluate the new optimization algorithm Photon Optimizer version 13 (Varian) (PO13). The purpose is to obtain a treatment planning protocol for prostate cancer treatments, reproducible which can be delivered by the machine and evaluate the dosimetric gain for patients (better organ at risk protection). Material and Methods In our department, the medical prescription dose is 75Gy (2.5Gy/fraction) on the prostate (PTV-T) and 46Gy (2Gy/fraction) on the right and left iliac lymph nodes (PTV-N). Treatment plannings are made of 2 plans: - 1 st plan in integrated boost delivering : 57.5Gy on the PTV-T and 46Gy on the PTV-N in 23 fractions, - 2 nd plan: 17.5Gy on the PTV-T in 7 fractions. Plans are optimized with PO13 and are calculated with AAA algorithm (Varian) on the Eclipse’s Treatment Planning System v13.6 and delivered on a Truebeam (Varian) v 2.5 MR2 in VMAT (6 MV, 600 UM/min). Several adjustable parameters with PO13 are tested on one patient: Normal Tissue Optimization (NTO) influence, calculation resolution during the optimization (1.25, 2.5, 5mm), one or more arcs, optimization volumes, PTV priority compared to OAR priority, PTV priority compared to NTO priority, use of the intermediate dose calculation. The methodology used is first to obtain the best PTV cover, changing one parameter at each optimization, and once parameters chosen and fixed, OAR constraints are introduced. These parameters will be tested with several patients (reproducibility). Plan evaluation is realised with DVH: - PTV: D95%, D98%, D2%, Dmin and Confomation Number (CN). - OAR: bladder and rectum V50Gy and femoral heads Dmax, Dmean/ The monitor units number (MU) per arc is also noted. The dosimetric benefit for the patient will be evaluated in comparison with the real treatment plan calculated with the PRO13 (Progressive Resolution Optimizer algorithm v13). Patient’s quality assurances were also performed using ionisation chamber and 2D gamma index. Results The protocol will be based on the plan presenting a sufficient coverage for the PTV with the best CN taking into account the DVH values on the OAR. The 1 st plan chosen is Plan_1 (table 1) with 2 arcs, PTV priority 200/NTO priority 500, calculation resolution 2.5mm, an intermediate dose calculation. Plan_2 with a single arc has D98% and D95% reduced by 1Gy compared to the other plans and Plan_4 has a CN too weak on the two PTV. Plan_3 with a calculation resolution of 1.25mm seems the same as Plan_1 but the optimization time is multiplied by 2, a resolution of 2.5mm is therefore preferred. The parameters are same for the 2 nd plan. With the use of this new protocol with PO13, we obtain a significant reduction of DVH values for rectum and bladder in comparison with PRO13 algorithm (table 2).

Results Differences between computed spot sizes at R 100

in all

at R 100

, based on measured

TPS and modelled values of σ z

σ 0 data were: < 0.3 mm in x-direction and < ±0.2 mm in y-direction for XiO; < 0.4 mm in x-direction and < 0.3 mm in y-direction for Eclipse; < 0.8 mm in x-direction and < 0.7 mm in y-direction for RayStation; < 1.4 mm in x- direction and < 1.1 mm in y-direction for Pinnacle. Benchmarked variation in TPS computed spot sizes at R 100 for all energies, spot ellipicity agreement, and variation as a function of SSD for 170 MeV are shown in Figure 1.

Conclusion Characteristics of the computed depth dependent spot size σ(z) at R 100 in both lateral directions, and at five different SSD, were compared to modelled values of σ z at R100, based on measured in-air data, for four commercially available TPS. All were within clinically acceptable tolerances, with XiO and Eclipse showing the closest agreement. Differences observed were attributed to TPS specific beam modelling. Further investigation will assess the cumulative impact of these discrepancies on verified clinical treatment plans.