ESTRO 2020 Abstract book

S104 ESTRO 2020

study we used an in-house system for fully automated, integrated optimization of IMRT profiles and beam geometries to investigate the dosimetrical impact of beam arrangement in the treatment of rectal cancer. Material and Methods The TPS for automated planning was configured for rectal MRL treatments with reduction of the D mean in a composite OAR (small-bowl-bag + bladder excluding PTV overlap) as main OAR objective, after PTV coverage. For 23 rectal cancer patients, treated with 25 fractions of 2 Gy, three types of plans were automatically generated: 1) BAO L : plans with L=7-12 individualized beam angles, generated with computerized selection from 56 coplanar, equiangular candidate beams, 2) EQUI M : plans with M=9- 56 fixed (non-patient specific)equiangular beam angles, 3) CLASS N : plans with fixed (non-patient specific) beam angle class solutions with N=7-12 beams. The class solutions were established by analyzing the BAO L beam angles of 10 training patients. For all plans, beams passing through the MRL cryostat pipe and the high attenuation parts of the couch were avoided. Prior to dosimetric comparisons, all plans were scaled for equal PTV coverage (PTV V 95% = 99%). The two sided Wilcoxon signed rank test was used for statistical significance testing (p<0.05). Results All 437 automatically generated MRL plans fulfilled all clinical requirements. Clinically relevant differences between plans were observed for the D mean of the composite OAR, as shown in Figure 1. Detailed comparisons between plans regarding OAR D mean are presented in Table 1. For all planning approaches (BAO, EQUI and CLASS), large improvements in plan quality were observed for increasing beam numbers; an OAR D mean reduction off 5.6 Gy (31%) between BAO 7 and BAO 12 . Figure 1 demonstrates clear advantages of BAO and CLASS over EQUI, e.g. an average reduction of 2.6 Gy (15%) ([1.0, 8.5], p<0.01) in OAR D mean was observed for the patient-specific 9-beam plans (BAO 9 ) compared to the non-patient-specific equiangular 9 beam plans (EQUI 9 ). Less well-selected beams lead to similar plan quality, e.g. BAO 12 and EQUI 24 had equal OAR D mean (similar for BAO 10 vs EQUI 15 , BAO 11 vs EQUI 19 ). Figure 1 shows equivalence of BAO N and CLASS N plans.

Conclusion Fully automated, integrated optimization of IMRT profiles and beam angles was used to robustly investigate the impact of beam directions and beam numbers on plan quality in rectal treatments with a Unity MRL. With optimized beam angles, plan quality is substantially enhanced compared to equiangular treatment with equal number of beams. Beam angle class solutions were developed that could replace individualized beam angle selection without plan quality loss, avoiding time- consuming individualized beam angle optimization . PD-0190 Update on yesterday’s dose – Use of delivery log files to refine the daily proton delivery M. Matter 1,2 , L. Nenoff 1,2 , L. Marc 3 , A.J. Lomax 1,2 , D.C. Weber 1,4,5 , F. Albertitni 1 1 Paul Scherrer Institute, Center for Proton Therapy, Villigen PSI, Switzerland ; 2 ETH Zürich, Departement of Physics, Zürich, Switzerland ; 3 EPF Lausanne, Departement of Physics, Lausanne, Switzerland ; 4 University Hospital Bern, Department of Radiation Oncology, Bern, Switzerland ; 5 University Hospital Zürich, Department of Radiation Oncology, Zürich, Switzerland Purpose or Objective Daily adaptive proton therapy (DAPT) aims to correct, on a daily basis, anatomical changes and positioning uncertainties. For this, the treatment should be re- optimized on-line as quickly as possible to help minimize the time between daily imaging and treatment delivery, a process that can be achieved in just a few seconds if a simple, but limited accuracy, dose algorithm is used (1). Off-line (i.e. between fractions) however, a slower and more accurate dose calculation, utilizing machine log- files, can be used to accurately reconstruct the delivered dose. This work investigates whether such off-line reconstructed doses, accumulated from previous fractions, can be fed-back into the daily re-optimization process to correct for systematic dose inaccuracies caused by fast on-line optimization, as well as inaccuracies resulting from machine delivery uncertainties. Material and Methods Five fraction treatments for three nasopharynx patients have been simulated, and log-file/Monte Carlo (MC) reconstructed doses of simulated previous deliveries were fed back into a fast on-line plan optimization process, performed using a GPU based ray-casting algorithm (1). Treatment delivery was simulated by adding realistic uncertainties to the planned spot positions and dose was reconstructed using MC (Topas-Geant4). The daily optimization target dose of the N th fraction was then optimized as follows: