ESTRO 38 Abstract book

S1073 ESTRO 38

Material and Methods Eligible patients, who had the intellectual and physical capacity to perform reproducible DIBH during training, underwent DIBH CT simulation and treatment delivery using the Varian surface-based real-time position management (RPM) tracking system. Treatment planning was carried out on the Varian Eclipse treatment planning system v.11. Patients received a prescription dose of 45Gy in 20 fractions. A second series of CT images was taken during FB for each patient, for which a second treatment plan was generated. Accordingly, a comparison between the DIBH and FB sets of plans, in terms of certain lung and heart dosimetric and geometric parameters, was conducted. Paired t-test was used to compare means for both data sets. Results Thirty consecutive breast cancer patients were treated with DIBH radiation therapy technique. Twenty patients (67%) had left breast cancer, nine patients (30%) were right-sided and one patient (3%) had bilateral disease. Dosimetric comparison between DIBH and FB plans revealed a lower heart V18 with DIBH at 4.7±4.7% vs. 12.1±11.1% in FB (p=0.001), a lower heart V28 with DIBH at 3±3.3% vs. 10.3±10% in FB (p=0.0002) and a lower mean cardiac dose of 4.1±3Gy with DIBH vs. 6.7±4.9Gy in FB (p=0.009). There was also a reduction in the left anterior descending coronary artery (LAD) V18 with DIBH at 43.5±42.1% vs. 68±43.2% in FB (p<0.0001), a lower mean LAD dose of 18.2±15Gy with DIBH vs. 28.3±18Gy in FB (p=0.006) and a lower maximum LAD dose of 27.5±17.8Gy with DIBH vs. 32.9±17.9Gy in FB (p=0.007). In addition, DIBH yielded a lower ipsilateral lung V18 at 20.5±4.8% vs. 25.7±8.1% in FB (p<0.0001) and a lower ipsilateral lung mean dose of 11.1±3.7Gy vs 12.3±3.2Gy in FB (p=0.026). With regard to the geometric parameters, DIBH reduced the cardiac contact distance with the chest wall, both in the axial (0.4±0.8cm vs. 3.9±2.2cm in FB; p<0.0001) and sagittal planes (2.5±1.9cm vs. 4.2±2.1cm in FB; p<0.0001), see figure 1. Moreover, DIBH significantly decreased the maximum heart distance inside the treatment field compared with FB (1.6±0.6cm vs. 2.7±0.5cm, respectively; p<0.0001).

1 Hospital Universitario HM Sanchinarro, Radiofísica, Madrid, Spain ; 2 Hospital Universitario HM Sanchinarro, Oncología Radioterápica, Madrid, Spain Purpose or Objective Quantify the experimental relationship between the uncertainty due to respiratory motion (RM) and the peak to peak amplitude (A) of this movement in liver, lung, pancreas and breast tumors treated with stereotactic body radiation therapy (SBRT) and ExacTrac Adaptive Gating® with intra-fractional IGRT. Material and Methods Fifty five fractions (15000 cycles) of liver, 41 fractions (14000 cycles) of lung, 76 fractions (34000 cycles) of pancreas and 70 fractions (14000 cycles) of breast tumors were treated with SBRT using ExacTrac Adaptive Gating® in a Novalis® linac. This system uses external markers to monitor the respiratory cycle and internal fiducial markers to set up the patient and measure the movement of the tumor. All the magnitudes of the respiratory motion of tumor were analyzed through a linear regression between the motion of external markers and internal fiducial markers (surrogate to tumor motion). The uncertainty due to respiratory motion (SD(RM)) is estimated measuring directly the standard deviation (SD) of the respiratory motion (RM). Results Table 1 shows the amplitude (A), the standard deviation of the amplitude [SD(A)] and the uncertainty due to respiratory motion {SD(RM)}. The SD(RM) is usually estimated as A/3, this is valid for an ideal respiratory cycle represented by RM= A*sen 2n (wt), in which the amplitude is not variable. But it may not be true if SD(A) is substantial. In this study SD(RM) is significantly larger than A/3 (p<0.001) in all directions of all localizations due to the intrafraction variability of A. The univariant analysis revealed a statistical significant linear correlation between SD(RM) and A and between SD(RM) and SD(A) both with p<0.001 in in all directions of all localizations. On the other hand, the multivariant regression SD 2 (MR) = a*A 2 +b*SD 2 (A) showed statistical significant correlation with SD 2 (A) and A 2 (p<0.05) in all cases, except in lung tumors in which the A 2 correlation was not significant. In addition, for liver and pancreas tumors the SD 2 (A) contribution to the respiratory motion variance is the greatest but not in a significant way. Conclusion Whereas for ideal respiratory cycles the uncertainty in tumor position due to respiratory motion can be estimated accurately with A/3, in real respiratory cycles is necessary to include the SD(A) beside A to avoid underestimation of this uncertainty. Further studies are necessary to demonstrate if those findings have any repercussion on PTV margin estimation. EP-1968 Respiratory-gated carbon-ion beam treatments of abdominal targets: clinical introduction of 4DMRI A. Vai 1 , G. Meschini 2 , S. Molinelli 1 , C. Paganelli 2 , D. Maestri 1 , G. Magro 1 , E. Mastella 1 , A. Mairani 1,3 , A. Mirandola 1 , S. Russo 1 , L. Preda 4 , G. Viselener 4 , A. Barcellini 5 , V. Vitolo 5 , A. Mancin 5 , G. Fontana 6 , G.

Conclusion Compared to FB, DIBH technique results in marked reduction of the doses to the heart, LAD and lungs. In addition, it decreases the cardiac volume in contact with the chest wall as well as the maximum heart distance inside the treatment field. Thus, DIBH appears to be a promising technique to mitigate long-term cardiac and pulmonary toxicities in breast cancer survivors. EP-1967 Relationship of uncertainty due to respiratory motion with amplitude in SBRT treatments J. García Ruiz-Zorrilla 1 , M.A. De la Casa de Julián 1 , P. García de Acilu Laa 1 , O. Hernando Requejo 2 , C. Rubio Rodriguez 2 , J. Martí Asenjo 1 , P. Fernández Letón 1 , D. Zucca Aparicio 1 , J.M. Pérez Moreno 1