ESTRO 2025 Congress Report | RTT track

Improving efficiency in adaptive RT
Bethany Williams, The Royal Marsden NHS Foundation Trust, London, UK.

Bethany Williams was an invited speaker who gave an overview of the challenges in improving efficiency in the field of adaptive radiotherapy. An example she gave was the optimisation of image acquisition and plan robustness, which influences the speed of the workflow but is subject to software and hardware, which are the responsibility of vendors. Artificial intelligence (AI) can help, but requires training to verify automation and guarantee high-quality treatment. Then there is the need to upskill radiation therapists (RTTs), which was a topic that came back a few times in other presentations, but resources for training are limited, and RTT-led credentialling is needed.
 

RTT-led adaptive radiotherapy credentialling
Meegan Shepherd, Northern Sydney Cancer Centre, Royal North Shore Hospital, St Leonards, Australia. Medical Imaging and Radiation Sciences, Monash University, Clayton, Australia.

The experience and research of the staff of the Northern Sydney Cancer Centre showed how important it was to equip RTTs with skills and knowledge that play a role in opening pathways for leadership roles. [A1] By leveraging the expertise of RTTs, healthcare can optimise patient care that involves workflow challenges and rising costs.

My next focus point was “simulation free” or “direct to treatment”. Following from Bethany her talk, simulation free is also a way of improving efficiency that was well explored by a few institutes. These two poster presentation showed that is feasible without influencing the quality of the treatment.[A2] 
 

Delivery of online adaptive palliative radiotherapy without pre-treatment CT imaging
Koen J. Nelissen, , Amsterdam UMC, Vrije Universiteit Amsterdam, The Netherlands. Cancer Treatment and Quality of Life, Cancer Centre Amsterdam, The Netherlands.

This talk was focused on how metastases can be treated through the use of online adaptive radiotherapy (oART) when the radiation oncologist does not have diagnostic images made previously with cone beam CT (CBCT). The quality of the results was not significantly different from that found in the FAST-METS trial, in which diagnostic CT was used. Both techniques have been widely investigated and even implemented as standard.
 

Simulation-free MR-guided prostate radiotherapy: using diagnostic MR for reference planning
Francis Casey, The Joint Department of Physics, The Royal Marsden Hospital and the Institute of Cancer Research, London, UK

This speaker had looked at using non-radiotherapy-dedicated MR scans as references for treatment planning for MR-only prostate oART on the Elekta Unity. Large anatomy changes were deformed in an adaptation workflow, but autosegmentation could be of use to eliminate manual recontouring. This was another fascinating study that illustrated the options that could improve the efficiency of the oART process.

Recontouring in oART is time-consuming, and various studies have considered how to improve the efficiency of this step.
 

Feasibility of therapist-driven MR-guided adaptive radiotherapy for oligometastatic disease: geometric accuracy and dosimetric impact
Amanda Moreira, Radiation Medicine Programme, Princess Margaret Cancer Centre, Toronto, Canada

This study evaluated geometric accuracy and dosimetric uncertainty between radiation oncologists and RTTs recontouring of all target and organ-at-risk volumes during stereotactic body radiation therapy of oligometastatic disease. The results showed that RTT recontouring was clinically acceptable, and the use of this method could lead to an RTT-led workflow, which would remove some resource burden and improve access to this technique.
 

Evaluation of unsupervised use of AI-generated structures in CBCT-guided online adaptive radiotherapy for patients with urinary bladder cancer
Lisette J. Sandt, Department of Oncology, Copenhagen University Hospital – Herlev and Gentofte, Copenhagen, Denmark

The use of AI in oART is increasing. In this study, Ms Sandt evaluated unsupervised AI structures (target and organs-at-risk). There was a strong similarity between the structures, with small clinical differences, so the results indicate that the use of AI structures may speed up oART treatments.

Lisa Wiersema
MRlinac and research RTT
Antoni van Leeuwenhoek – The Netherlands Cancer Institute