ESTRO 2021 Abstract Book

S1271

ESTRO 2021

Conclusion A dual isocentre technique for cervix treatments can enable an adaptive workflow that recovers the required plan quality. This illustrates the feasibility of a dual isocentre technique for the MRL.

PO-1546 Quality of adaptive treatment plan on a 0.35 T MR-Linac I. Bessieres 1 , A. Petitfils 1 , L. Aubignac 1 , O. Lorenzo 1 , J. Boudet 1 1 Centre Georges-François Leclerc, Medical Physics, Dijon, France

Purpose or Objective Stereotactic MR-guided adaptive radiation therapy (SMART) is possible on the 0.35 T MRIdian MR-linac from Viewray company. The adaptive workflow makes possible the re-optimization of the step-and-shoot IMRT plan according to a daily evaluation of the contours. In comparison to a non-adaptive workflow, no patient specific QA (PSQA) measurement is possible before the treatment delivery. Consequently, the quality and the deliverability of the adapted plan cannot be insured before the treatment. In this context, this work focused on a retrospective evaluation of the quality of our adapted plans. Materials and Methods Adapted plans of 26 patients treated for abdominal lesions (pancreas and liver) have been included. A number of 112 fractions have bee adapted (86%) resulting on a new plan. For each adapted plan, PSQA measurements have been performed a posteriori to its delivery with the 3D-matrix ArcCHECK and analyzed with 2%/2 mm gamma index criteria. The number of MU and the PSQA results of adapted plans have been analyzed as quality endpoints in comparison to the initial plan. Results For almost 50% of the patients, adapted plans had the same number of MU (±10%) and similar PSQA pass rate (±1%) in comparison to the initial plan. For 25% of the patients, adapted plans had less MU (-10% or less) and similar or better PSQA pass rate. For 25% of the patients, adapted plans had more MU (+10% or more) and decreased PSQA pass rate. These changes have been mostly correlated to significant organ at risk shape or position changes. For very few adapted plan, PSQA results were decreased to non-acceptable value in a non- adaptive workflow. Conclusion In most of the case, an equivalent or improved level of quality is reached by adapting a plan. The quality can be decreased (25%) but only in a very limited proportion up to a non-acceptable level. PO-1547 Dosimetric study of Adaptive radiotherapy (ART) for locally advanced head and neck cancer J. Das 1,2 , B. Goswami 3 , S. Goswami 3 , K.J. Deka 3 , G. Bora 3 , L. Das 3 1 State Cancer Institute, Gauhati Medical College and Hospital, Radiation Oncology, Guwahati, India; 2 state cancer institute, Gauhati Medical College and hospital, Assam, India, Radiation Oncology, Guwahati, India; 3 State Cancer Institute, Gauhati Medical College and hospital, Guwahati, Radiation Oncology, Guwahati, India Purpose or Objective The objective of this study is to evaluate the dosimetric parameters in adaptive radiotherapy for locally advanced head and neck cancers treated in a tertiary care cancer centre of Assam, India. Materials and Methods This is a hospital based prospective study conducted in the period from April 2020 to Jan 2021. Histologically proven Head and Neck Carcinoma patients with Stage III to IV (locally advanced) were selected for the study. A total of 10 patients receiving definitive, intensity-modulated radiation therapy (IMRT) to the head and neck region were evaluated for the study. Images were transferred electronically to MONACO (version 5.1) for contouring and planning. After the acquisition of CT images, target volumes, OARs were contoured in the planning CT. Images were again acquired midway during the planned course of radiation therapy. Body contours, target volumes, and organs at risk were redrawn on the new set of images. Two sets of additional treatment plans were generated: 1) a non-optimized plan (plan 2), which is an overlay of the original plan (plan 1) on the new set of contours, and 2) an optimized plan (plan 3) with the new set of contours. These 3 sets of plans were then compared for dosimetric differences. Results Four patients had locally advanced nasopharyngeal cancers, 4 patients had locally advanced oropharyngeal cancers, 2 patients had locally advanced hypopharyngeal cancer. The average reduction in gross tumour volume was 37.1 ml. The average changes in right and left parotid volume were 5.94 and 5.49 ml, respectively. In plan 1, the mean V26 of right and left parotids are (67.20±21.74)cc and (69.76±23.39)cc respectively. In plan 2(non-optimised), mean V26 of right and left parotids are (74.96±22.52)cc and (77.97±17.86)cc respectively. In plan 3(optimised) V26 of right and left parotids are (77.41±17.92) cc and (68.57±17.45) cc respectively. In plan 2, with the non-optimized plan, the average increase in maximum dose to the spinal cord was 9.8% (58.96-68.76; p= 0.156). With reoptimization, the maximum dose to the spinal cord decreased from 68.76% to 54.97% (mean difference, -13.79%, p=0.03). Average D99 for the planning target volume( dose received by 99% of the target volume) was 98.68% and 98.65% with the original and reoptimized plans, respectively.