Abstract Book

S1245

ESTRO 37

Results The group mean of the initial patient setup error for both WBRT and WBRT-LN patients was less than 5 mm in all directions, though systematic errors up to 4 mm were demonstrated (Table 1). Systematic residual error of the breast/chest wall contour and the surgical clip position were within 3.5 and 4.0 mm, respectively. For WBRT-LN, the residual systematic error of the nodal volume was less than 1.5 mm. Residual systematic rotational error of the chest wall was error of rotation 1.2°. Re-matching on the humeral head resulted in systematic errors of 3.7 and 2.7 mm in the SI and LR directions, respectively. Conclusion This study demonstrates the feasibility of a customised online imaging protocol using non-cardinal imaging angles. Considerable variation was observed from patients’ initial set up position based on skin marks. Clinical coverage of the anterior breast/chest wall contour and the nodal region for WBRT-LN was maintained with acceptably small residual errors. Furthermore, the implementation and evaluation of online imaging has enabled us to benchmark our accuracy in patient positioning during WBRT for future practice development. EP-2378 Comparison of Patient Setup Using 2D Bone with 3D Soft Tissue Matching For IGRT of Head and Neck Cancers R. Pasricha 1 , S. Goyal 2 , S. Goyal 2 , N. PATNI 3 1 All India Institute of Medical Sciences AIIMS-Rishikesh, Radiation Oncology, Rishikesh, India 2 Gujrat Cancer Research institute GCRI, RADIOTHERAPY, AHMEDABAD, India 3 BHAGWAN MAHAVEER CANCER HOSPITAL & RESEARCH CENTERBMCHRC, RADIOTHERAPY, JAIPUR, India Purpose or Objective Compare the 2D-2D manual matching using bony landmarks with 3D-3D soft tissue auto-matching and its effect on PTV margins and locoregional failures at the field edge (margins failure) in the patient of Oropharyngeal malignancy. Material and Methods In this study, 36 consecutive patients treated for stage III-IV ca oropharynx using concomitant chemo-IMRT protocol were image matched. To study the clinical consequences of 2D manual or 3D auto correction methods, both KV images and CBCT were obtained for each patient before IMRT and shifts were also calculated for both.Half of the patients were treated after applying shifts of 2D-2D matching and other half were treated after correcting for the shifts of 3D-3D matching in a random manner using the computer-generated numbers.Systematic and random errors were calculated separately for 2D-2D and 3D-3D matching. PTV margins for 2D and 3D matching were obtained for X, Y and Z axes Patients were assessed clinically for local control of tumor at one, three and six months, and radiologically by MRI/CT scans at six months.Unpaired student’s ‘t’ test was used for comparing mean between two groups while Chi-square test was used to compare proportions for two or more groups and to find out associations between two variables. Repeated Measure ANOVA was used for repeatedly measured continuous variables Results A total of 1080 individual CBCT and 1080 individual KV portal image were analyzed.Mean weight loss was insignificant in 2D & 3D groups when compared. On comparison of systematic and random errors obtained by

2D-2D and 3D-3D match along the three (X, Y, Z) respective axes were insignificant irrespective of the type of matching applied.PTV margins were (3.2, 3.2, 3.2) and (3.3, 3.9, 3.8) using 2D-2D and 3D-3D matching. At six months follow-up, the 3D-3D group had 5(27.78%) in field recurrences while the 2D-2D group had 3(16.67%). 2 patients in each group failed at PTV margins. All the marginal failures were in contralateral neck field. Conclusion 3D-3D volumetric soft tissue automatic match and 2D-2D KV image manual match using bony landmarks are equally effective for assessment of setup errors in oropharyngeal malignancy when done on daily basis.PTV margin obtained from 3D-3D soft tissue auto-matching were slightly larger than those obtained from 2D-2D manual matching. EP-2379 High-dose adrenal radiotherapy on a LINAC: planning and IGRT considerations and treatment times I.T. Kuijper 1 , W.F.A.R. Verbakel 1 , B.J. Slotman 1 , M. Dahele 1 1 VU University Medical Center, Department of Radiation Oncology, Amsterdam, The Netherlands Purpose or Objective Hypofractionated radiotherapy (e.g. stereotactic radiotherapy) is a radical local treatment option for adrenal metastases, associated with good local control and limited toxicity. This study reports practical considerations relevant to treatment on a conventional Planning and image-guidance (IGRT) strategies designed to reduce toxicity and risks associated with high-dose treatment are described (n=8 pts). In addition, (1) total treatment time (first imaging to end of last arc), (2) beam-on time and (3) image-guidance (IGRT) time (total treatment time minus beam-on time) are presented. Results For planning, in general two volumetric modulated arcs were used (RapidArc®, usually with avoidance sectors/partial arcs). Fractions ≤5Gy were delivered with 10MV flattening-filter free beam. Target and organ-at- risk (OAR) delineation was performed on the average intensity projection, using all relevant phases of a non- contrast free-breathing 4DCT with 2.5mm slice thickness. Mean free-breathing cranial-caudal target motion was 1.2cm (n=8). Breath-hold treatment (n=1/8) was considered if respiratory motion caused excessive overlap of PTV with OARs such as stomach and bowel. If available, co-registered FDG PET/CT was used to aid target identification. Contours for mobile dose-limiting OARs such as stomach and bowel were generated using the 0 and 50% phases and a 3 or 5mm planning OAR volume (PRV) margin added. Constraints were interactively changed for OAR and PRV to control dose gradients and reduce excessive dose next to the organ. An internal target volume (ITV) and PTV (typically ITV+5mm) were created (mean ITV/PTV=33/79cm 3 ). Optimization objectives were placed on the PTV-in-OAR overlap region to limit maximum dose. PTV overlapped with stomach/bowel PRV in 3 and 7 patients respectively. Mean ITV/PTV coverage with prescription dose was 93/88%. Patients were treated with knee and arm supports (no near-rigid immobilization). IGRT was cone-beam CT (CBCT) based, with registration on the adrenal gland and verification of OAR position. Excessive gas and sub-optimal acquisition parameters can impair target visualization on CBCT. However, treatment LINAC (TrueBeam®). Material and Methods