ESTRO 38 Abstract book

S1086 ESTRO 38

Results

Results: Preliminary results (4 patients) showed that for 4 out of 5 high-dose PTVs, D 98% during treatment was at least 2% lower than planned in both margin plans. The high-dose CTVs’ D 99% generally changed less than 1% relative to the planned dose. However, for two 3 mm plans the high-dose CTV D 99% was systematically 2% lower (Fig.1). For all high- dose CTVs, the D 99% of the reconstructed dose at the end of treatment was always 96% of the prescribed dose or higher in both 3 and 5 mm margin plans. For the elective low-dose CTVs, the reconstructed D 99% was at least 99% of the prescribed dose for all fractions in both plans. On average, the D mean for the ipsilateral parotid glands was 4 Gy lower (range; 1 to 10 Gy) in 3 mm margin plans compared to the 5 mm plans. For submandibular glands, the average dose reduction was 2 Gy. The difference between reconstructed and planned D mean of the parotid glands was not significantly different for the 3 mm (range; -1.2 to 1.1 Gy) and the 5 mm (range; -1.4 to 1.2 Gy) plans (p=0.67). Conclusion Changes in delivered dose due to anatomical changes can be accurately reconstructed using DIR. Initial results indicate that 3 mm PTV margins are robust for anatomical changes occurring during HNRT with CTV D 99% ≥ 96% of the prescribed dose. PTV margin reduction resulted in increased OAR sparing. EP-1988 Statistical process control to monitor anatomical changes during head and neck radiotherapy N. Lowther 1 , D. Hamilton 1 , H. Kim 1 , J. Evans 1 , S. Marsh 2 , R. Louwe 1 1 Wellington Blood & Cancer Centre, Department of Radiation Oncology, Wellington, New Zealand ; 2 University of Canterbury, Department of Medical Physics, Christchurch, New Zealand Purpose or Objective Reduced toxicity while maintaining loco-regional control rates have been reported after reducing planning target volume (PTV) margins for head-and-neck radiotherapy (HNRT). In this context, quantifying anatomical changes to monitor the patient during treatment is preferred. This retrospective feasibility study investigated the application of deformable image registration (DIR) and Exponentially Weighted Moving Average (EWMA) Statistical Process Control (SPC) charts for this purpose. Material and Methods DIR was performed between the computed tomography for treatment planning (pCT) images of 12 patients and their daily on-treatment cone beam computed tomography (CBCT) images to quantify changes in patient pose and anatomy during treatment. EWMA charts were used to investigate trends in patient positioning reproducibility and soft tissue changes of various structures. The 90% confidence limits for both the EWMA trends and the SPC process limits were obtained using a comprehensive uncertainty analysis. These confidence limits were used to confirm whether a trend breached either an SPC limit or an a priori set clinical limit of 2 mm at a previous fraction

Trends in patient positioning reproducibility relative to the first week of treatment that were outside SPC process limits before the end of treatment week 4 occurred in 54% of cases. Only 24% of these cases could be confirmed at a 90% confidence level before the end of treatment. Using an a priori clinical limit of 2 mm, absolute changes in patient pose were detected in 39% of cases, of which 82% were confirmed. Soft tissue trends outside SPC process limits occurring before the end of treatment week 4 could be confirmed in 90% of cases. Conclusion EWMA trends based on DIR data combined with structure specific action thresholds enabled detection of systematic changes in patient pose and anatomy during the first four weeks of treatment. This approach may facilitate timely treatment adaptation and provide a safety net for PTV margin reduction. EP-1989 Mesorectal variation and PTV margins for irradiation of rectal cancer patients using belly-board M. Cox 1 , M. Wendling 1 , R. Van Leeuwen 1 , H. Rütten 1 , P. Braam 1 1 Radboud University Medical Centre, Radiation Oncology, Nijmegen, The Netherlands Purpose or Objective In our centre rectal cancer patients are currently treated with VMAT in a prone position, using a belly-board and full bladder protocol. Current PTV margins are 15 mm in anterior direction and 10 mm in other directions. These margins are based on studies in which patients were irradiated either in supine position or prone position without the use of a belly-board. In the mesorectal part of the CTV, shape variation is expected to be most substantial and heterogeneous. Therefore, the purpose of this study was to investigate the inter-fraction shape variation of the mesorectum and determine PTV margins in rectal cancer patients irradiated in prone position using a belly-board. Material and Methods For 18 patients a planning CT (pCT) and five cone-beam CT (CBCT) scans were acquired in prone position using a belly-board (Pelvic Prone Board, MacroMedics, The Netherlands). The mesorectal part of the CTV was delineated on all scans. These delineations were interpolated on the craniocaudal axis to 50 slices, and 100 equidistant dots were placed and numbered on each slice, starting at the posterior side of the patient via left, anterior, right and back to posterior. The mesorectal shape variation was quantified for each patient by measuring the 1D distance between corresponding dots on the pCT- and the five CBCT delineations. For dots on the left/right side the distance was measured in lateral direction and for the anterior/posterior side in ventrodorsal direction. For each patient the mean and standard deviation (SD) were calculated for each dot and stored in 2D surface maps. Subsequently, local group mean (signed for inside/outside), systematic- and random error maps of the total group were calculated (figure 1). In

or not. Results