7th ICHNO Abstract book

page 68 7 th ICHNO Conference International Conference on innovative approaches in Head and Neck Oncology 14 – 16 March 2019 Barcelona, Spain __________________________________________________________________________________________ 7th ICHNO

CBCT SIM CBCT SIM .This step ensures that patient anatomy in agrees with anatomy in re-CT. The workflow includes four steps: first, a CT-to-CBCT DIR is performed to warp the CT Hounsfield units onto the CBCT SIM anatomy. This step creates the Middle CT . Second, the pCT-to- CBCT RIR is performed using only the superior 5 cm portion CBCT. Following this RIR, the portion of registered pCT superior to the CBCT is saved as Superior CT . Third, the pCT-to-CBCT RIR is performed using only the inferior 7 cm portion CBCT. Following this RIR, the portion of registered pCT inferior to the CBCT is saved as Inferior CT . Finally, the Virtual CT and structure set is created as the composite of Inferior CT + Middle CT + Superior CT . All image registrations were applied using the commercially available MIM algorithm and were fully automatic without any operator adjustment. Two independent tests were performed to evaluate the novel workflow. Specifically, the first test measured accuracy of the Virtual CT structures (brain, brainstem and spinal canal) by calculating the DICE coefficient (DSC) between OARs on the Virtual CT and a re-CT. The second test compared high dose metrics including maximum dose (D max ) and the most exposed 1 % (D 1% ) between the dose distributions from the clinical plan calculated on the Virtual CT versus on the re-CT. Results Contour accuracy was assessed by comparing Virtual CT structures versus three independent observer contours directly on the re-CT. Mean DSC values are shown for each structure in Table 1. High dose metrics were assessed by comparing the dose distribution from the clinical plan calculated on the Virtual CT versus on the re-CT. Dose metrics for all patients are summarized in Figure 1.

(CC) and 0.6°(LR-axis), 1°(AP-axis), and 0.7°(CC-axis). Median pre-treatment height, weight and BMI were 170.2 cm (148-186), 77.97 kg (42-114) and 26.72 (18.58-34.58), respectively. All patients presented weight loss. Median weight variation at the 15 th treatment session was 4.3% (1.28-11.1) and at the end of treatment was 7% (2.56- 17.46). Statistical analysis showed no statistically significant correlation between mean displacements of first four CBCTs and mean of following CBCTs and no statistically significant correlation was observed between set-up errors and both BMI or weight change during treatment sessions.

Conclusion Translations and rotations recorded in this study were in agreement with literature data. Weight loss occurred in all treated patients but was not found any correlation between patient weight, BMI or weight loss. Therefore, we cannot support the hypothesis that set-up errors during radiotherapy are correlated to patient weight, BMI or weight loss. PO-130 Evaluation of a ‘dose-of-the-day’ workflow for head and neck cancer patients E. Berthelet 1 , L. Nica 2 , S. Hamilton 1 , E. Tran 1 , J. Wu 1 , M. Welsh 2 , N. McVicar 3 1 BC Cancer Agency - Vancouver, Radiation Oncology- Faculty of Medicine- Department of Surgery- UBC, Vancouver, Canada; 2 BC Cancer Agency-Vancouver, Radiation Therapy, Vancouver, Canada; 3 BC Cancer Agency - Vancouver, Medical Physics, Vancouver, Canada Purpose or Objective At our centre, the brain, brainstem and spinal canal are typically the top priority organs at risk (OARs) in Head and Neck patients. This study aims to evaluate novel workflow that employs computed tomography (CT) to cone‐beam CT (CBCT) deformable (DIR) and rigid image registrations (RIR) for the application of calculating the 'dose of the day' received by these OARs throughout treatment without requiring a second CT simulation (re-CT). Material and Methods Five patients were included in this study. Each patient had a planning CT, a re-CT and a CBCT acquired within ± 3 days of the re-CT. A novel MIM (MIM Software Inc v6.6.8, Cleveland, OH) workflow was developed to generate a virtual CT using the planning CT (pCT) and a CBCT. Prior to beginning the workflow, CBCT-to-re-CT DIR was performed to create

Conclusion The dose differences were clinically acceptable and the results obtained support further investigation into the use of the novel workflow to monitor high priority OARs during treatment. Clinical data collected in this study may enable us to improve clinical strategies for the delivery of adaptive radiotherapy in this patient population. It may also help refine the thresholds, timing, frequency and clinical situations in which re-planning is indicated. PO-131 Prognostic radiological features in early stage HPV positive oropharyngeal cancer M. Echevarria 1 , A. Trotti 1 , L. Harrison 1 , J. Caudell 1 1 H. Lee Moffitt Cancer Center, Radiation Oncology, Tampa, USA