ESTRO 37 Abstract book

S1055

ESTRO 37

the OARs, moreover the average doses of each target and OAR for 97 patients were evaluated. The dose for Oars and for targets obtained for AU and MA plans were calculated and compared. Paired t test was used for statistical analysis of AU and MA plan. Results There is no significant difference in the coverage rate of target between AU and MA Treatment plans. But there are significant differences between AU and MA plan for most of OARs in 97 patients: the average mean dose of OARs generated with AU plan is 270cGy to 1870cGy lower than those generated with MA plan. For OARs for which no significant difference between AU and MA plans is highlighted, the mean dose of OARs in AU plans were at least not higher than MA plans. Conclusion Nasopharyngeal carcinoma IMRT plans made by automatic planning tool meet the clinical requirements for target prescription dose, moreover, the dose of normal tissues was lower, and therefore the quality of radiotherapy planning was significantly enhanced. AU treatment plans were generated with higher quality without any human intervention; the time of dosimetrists can be saved and the influence of factors as the lack of experience on the treatment planning can be avoided. In conclusion, for radiotherapy of nasopharyngeal carcinoma AU plans might accurately replace MA plans. EP-1939 The Impact of Tumor Volume on Dose Fall-Off of Cyberknife With MLC vs. Fixed Collimator in Brain SRS K.S. Park 1 , J.H. Kim 1 , J.Y. Woo 1 , J.G. Baek 1 , M.J. Kim 1 , S.Y. Kim 1 , R.H. Park 1 , T.S. Jeong 1 , D.M. Jung 1 , J.H. Cho 1 1 Yonsei Cancer Center- Yonsei University, Radiation Oncology, Seoul, Korea Republic of Purpose or Objective The sharp dose fall-off is a feature of brain stereotactic radiosurgery (SRS) that deliver high radiation dose to subject with minimal dose to brain normal tissue. The aim of this study is to compare treatment plans using Cyberknife with InCiseTM multi-leaf collimator (CK_M) with treatment plans using Cyberknife with fixed collimator (CK_F) for brain SRS in various tumor volumes. Material and Methods Fifteen patients with brain SRS were planned us ing CK_M and CK_F. Prescription was 24 Gy in 1 fraction covering at least 95% of the target volume. The dose was prescribed at the 70% isodose level. Homogeneity index (HI), Conformity index (CI) and dose gradient index (DGI) were used as dosimetric indices. DGI is the distance from the average radius enclosed by the original prescription isodose to the average radius enclosed by the isodose at 50% of the prescription dose. Results The mean tumor volume was 1.41 cm3 (range, 0.27- 6.19). Overall, comparing CK_M and CK_F, the respective parametric values (mean ± standard deviation) were : same HI (1.44 ± 0.02 vs. 1.44 ± 0.01); similar CI (1.11 ± 0.06 vs. 1.20 ± 0.09, p=0.005); non statistically significant DGI (2.99 ± 0.55 vs. 3.12± 0.88 mm, p=0.206). Comparing CK_M and CK_F by tumor volume, CI (1.15 ± 0.05 vs. 1.26 ± 0.11, p=0.063), DGI (2.60 ± 0.16 vs. 2.37 ± 0.14 mm, p=0.014) for tumor volume less than 0.5 cm3(N=7, mean 0.38 ± 0.07); CI (1.07 ± 0.04 vs. 1.15 ± 0.03, p=0.028), DGI (3.31 ± 0.57 vs. 3.78 ± 0.69 mm, p=0.011) at more than 0.5 cm3(N=8, mean 2.31 ± 1.77).

Conclusion If the tumor volume is more than 0.5 cm3, Cyberknife with InCise multi-leaf collimator can achieve plan quality with sharper peripheral dose fall-off as well as better conformity than Cyberknife with fixed collimator for brain stereotactic radiosurgery. EP-1940 Hippocampal avoidance whole brain radiotherapy: a comparison of contouring and planning techniques G. Kothari 1 , P. Rich 2 , K. Amin 3 , G. Ross 4 1 Royal Marsden Hospital Trust & Institute of Cancer Research, Breast unit, London, United Kingdom 2 The Royal Marsden NHS Foundation Trust, Department of Diagnostic Radiology, London, United Kingdom 3 The Royal Marsden NHS Foundation Trust, Joint Department of Physics, London, United Kingdom 4 The Royal Marsden NHS Foundation Trust, Breast unit, London, United Kingdom Purpose or Objective Hippocampal avoidance whole brain radiotherapy (HAWBRT) has been shown to reduce the adverse neurocognitive effects of WBRT. This is relevant in patients with breast cancer (BC) for whom improvement in systemic therapy has seen a rise in the incidence of brain metastases (BM), as well as survival, such that late neurocognitive effects of WBRT are becoming increasingly relevant. This planning study compares manual with auto- segmentation contouring of the hippocampus, and intensity modulated radiotherapy (IMRT) with volumetric modulated arc therapy (VMAT) in the delivery of HAWBRT in patients with BC and BM. Material and Methods 10 patients with BC and BM who had available diagnostic MRI and planning CT images from prior RT were chosen. Manual contouring of hippocampal volumes was performed by a radiation oncologist and neuroradiologist, with reference to the Radiation Therapy Oncology Group (RTOG) hippocampal contouring atlas. Auto-segmentation of the hippocampus was performed using Brainlab Elements (Brainlab, Feldkirchen, Germany) auto- segmentation software. Manual and auto-segmentation of hippocampal contours were compared using the Dice Similarity Coefficient (DSC). A DSC value of > 0.8 was pre- determined to be a clinically acceptable value from available literature. IMRT and VMAT plans were generated using Eclipse treatment planning system TM (Varian Medical Systems Inc.). Plans were prescribed to a median of 30Gy in 10 fractions, using the HIPPO trial protocol (ClinicalTrials.gov ID: NCT02147028) to guide planning target volume (PTV) coverage and organ at risk (OAR) constraints. OARs included the lens, eyes, optic nerves, optic chiasm, parotids, and cochlea. Results The average DSC for manual compared to auto- segmentation contouring of the left and right hippocampus was 0.66 and 0.67 respectively. The mean