ESTRO 2025 - Abstract Book

S1344

Clinical - Lung

ESTRO 2025

Conclusion: SBRT in LM from OM/OP CRC results in high LC rates, particularly with intensive regimens in non-heavily pretreated patients that experience longer DFS and OS. Due to possible occurence of secondary mutations in the pretreated population, a subset analysis in the chemotherapy-naive/first-line subset suggests an impact of KRASmut in radioresistance that should be further explored in a dose-escalation perspective.

Keywords: KRAS mutation, SBRT, lung oligometastases

2231

Digital Poster Streamlining Lung Cancer Treatment Planning: AI-Enhanced ATS Nodal Contouring Efficiency Ciaran Malone 1 , Darragh Browne 1 , Pierre Thirion 1,2 1 Department of Radiation Oncology, St.Luke's Radiation Oncology Network, Dublin, Ireland. 2 School of Medicine, Trinity St.James’s Cancer Institute, Trinity College Dublin, Dublin, Ireland Purpose/Objective: Artificial Intelligence is increasingly used in radiotherapy but its clinical safety, workflow integration, impact on practice, and patient benefit need further evaluation. In conventional thoracic radiotherapy for advanced NSCLC and limited-stage SCLC, the CTV includes initially involved nodes. Guidelines recommend optionally including the whole pathologically affected lymph node station based on the American Thoracic Society(ATS) nodal levels atlas. Accurate contouring of ATS nodal levels is essential in radiotherapy planning, particularly when using 4DCT imaging to account for respiratory motion. Manual contouring is time-consuming, error-prone, and challenging due to the 17 distinct nodal stations that need to be contoured across multiple breathing phases. This study evaluates the efficacy and time-saving potential of the MVision AI autosegmentation software's latest thoracic model in contouring ATS nodal levels compared to manual expert contouring.