ESTRO 2021 Abstract Book

S1273

ESTRO 2021

fraction Internal Margin (IM) to extend PTV margin.

Conclusion Our results demonstrated that , although breath hold resulted an effective technique for liver SRT with a residual organ motion < 5 mm, liver deformability represented a residual error not online correctable to consider also in breath hold patients. Deformability was offline quantified for all patients and considered as IM for PTV definition, with an additional margin of at least 2 mm; our data will be validated by a larger sample. PO-1549 The value of online adaptive proton therapy based on robust dose restoration for head and neck E. Borderias-Villarroel 1 , V. Taasti 2 , W. Van Elmpt 2 , X. Geets 3 , E. Sterpin 4 1 MIRO, UCLouvain, Brussels, Belgium; 2 MAASTRO Clinic, Radiation Oncology, Maastricht, The Netherlands; 3 MIRO - UCLouvain, Radiotherapy Cliniques Saint Luc , Brussels, Belgium; 4 MIRO , UCLouvain, Brussels, Belgium Purpose or Objective Intensity modulated proton therapy (IMPT) is highly sensitive to anatomical variations which can cause inadequate target coverage during treatment. Different techniques, such as robust treatment planning, online or offline adaptive IMPT, can mitigate the distortion of the initial plans due to anatomical deformations throughout the treatment fractions . This study compares a robust planning strategy (RO) to two adaptive IMPT strategies - a fully-offline adaptive (FOA) and an online adaptive strategy (dose restoration (DR)) to determine the benefit of dose restoration (DR) in head and neck (HN) patients. Materials and Methods Robust optimization (setup error = 3mm, range error = 3%) was performed for 5 patients in a head and neck (HN) cancer cohort including 24 repeated-CT (rCTs). Dose was prescribed as 70 Gy(RBE) to the primary tumour (CTV2) and as 54.25 Gy(RBE) to the nodal target (CTV1). For each rCT, we computed the evaluation of the initial robust plan (not-adapted (NA) plan), the online dose restoration (DR) strategy and a full off-line adaptation (FOA) method. DR used isodose contours generated from the initial dose and patient specific minimum and maximum objectives to re-optimize the plan and reproduce the initial dose in every rCT. Robust evaluations (1mm, 3%) of not-adapted, restored and FOA plans were compared for each rCT. Results Differences to clinical plan in DVH-metrics (nominal and worst-case values) were analysed for not-adapted (NA), restored and FOA plans (Figure 1). An improvement in DVH-metrics and robustness was seen for restored and FOA plans compared to not-adapted plans. For not-adapted plans, 29%(7/24) of rCTs did not fulfill the CTV coverage criteria (D98%> 95%Dprescription). By implementing DR, the median of the D98% over the 24 rCTs was increased by 0.18 Gy and 0.30 Gy in CTV2 and CTV1, respectively. Dose restoration reduced the variability on target coverage and improved the homogeneity. Variability on critical risk organs such as the spinal cord was also reduced in restored plans. Because of large anatomical changes and/or inaccurate patient repositioning, 16%(4/24) of rCTs still needed full offline adaptation (FOA) to ensure an optimal treatment since dose restoration was not able to re-establish the initial plan quality on the rCTs.