ESTRO 2023 - Abstract Book

S698

Monday 15 May 2023

ESTRO 2023

molecular subtypes. Furthermore, no evidence was found of a radiotherapy-induced upregulation of PD-L1 in early-stage BC. This latter result calls into question the therapeutic strategy of combining neoadjuvant PD-L1 inhibitors with a RT boost in high-risk early-stage BC, which is currently under investigation in several clinical trials.

Proffered Papers: Highlights of Proffered Papers - Latest clinical trials

OC-0828 Benefit of AI-assisted organ-at-risk contouring in head-and-neck cancer: A global randomized study M.E. Rasmussen 1 , K. Akbarov 2 , E. Titowich 2 , K. Wakeham 2 , J.A. Nijkamp 3 , W. Van Elmpt 4 , A.K.U. Ahmed Mohamed 5 , B.P. Brohet Kartika Erida 5 , C.N. Darejan Lomidze 5 , G.P. Gustavo Ferraris 5 , H.M. Igor Stojkovski 5 , I.I. Issa Mohamad 5 , L.S. Lotfi Kochbati 5 , L.E. Maksim Piatkevich 5 , M.M.B.J. Orges Spahiu 5 , R.A. Raushan Zakirova 6 , S.S.S.K. Uranchimeg Tsegmed 5 , S.S. Korreman 7 , J.G. Eriksen 7 1 Principal Investigator, The ELAISA Consortium, Aarhus, Denmark; 2 International Atomic Energy Agency, The ELAISA Consortium, Vienna, Austria; 3 Consultant, The ELAISA Consortium, Aarhus, Denmark; 4 Consultant, The ELAISA Consortium, Maastricht, The Netherlands; 5 Chief Scientific Investigators, The ELAISA Consortium, Vienna, Austria; 6 Chief Scientific Investigators, The ELAISA Consortium , Vienna, Austria; 7 Principal Investigator and shared last author, The ELAISA Consortium, Aarhus, Denmark Purpose or Objective Global roll-out of artificial intelligence-assisted (AI-assisted) contouring has immense potential for the access and quality of care in radiotherapy. However, the current evidence is primarily from high-income countries. The purpose of this study was to compare inter-observer variation (IOV) between radiation oncologists (ROs) in low- and middle-income countries (LMICs) using either AI-assisted or manual contouring of organs-at-risk in head-and-neck cancer. Materials and Methods 97 ROs from 23 institutions in 22 nations across 5 continents were invited to the study. The institutions were randomized to either manual contouring (control) or AI-assisted contouring (intervention) of 8 common organs-at-risk of head-neck cancer. Randomization was balanced on (1) the annual number of head-neck cases treated at institutions and (2) the availability of auto-contouring at institutions. Four head-neck cases were randomly assigned to ROs institution-wise resulting in 2-3 institutions (8-15 ROs) per case in each group. Deep learning-based auto-contours were made with MVision AI Oy, Helsinki, Finland. Contouring was performed online in EduCase™ (RadOnc eLearning Center, Inc). ROs were informed about the contouring guidelines used in the study and that their task was to “generate clinically acceptable contours”. ROs were also asked to complete in a short survey on their professional backgrounds. IOV was quantified as medians of Dice coefficients and Hausdorff distances 95th percentile (HD95) between contours of the ROs and a median contour within groups. Median contours were made by summing structures and thresholding where the median number of ROs or more had included a given voxel. High Dice and low HD95 indicate low IOV. Confidence intervals were estimated with bootstrap and groups were compared with the Mann-Whitney U test. Results 89 ROs handed in the contours and 74 completed the survey. The four cases were contoured respectively by 18 (7/11), 24 (12/12), 23 (11/12) and 24 (12/12) ROs (Manual/AI-assisted) (See Table 1 for details). Table 1