ESTRO 2024 - Abstract Book

Radiation Oncology: Bridgingthe Care Gap 3-7M ay2024 Glasgow,UK

Radiotherapy &Oncology Journal of the European SocieTy for Radiotherapy and Oncology

Volume 194 Supplement 1 (2024)

Radiotherapy & Oncology is available online: For ESTRO members: http://www.thegreenjournal.com For institutional libraries: http://www.sciencedirect.com

Amsterdam  Boston  London  New York  Oxford  Paris  Philadelphia  San Diego  St. Louis

International Standard Serial Number 0167-8140

© 2024 Elsevier B.V. All rights reserved.

This journal and the individual contributions contained in it are protected under copyright, and the following terms and conditions apply totheir use in addition to the terms of any Creative Commons or other user license that has been applied by the publisher to an individual article: Photocopying Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission is not required for photocopying of articles published under the CC BY license nor for photocopying for non-commercial purposes in accordance with any other user license applied by the publisher. Permission of the publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Derivative Works Users may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions or companies. Other than for articles published under the CC BY license, permission of the publisher is required for resale or distribution outside the subscribing institution or company. For any subscribed articles or articles published under a CC BY-NC-ND license, permission of the publisher is required for all other derivative works, including compilations and translations. Storage or Usage Except as outlined above or as set out in the relevant user license, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publisher. Permissions For information on how to seek permission visit www.elsevier.com/permissions Author rights Author(s) may have additional rights in their articles as set out in their agreement with the publisher (more information at http://www.elsevier.com/ authorsrights). Notice Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, com pounds or experiments described herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. To the fullest extent of the law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality or value of such product or of the claims made of it by its manufacturer

Orders, claims, and journal inquiries: Please visit our Support Hub page https://service.elsevier.com for assistance. Funding Body Agreements and Policies

Elsevier has established agreements and developed policies to allow authors whose articles appear in journals published by Elsevier, to comply with potential manuscript archiving requirements as specified as conditions of their grant awards. To learn more about existing agreements and policies please visit https://www.elsevier.com/fundingbodies

Available online at www.sciencedirect.com ScienceDirect

3-7 May 2024 Glasgow, UK

Table of contents

Invited Speaker..................................................................................................................................................................1-150

BRACHYTHERAPY

Breast.............................................................................................................................................................................. 151-167 GI, paediatric, miscellaneous.......................................................................................................................................168-198 General........................................................................................................................................................................... 199-206 Gynaecology................................................................................................................................................................... 207-287 Head & neck, skin, eye..................................................................................................................................................288-325 Physics............................................................................................................................................................................ 326-391 Urology ..........................................................................................................................................................................392-432

CLINICAL

Biomarkers..................................................................................................................................................................... 433-445 Breast.............................................................................................................................................................................. 446-734 CNS.................................................................................................................................................................................. 735-972 Gynaecology.................................................................................................................................................................973-1126 Haematology..............................................................................................................................................................1127-1176 Head & neck...............................................................................................................................................................1177-1461 Lower GI.....................................................................................................................................................................1462-1571 Lung............................................................................................................................................................................1572-1828 Mixed sites, palliation...............................................................................................................................................1829-2013 Paediatric tumours....................................................................................................................................................2014-2071 Sarcoma, skin cancer, malignant melanoma..........................................................................................................2072-2125 Upper GI.....................................................................................................................................................................2126-2286 Urology.......................................................................................................................................................................2287-2649

INTERDISCIPLNARY

Education in radiation oncology..............................................................................................................................2650-2721 Global health..............................................................................................................................................................2722-2761 Health economics & health services research........................................................................................................2762-2871 Other ..........................................................................................................................................................................2872-2987

PHYSICS

Autosegmentation.....................................................................................................................................................2988-3166 Detectors, dose measurement and phantoms......................................................................................................3167-3411 Dose calculation algorithms.....................................................................................................................................3412-3458 Dose prediction, optimisation and applications of photon and electron planning............................................3459-3740 Image acquisition and processing including ML based methods.........................................................................3741-3972 Inter-fraction motion management and offline adaptive radiotherapy .............................................................3973-4132 Intra-fraction motion management and real-time adaptive radiotherapy..........................................................4133-4409 Machine learning models and clinical applications...............................................................................................4410-4580 Optimisation, algorithms and applications for ion beam treatment planning...................................................4581-4731 Quality assurance and auditing . .............................................................................................................................4732-4954 Radiomics, functional and biological imaging and outcome prediction..............................................................4955-5157

RADIOBIOLOGY

Immuno-radiobiology. ..............................................................................................................................................5158-5191 Microenvironment.....................................................................................................................................................5192-5231 Normal tissue radiobiology......................................................................................................................................5232-5310 Tumour radiobiology................................................................................................................................................5311-5440

RTT

Patient care, preparation, immobilisation and IGRT verification protocols ........................................................5441-5630 Patient experience and quality of life .....................................................................................................................5631-5735 Education, training, advanced practice and role developments..........................................................................5736-5857 Service evaluation, quality assurance and risk management ..............................................................................5858-5934 Treatment planning, OAR and target definitions ...................................................................................................5935-6038

SPEAKER ABSTRACT Invited Speaker

S2

Invited Speaker

ESTRO 2024

3276

3D printing

Bertrand Dewit

KU Leuven, Laboratory Experimental Radiotherapy, Leuven, Belgium

Abstract:

