ESTRO 36 Abstract Book

S147 ESTRO 36 2017 _______________________________________________________________________________________________

the transponder motion and treatment delivery were used to calculate the motion-induced geometrical errors during beam-on in the actual gated treatments and in simulated non-gated standard treatments with CBCT-guided setup to the mean transponder centroid position before each fraction. The observed motion was used to reconstruct the actually delivered CTV dose distribution with gating and the would-be dose distribution without gating. Results Fig. 1 shows the internal tumor motion during a single fraction. Due to drift and respiratory motion the mean (+/- SD) geometric error during non-gated treatment at this fraction (Fig1A) would have been 1.3mm (1.7) LR, 5.0mm (7.7) CC, and -2.0mm (1.8) AP. The gated treatment, including 5 couch shifts to counteract drift (Fig1B), reduced the errors to 0.7mm (0.7) LR, 0.4mm (1.9) CC, and -0.1mm (0.9) AP. Fig. 1C shows the CC geometrical errors for all patients. The mean (range) number of couch corrections for drifts during each gated fraction was 2.8 (0-7). The mean duty cycle during gated treatment was 60.8% (31.7-72.7%). As shown in Fig 2A, gating markedly reduced the population based PTV margin needed for intrafraction motion. Motion-including dose- reconstruction provided the CTV-DVHs of all fractions of planned, actual gated delivered, and simulated non-gated delivered doses. Mean CTV-DVHs are shown in Fig 2B. Note the large DVH variation for non-gated treatments. The mean (range) reduction in CTV D 95 relative to the planned dose was 0.9 percent points (0.1-2.3) with gating and 6.8 percent points (0.9-29.6) without gating. Conclusion Gating based on internal motion monitoring markedly reduced geometric and dosimetric errors in liver SBRT compared to non-gated standard treatment. Results of the full trial (15 patients) are expected for presentation at ESTRO.

PV-0284 3D Performance Analysis of Cyberknife Synchrony® Respiratory Tracking System M.C. Sahin 1 , P. Hurmuz 1 , M. Yeginer 1 , G. Yazici 1 , G. Ozyigit 1 1 Hacettepe University Faculty of Medicine, Radiation Oncology, Ankara, Turkey Purpose or Objective Tumor movement is a challenging issue for the precise delivery of radiation for thoracic tumors. The Synchrony respiratory motion tracking system (RMTS) of Cyberknife® robotic radiosurgery unit synchronizes radiation beam delivery with the respiration induced tumor motion. This study aims to investigate the performance of Synchrony RMTS for different movement widths using polymer gel dosimetry. To the best of our knowledge this is the first study to make the three dimensional performance analysis of Synchrony RMTS. Material and Methods The MultiPlan® treatment planning system (TPS) of Cyberknife® was used to deliver 4 Gy to a tumor of 1X1X1 cm 3 . BrainLab Gating lung phantom was used to simulate lung movements with three different amplitudes (1 cm, 2 cm and 3 cm). Three fiducials were inserted to the phantom for tracking. Radiochromic film and polymer gel dosimetry were used and measurements were compared with the dose distributions acquired from the TPS. The dose information of irradiated gel were read out using 1.5 T magnetic resonance imaging. The gamma index values were analysed using the Ashland FilmQA Pro 3.0 software for film dosimeters and Polygevero software for gel dosimeters using the 3mm/3% criteria. PolyGevero gamma index value of ≤1 is accepted as a passing criteria according to the literature. Results The mean 3 mm/3% gamma index values of film dosimetry were 92.6±1.94%, 91.0±4.00%, 90.3±2.04% for tumor motions of 1 cm, 2 cm and 3 cm, respectively (p<0.001). For polymer gel dosimetry, the mean gamma index values calculated over almost three million points were 0.56±0.10, 0.60±0.24 and 0.65±0.30 for tumor motions of 1 cm, 2 cm and 3 cm, respectively (p<0.001). Although the difference was statistically significant for 3 different amplitudes, the performance of the system was within the acceptance limits