ESTRO 38 Abstract book

S1092 ESTRO 38

cavity fillings were created by filling the nasal cavities onion layer wise from outside to inside. Each nasal cavity was filled separately ranging from empty to completely filled. Each artificial CT was combined with 5 random setup errors (1.57 mm σ for translation and 0.57° σ for rotation in all 3 directions), leading to between 180 and 625 clinical scenarios per patient and plan approach, depending on the size of the nasal cavities. Plans were recalculated and reoptimized for each scenario, simulating non-adapted and DAPT fractions, respectively. 100 times 30 randomly selected fractions were combined to treatment doses. Target coverage, dose to organs at risk (OARs) and integral dose were evaluated.

Figure 1: Example of planned “star” (left) and “conformal” (right) dose distribution. Results By using conformal field directions the integral dose could be reduced by 43% on average (minimum reduction patient 4: 29%, maximum patient 2: 53%, figure 2a). Also OAR doses were reduced, if they were not directly attached or included in the target volume. In the non-adapted scenario the loss of target coverage is mainly patient dependent (V95 reduction of up to 34%, figure 2b) while the field direction has no systematic influence. DAPT can restore target coverage in all cases and in some cases even reduce the dose to OARs.

Conclusion The deviations of the BD-CT dose distributions from the original dose distribution were small for the considered metrics in the three cases. It is necessary to include more patient cases to this work in progress for more significant results. However, our results indicate the applicability of a BD approach for sCT generation from MR for proton therapy. In a next step, we will identify MR sequences suited to identify respective tissue classes, transfer the gained knowledge to an MR-based workflow and compare the resulting predicted dose distribution to the calculation based on the planning CT. EP-1997 Daily adaptive proton therapy: A clear potential to reduce dose to healthy tissue and integral dose L. Nenoff 1,2 , M. Matter 1,2 , J. Hedlund Lindmar 1,2 , T. Lomax 1,2 , D.C. Weber 1,3,4 , F. Albertini 1 1 Paul Scherrer Institute, Center for Proton Therapy, Villigen PSI, Switzerland ; 2 ETH Zurich, Department of Physics, Zurich, Switzerland ; 3 University Hospital Bern, Department of Radiation Oncology, Bern, Switzerland ; 4 University Hospital Zurich, Department of Radiation Oncology, Zurich, Switzerland Purpose or Objective In some anatomical areas, for example the nasal cavities, anatomical changes can be observed on a daily base. This could lead to large dose distortions and is therefore covered by large target margins and a careful selection of field directions. However, this increases the dose to healthy tissue. With a daily 3D image in treatment position target coverage can be ensured by adapting on the daily geometry. Consequently, target margins can be reduced and alternative, less robust but more conformal field arrangements could be used. In this study we investigate the dosimetric effect of daily adaptive proton therapy (DAPT) combined with conformal field arrangements. Material and Methods Planning CTs of 5 Patients with tumors involving parts of the nasal cavities were used to generate an anatomically robust “star” plan with the clinically used PTV margin and a “conformal” plan with a field specific PTV with covering 3% uncertainty in range and 1 mm in setup (60 GyRBE, 30 fx, figure 1) each. Also synthetic CTs with different nasal

Figure 2: a) Integral dose of initial treatment plans. b) CTV V95 differences between treatment doses and initial treatment plan. Conclusion DAPT combined with an adequate margin adaption and conformal field directions reduces the integral dose and