ESTRO 2021 Abstract Book

S1291

ESTRO 2021

radiotherapy system. The majority of key landmarks identified were within 5mm, however visual inspection identified discrepancies not apparent within the clinical system. Independent validation of the underlying dose accumulation DIR highlighted non-physical deformations indicating care is required when interpreting the reported total treatment dose. Investigation of more patients and sites will follow. PO-1567 Initial dosimetric experience using daily AI-assisted adaptive radiotherapy for laryngeal cancer N. Weizman 1 , P. Blumenfeld 1 , M. Wygoda 1 , I. Darrs 1 , A. Meirovitz 1 , J. Menhel 1 , J. Feldman 1 , A. Popovtzer 1 1 Hadassah medical center, Radiation department, Jerusalem, Israel Purpose or Objective Adaptive radiation therapy (ART) provides a method to modify the radiation treatment plan secondary to structural and spatial changes that occur during a treatment course. Specifically, ART is relevant in head and neck cancers as many patients will experience changes of the target volumes and organs-at-risk (OARs) during treatment due to combination of treatment response, weight loss, inflammation and effects on normal tissues. Our institution recently began performing daily ART for select cases of locally advanced larynx cancers using a commercially available, online adaptive platform (ETHOS Varian) with AI-assisted workflows on daily cone-beam computed tomography (CBCT). Herein we report our initial dosimetric experience using this novel technique for laryngeal cancers. Materials and Methods All patients were presented in the weekly physician conference. Patients potentially benefiting from daily adaptation underwent CT-simulation and were planned on the AI system. A total of three patients being treated (105 radiotherapy sessions/plans) were imaged using kV-CBCT prior to treatment. Subsequent workflow included: 1) patients’ set up, positioning and image approval, 2) AI-based auto-contour and structure deformation of OARs and Targets on CBCT, 3) editing of contours by treating physician, 4) two plans are generated, a CT sim-based plan with deformed structures(scheduled) and a re-optimized structures deformed plan (adaptive), 5) plans evaluation and approval of best option, 6) plan QA and 7) plan delivery. Data was collected using Ethos TPS and statistical analysis comparing the scheduled versus adaptive plans was performed using STATA 13.0. Results 105 adaptive and scheduled radiotherapy plans were generated of which 78 adaptive were delivered. Average daily change in PTV size was: 1.5 cc (95% CI: -0.9cc - - 2.0 cc, p<.001). In all cases, cumulative adaptive plans demonstrated superior PTV coverage by 1.1% (95% CI: 0.6%-1.5%, p<.001) and a reduction in maximum dose by -2.4% (95% CI: -1.9% - 2.8% - , p<.001). High priority OARs demonstrated a reduction in mean dose for the laryngeal constrictors by 2.2Gy ± 0.1Gy, submandibular gland right and left, 2 Gy ± 0.1Gy and 2.5 Gy ± 0.1Gy respectively. Conclusion AI-assisted adaptive radiation therapy for larynx cancer is feasible and may potentially improve the therapeutic index for different tumor types. Prospective clinical evaluation is indicated to further this novel approach. PO-1568 Accumulated dose basing on Extended CBCT in adaptive radiotherapy for head and neck patients M. Gruda 1 , M. Murawska 1 , D. Kopeć 1 , D. Bodzak 1 , M. Piszczek 2 , D. Kiprian 3 , A. Zawadzka 1 1 The Maria Sklodowska-Curie National Research Institute of Oncology, Medical Physics Department, Warsaw, Poland; 2 The Maria Sklodowska-Curie National Research Institute of Oncology, Radiotherapy Department I, Warsaw, Poland; 3 The Maria Sklodowska-Curie National Research Institute of Oncology, Head and Neck Cancer Department, Warsaw, Poland Purpose or Objective Recalculating dose distribution using imaging performed during a treatment session allows for monitoring patient anatomy changes and their impact on dose delivery. The purpose of this study was to minimize the difference between planned and delivered dose for head and neck patients undergoing definitive radiotherapy. Materials and Methods Adaptive radiotherapy protocol was applied for 11 patients treated for head and neck cancer with a total treatment time ≥ 6 weeks. Treatment plans were generated using RayStation TPS (RaySearch Laboratories, Sweden) in VMAT technique (6MV photon beams). During treatment sessions 2, 7, 12, 17, 22, 27 and 32 additional Extended Scan CBCT was performed. Subsequently, the dose distribution was recalculated in TPS using the Dose Tracking option and deformed to a pre-treatment CT scan. Treatment delivery was estimated using Accumulated Delivery Dose tool and the remaining dose was predicted using current patient anatomy and Estimate Fraction Dose tool. For each patient new CT scan was performed after the 20 th session of RT (CT2). Subsequently, a new treatment plan was generated, using a delivered dose for OARs as a Background Dose. Monitoring the delivered dose using CBCT was continued for the rest of the treatment. On completion of RT total delivered dose was estimated for PTVs, elective CTVs (CTV1 and CTV2) and organs at risk. The Wilcoxon signed pairs rank test was used to test the statistical differences between the analyzed parameters. Results For all patients, target volumes were lower at the time of the second CT scan compared to the pre-treatment scan (average difference of 14.2% for GTV, 6.8% for CTV1 and 5.2% for CTV2). Left and right parotid gland volumes were lower by 14.2% and 13.8% respectively. CBCT accumulated dose (pre-CT2) showed an increase for parotids D mean by 1.5%, spinal cord D max by 2%, mandible D max by 1%. An increase in D98%, D95% and D mean for GTVs was observed (0.7%). For elective CTV2 D95% was lower by 0.5% on average. Replanning treatments using CT2 after the 20 th fraction allowed to reduce dose differences between planned and total delivered dose to statistically insignificant levels (p<0.05). For OARs an improvement in dose distribution was observed in 8 of 11