ESTRO 38 Abstract book

S962 ESTRO 38

OF calc ] was considered. A Multivariate analysis of Variance was performed to test the effect of G, S, field size (FS), calculation algorithm and measurements setup on the uncertainties of OF (Dev [%]) giving the p-values of each variance component. For each factor identified as a significant predictor, a post-hoc test was carried out in order to assess the significance of intragroup differences. Results Results showed that OFs calculated by TPSs were generally larger than OFs measured. Statistical analysis shows that the uncertainties on OFs are not correlated with slice thickness alone, while they increase as the size of the calculation grid increases with a greater dispersion of the data dependent on the algorithm (Fig. 1 panel a) (p < 0.01) and as the FS decreases (Fig. 1 panel b) (p<0.001). Also algorithm and measurement setup affect results (p < 0.01) with a lower dependence on the setup for systems that have the collapse cone as an algorithm (Fig. 1 panel c). A post-hoc test has confirmed the effect on OF’s calculation of field, grid, algorithm and measurement setup (p<0.01). )/OF meas Fig 1. Box-Plot of the absolute difference of calculated and measured OFs in terms of calculation grid (panel a), field size (panel b) and measurement setup (panel c), grouped by algorithm. Conclusion Our results indicate that modern TPS overestimate the calculated small fields OFs compared with measured ones. The overestimation increases with the grid resolution and decreases with the FS. Therefore, we suggest to use, in the SBRT treatment plans, a dimension of the calculation grid as lowest as possible taking account of the hardware capability of TPS and the calculation algorithm. EP-1780 Correlation between VMAT plans complexity indices and gamma passing rate by using three QA phantoms I. Vacchieri 1 , P. Tabarelli De Fatis 2 , M. Liotta 2 1 University of Milan, post graduate Medical Physics School, Milan, Italy ; 2 IRCCS Istituti Clinici Scientifici Maugeri, Medical Physics Unit, Pavia, Italy Purpose or Objective to compare the correlation between VMAT plans complexity indices and local gamma passing rate obtained from three phantoms: ArcCHECK TM - Sun Nuclear Corp. (A), Octavius® 4D- PTW (O) and Delta4- ScandiDos (D 4 ). Material and Methods 50 VMAT clinical plans that include 15 prostate tumor bed, 15 pelvis, 10 rectums and 10 head and neck have been planned with Philips Pinnacle 3 v9.0 for Varian Clinac iX with 120 Millenium MLC. For all plans, pre-treatment verifications have been performed with the three devices and a local gamma analysis (3%-3mm) has been made with a threshold of 10%. RTPlan files were analyzed with a homemade MATLAB (MathWorks®) program to obtain plan complexity indices: Modulation Complexity Score (MCS) and Normalize Leaf Travel LTi defined by Masi et al. [1], and Monitor Units (MU). Then, the correlations between gamma passing rate of the three devices and MCS, LTi and MU have been investigated by Pearson coefficient. The correlation were considered weak for | r |<0.4, moderate for 0.4≤| r |≤0.7 and strong for | r |>0.7. Furthermore statistical significance of the difference between the means of the gamma passing rate of the three phantoms

In clinical practice, the movers were on average in 50% of fractions irradiated with the full bladder plan and in the other 50% with the empty bladder plan. Therefore, in this study we investigated correlations (R 2 ) between mean planned OAR dose parameters for the full and empty structure sets with the corresponding estimated delivered parameters derived with the CBCTs and dose accumulation. Results Obtained R 2 are presented in the Table with an overall average of 0.68. For rectum and bladder, V30 showed the strongest correlation. For the small bowel V20 correlated best, while R 2 f or V30 was still 0.74. The V40 R 2 for bladder, rectum and bowel were 0.75, 0.78 and 0.64, respectively. For bowel cavity V45, R 2 was rather low. This may be related to the involved relatively small volumes combined with large mobility.

Conclusion Especially the planned bladder, rectum and bowel cavity V30 and V40 correlated with accumulated dose. Reference: 1. A symmetric nonrigid registration method to handle large organ deformations in cervical cancer patients. Med Phys . 2010 Jul;37:3760– 3772. EP-1779 Impact of calculation grid resolution and CT slice thickness on TPS calculated small fields OF M.D. Falco 1 , M. Fusella 1 , C. Fiandra 1 , S. Clemente 1 , C. Garibaldi 1 , M. Casati 1 , F.R. Giglioli 1 , E. Gallio 1 , T. Malatesta 1 , A. Delana 1 , C. Marino 1 , A. Soriani 1 , S. Linsalata 1 , P. Bagalà 1 , G. Benecchi 1 , R. Consorti 1 , M. Casale 1 , G. Reggiori 1 , E. Villaggi 1 , S. Russo 1 , P. Mancosu 1 1 SBRT Working Group, AIFM - Italian Association of Medical Physics, Piazza della Repubblica- Milan, Italy Purpose or Objective Stereotactic Body Radiation Therapy (SBRT) is increasing thanks to modern technologies and integrated image- guided systems. The main objective of the study was to assess the influence of calculation grid size resolution and CT slice thickness on calculated small fields output factors (OFs) among a large number of centers, equipped with a wide variety of Linacs and Treatment Planning System (TPS). Material and Methods A total of 700 measurements from twenty centers were collected. On three homogeneous phantoms of different slice thickness, i.e. 1,2 and 3 millimetres, small fields OFs were calculated using data commissioned in the TPS. OFs for 1x1 cm 2 , 2x2 cm 2 , 3x3 cm 2 were calculated in the following combination of grid-slice thickness (G-S): G1-S1, G1-S2, G2-S2, G1-S3, G2-S3 and G3-S3, respectively. Four different calculation algorithms were analyzed: AcurosXB (AXB), Analytical Anisotropic Algorithm (AAA), Collapsed Cone Convolution (CCC) and VMC MonteCarlo (MC). The measurement set-ups were: SSD=90,95,100 cm with depth= 10, 5 and 10 cm, respectively. Calculated and measured OFs were compared and the Dev[%]=ABS[(OF meas -