ESTRO 2024 - Abstract Book

S4466

Physics - Machine learning models and clinical applications

ESTRO 2024

We established the RADIANT database by combining the multi-omics data from a phase III clinical trial with 589 LARC patients. AI models were developed and deployed to enhance the richness of the database. The database provides the basis for hypothesis-driven and hypothesis-free analyses. We aim to expand the database by collecting and processing the data from other rectal cancer trials and including real-world cohorts in the future.

Keywords: Rectal Cancer, Artificial Intelligence

References:

[1] ARISTOTLE: a phase III trial comparing standard versus novel chemoradiation treatment (CRT) as pre-operative treatment for magnetic resonance imaging (MRI)-defined locally advanced rectal cancer. https://doi.org/10.1186/ISRCTN09351447.

[2] J. Wasserthal, H. C. Breit, M. T. Meyer et al. TotalSegmentator: robust segmentation of 104 anatomical structures in CT images, arXiv, 2022.

[3] J. N. Kather, N. Halama, A. Marx et al. 100,000 histological images of human colorectal cancer and healthy tissue, Zenodo, 2018.

[4] S. Graham, M. Jahanifar, A. Azam et al. Lizard: A Large-Scale Dataset for Colonic Nuclear Instance Segmentation and Classification, arXiv, 2021.

1317

Proffered Paper

Identifying Lymphocyte Infiltration for Rectal Cancer Outcome Prediction Using Phase III Trial Data

Zhuoyan Shen 1 , Mikael Simard 1 , Douglas Brand 1,2 , Ying Zhang 1 , Gary Royle 1 , Andre Lopes 3 , Rubina Begum 3 , Nicholas West 4 , Ane Appelt 4 , Alexandra Gilbert 4 , Tim Maughan 5 , David Sebag-Montefiore 4 , Maria A Hawkins 1,2 , Charles-Antoine Collins Fekete 1 1 University College London, Department of Medical Physics and Biomedical Engineering, London, United Kingdom. 2 University College London Hospitals NHS Foundation Trust, Department of Radiotherapy, London, United Kingdom. 3 University College London, Cancer Institute, London, United Kingdom. 4 University of Leeds, School of Medicine, Leeds, United Kingdom. 5 University of Oxford, Department of Oncology, Oxford, United Kingdom

Purpose/Objective:

The prognostic relevance of the tumour microenvironment is well-established across human cancers. Lymphocyte infiltration is a key component and is associated with favourable outcomes across a variety of cancer types, including rectal cancer. However, in most published studies it is assessed manually in selected tumour regions. In this study, we sought to apply artificial intelligence (AI) across the whole tumour area to explore the prognostic impact relating