ESTRO 37 Abstract book

S46

ESTRO 37

optimization towards a Pareto optimal solution for the new anatomy. The method was evaluated using the planning CT and 7-9 repeat CTs of 11 prostate cancer patients. Prostate, lymph nodes and seminal vesicles were delineated as target structures. Dose was prescribed according to a simultaneously integrated boost scheme assigning 74 Gy to the high-dose PTV, (prostate) and 55 Gy to the low- dose PTV (lymph nodes and seminal vesicles). Re- optimization was performed for each repeat CT using tight margins of 2/3.5 mm (high-dose/low-dose PTV), only meant to account for intra-fraction motion. The original plan was created with enlarged margins (4/7 mm) to provide a spot placement covering most target deformations in the repeat CTs. The original and re- optimized spot positions and weights were evaluated on the repeat CTs using the 2/3.5 mm-PTVs. Results Evaluated on the repeat CTs, the original plans achieved sufficient target coverage (PTV V 95% ≥ 98% and V 107% ≤ 2%) in only 1 out of 88 CTs (see Fig 1). With re-optimization this improved to 86/88 CTs. Fig 2 shows the differences between the original and re-optimized plans for the OARs. Median improvements up to 12%-point for the rectum and bladder V 45Gy were observed, with outliers up to 35%-point. For most CTs, all dosimetric parameter values were lower for the re-optimized plans. Four CTs had a lower bladder D mean for the original plan. All re- optimizations were completed within 3.5 minutes (average 3 min).

Conclusion Geometric uncertainties induced by motions estimated from 4D imaging using DIRs can introduce pronounced systematic and random errors in 4DDCs, which are significant for the clinical evaluation of liver 4D PBS-PT plans. Therefore, for an accurate and safe treatment, we recommend to interpret individual 4D dose distributions with caution. OC-0090 A Planning Strategy for Near Real-Time Adaptive Proton Therapy T. Jagt 1 , S. Breedveld 1 , B. Heijmen 1 , M. Hoogeman 1 1 Erasmus Medical Center Rotterdam Daniel den Hoed Cancer Center, Radiation Oncology, Rotterdam, The Netherlands Purpose or Objective Intensity-modulated proton therapy allows for highly localized dose deposition, but is also very sensitive to daily variations in tissue density along the pencil beam paths and variations in target, OAR and patient shape. This study develops and evaluates a method to automatically adapt the treatment plan in near real-time to the anatomy of the day. This allows for tight dose distributions around the target. Our method simultaneously aims for target dose reconstruction, as well as minimizing the dose to the OARs as far as possible, while maintaining the multi-criteria trade-offs of the treatment plan, generated for the planning CT (original plan). Material and Methods The re-optimization method consists of two steps. 1) Restoration of the spot positions (Bragg peaks) from the original plan by adapting the energy of each pencil beam to the water equivalent path length (WEPL) in the repeat CT. To compensate for deformed target shape, new energy layers and spots are added. 2) Using the reference point method (RPM) for optimizing the spot weights, while respecting the trade-offs made in the original plan. The RPM uses the achieved clinical objective values of the original plan as reference point to guide the re-