ESTRO 2025 - Abstract Book

S2310

Interdisciplinary – Health economics & health services research

ESTRO 2025

Conclusion: The cost of AI tools varied with purchase model and caseload. Following NICE’s cost estimates 2 , our cost comparison showed AI OAR auto-contouring could save money across tumour sites, particularly for breast where case volume and contour acceptability were high (~£126 per patient). AI was found to increase efficiency, but pathway audit is required to realise benefits. Further research is needed to describe the health economic impact of the additional activities and advanced treatment options enabled by use of AI auto-contouring.

Keywords: AI, auto-contouring, health technology assessment

References: 1. Van Dijk, L.V et al (2020). Improving automatic delineation for head and neck organs at risk by Deep Learning Contouring. Radiotherapy and Oncology. 142: 115-123. 2. NICE (2023). Artificial intelligence technologies to aid contouring for radiotherapy treatment planning: early value assessment. HTE11.

2710

Digital Poster Follow-up study on financial toxicity in cancer survivors having undergone proton therapy in Switzerland Léonie Grawehr 1 , Barbara Bachtiary 2 , Damien C Weber 2,1,3 1 Faculty of Medecine, University of Zurich, 8006 Zurich, Switzerland. 2 Centre for Proton Therapy, Paul Scherrer Institut, ETH Domain, 5232 Villigen, Switzerland. 3 Department of Radiation Oncology, University Hospital of Zurich, 8006 Zurich, Switzerland Purpose/Objective: This is a follow-up study on financial toxicity in cancer survivors treated with proton therapy (PT) between 2019 and 2021 in Switzerland.(1) The aim is firstly to evaluate the long-term financial status of cancer survivors several years post-treatment and secondly to identify risk factors for financial toxicity, to ultimately allow early intervention in the future.