S233
Clinical - Breast
ESTRO 2026
generated risk predictions for developing CVD, osteoporosis, COPD, and unfavorable body composition. Responders were grouped by European regions in accordance with UN Geoscheme. Results: Of the 130 individuals who accessed the survey, 96 radiation oncologists provided complete or partial responses. Median age was 47 years (range: 24-67). Responders represented 22 European countries (83.3% of all responders), as well as non-European regions (16.7% of all responders), with 92.7% being affiliated with urban institutions.Responders expressed largely positive attitudes toward the relevance of AI-generated risk predictions: 97.9% (CI95% 92.7-99.7) for CVD, 91.5% (CI95% 83.9-96.3) for osteoporosis, 62.3% (CI95% 51.7-72.2) for COPD, and 77.3% (CI95% 67.1-85.5) for unfavorable body composition. Perceived usefulness was slightly lower: 91.7% (CI95% 84.2-96.3) for CVD, 81.9% (CI95% 72.6- 89.1) for osteoporosis, 58.1% (CI95% 47.4-68.2) for COPD, and 70.5% (CI95% 59.8-79.7) for unfavorable body composition. Attitudes were more positive for CVD and osteoporosis than for COPD and unfavorable body composition. No significant regional differences were observed (Figure 1).
Situ: An ASTRO Clinical Practice Guideline. Shaitelman, Simona F. et al. Practical Radiation Oncology, Volume 14, Issue 2, 112 - 132 Keywords: Breast, intraoperative, partial radiotherapy
Digital Poster 1214 Radiation oncologists’ attitudes toward AI- generated risk prediction of chronic diseases in patients with early breast cancer: An international survey Frederik V Carstensen 1 , Belinda B Irankunda 1 , Eva Batista 2 , Desirée HJG van den Bongard 3,4 , Tanja Spanic 5,6 , Sofie AM Gernaat 7 , Helena M Verkooijen 7 , Maja V Maraldo 1 1 Department of Oncology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark. 2 Breast Unit, Champalimaud Foundation, Lisbon, Portugal. 3 Department of Radiation Oncology,, Amsterdam University Medical Centers, Amsterdam, Netherlands. 4 Cancer Treatment and Quality of Life / Cancer Biology and Immunology, Cancer Center Amsterdam, Amsterdam, Netherlands. 5 Europa Donna, European Breast Cancer Coalition, Milan, Italy. 6 Europa Donna, Slovenia, Ljubljana, Slovenia. 7 Division of Imaging and Oncology, University Medical Center Utrecht, Utrecht, Netherlands Purpose/Objective: In Europe, approximately 375.000 patients are diagnosed with breast cancer each year. While they generally have a favorable prognosis, most are diagnosed after age 50 and face increased risk of chronic diseases, partly due to treatment-induced side effects. Radiation therapy is a standard component of adjuvant breast cancer treatment. Funded by Horizon Europe, this research cooporative aims to develop, evaluate and implement AI-models to predict the risk of developing cardiovascular disease (CVD), osteoporosis, chronic obstructive pulmonary disease (COPD), and unfavorable body composition based on RT planning CT-scans. This study explored European radiation oncologists’ attitudes toward such AI- generated risk predictions and examined demographic differences. Material/Methods: An online survey was developed by the authors and subsequently approved by the ESTRO Scientific Council for distribution. In May 2025, the survey was distributed by ESTRO via e-mail to breast cancer radiation oncologists and promoted at two breast cancer sessions at ESTRO 2025. Access was provided via direct link or QR code to a REDCap database.The survey included demographic items (age, country, location of institution) and eight questions assessing the perceived relevance and usefulness of AI-
Conclusion: European radiation oncologists generally support AI- based risk predictions of chronic diseases in early breast cancer, particularly CVD and osteoporosis. These findings highlight readiness within the community to integrate AI into survivorship risk assessment. Such implementation could enable more personalized follow-up strategies and targeted prevention of late side effects. Keywords: Breast, AI, Risk prediction
Made with FlippingBook - Share PDF online