The integration of 3D printing technology into radiotherapy practice represents a promising avenue for innovation and advancement. This emerging technology is poised to significantly transform various aspects of radiotherapy treatment offering the capability to produce patient-specific devices such as 3D printed bolus, brachytherapy applicators, electron apertures and immobilization solutions within a clinical timeframe. This integration not only introduces new tools but also necessitates a reconfiguration of workflows and roles within radiotherapy teams. The digital nature of 3D printing alters the creation process of these devices, thereby shifting the roles and responsibilities of Radiation Therapists (RTTs). For instance, with 3D-printed bolus, RTTs are no longer required to physically scan patients with bolus; instead, they utilize digital bolus data from treatment plans. However, ensuring accurate placement of the patient-specific bolus becomes even more paramount, thereby shifting the role of RTTs. In this context, SGRT tools and workflows can for example assist RTTs in ensuring precise device placement and treatment delivery. As 3D printing continues to evolve and slowly integrate into radiotherapy workflows, it provides insight into anticipated shifts in practice, underscoring the importance of adaptability among RTTs to embrace innovation to optimize patient care and treatment outcomes.

3277

A cyber-attack response plan: How to ensure that patients continue to be treated (correctly)

Samuel Peters

Kantonsspital St.Gallen, Klinik für Radio-Onkologie, St.Gallen, Switzerland

Abstract:

The threat of cyber attacks is on the rise, posing significant challenges to healthcare services, including those involved in radiation therapy. For example over the period from 2016 to 2021, the incidence of ransomware attacks in the healthcare sector in the US witnessed a more than twofold increase, impacting 42 million patients. The escalation of such attacks can be attributed to the widespread accessibility of personal and financial data, providing cybercriminals with opportunities to demand substantial ransom amounts in exchange for stolen information. Most experts agree that it is not a question of "if", but "when" you will be affected. In the event of a cyberattack on a hospital, the radiation oncology department is most severely affected as the daily routine relies fully on the availability of digital data. This means a complete standstill of the department and all the treatments. This may be due to a complete encryption of servers, workstations and even treatment devices by

S3 ESTRO 2024 malware. It may also be possible, that all servers and the entire network are shut down preventively by the central IT department to prevent further spread of the malware. In both cases, "normal" treatment of patients is no longer possible. Past events showed, that such a situation can last up to several weeks. To ensure that treatment can still continue during such an event, 6 things need to be considered – some preventative, others during this extraordinary situation: 1. Preparedness in the event of a cyberattack (having a business continuity and recovery plan (BCP), identify systems and affected personnel, hold backups, regular testing of BCP) 2. Preventive measures (User awareness, antivirus policy, system patching) 3. Detection and reaction (detection tools, data breach identification, system isolation) 4. Respond (activation of a BCP: communication, treatment and data handling in case of an event) 5. Recovery (activation of recovery plan, check recovered data, merge recovered and during event created data) 6. Debriefing (recap past events, adapt reaction and recovery plan) The key inquiries that every department must address are: What preparations and measures are currently in place? How can we guarantee the safe treatment of patients? To streamline the approach to this often challenging task, we present a few user-friendly methods, primarily focusing on preparedness and the steps to be followed during the response phase. Preparedness involves the development of strategies, implementation of security measures, and creation of incident response plans to mitigate the impact of cyber threats. This encompasses activities like creating data backups (online, offline, even on paper) and establishing a clear plan for detecting, responding to, and recovering from cyberattacks. The response phase entails activating the previously defined business continuity plan. This initiation follows predetermined procedures and protocols to ensure that patient treatment can resume swiftly, minimizing downtime. It includes tasks such as informing relevant parties like employees and patients, conducting regular meetings and treatments with limited (or no) access to initial data. This may involve redirecting patients to other facilities temporarily or suspending treatments. Rigorous documentation of all actions taken is imperative during this phase. Invited Speaker

3278

Accumulation across different modalities

Monica Serban

Radiation Medicine Program, Princess Margaret Cancer Centre, Department of Radiation Oncology / Medical Physics, Toronto, Canada

Abstract:

The integration of EBRT and BT significantly enhances the therapeutic outcomes for gynecological cancer patients. Accurately accumulating radiation doses from BT and EBRT modalities, whether delivered sequentially over a single course of multi-fractionated RT or in the context of re-irradiation, has become increasingly important. However, the prevailing solutions for dose accumulation, typically done via rigid or deformable image registration, suffer from large uncertainties. This presentation reviews the complexities associated with dose summation between BT and EBRT and offers practical clinical solutions while exploring emerging methodologies and future directions We discuss the challenges inherent in dose summation between EBRT and BT, focusing on two components. The first component covers the uncertainties in current image registration methods between EBRT and BT images due to difference in patient setup, changes in soft tissue in the presence or absence of applicators, physiological changes in

S4 ESTRO 2024 bladder and rectum filling, or biological response of tumor to treatment. The second component covers the uncertainties arising from combining doses with varying fractionation, including the limitations of conventional radiobiological models for dose accumulation. The unique difficulties encountered in reirradiation scenarios involving BT further underscore the necessity for robust summation methodologies Practical clinical solutions for addressing these challenges are presented, including direct addition of EQD2 doses between BT and EBRT, with a worst-case scenario assumption regarding hotspot doses. Furthermore, evidence supporting the use of organ D2cm 3 in gynecological cancer for establishing dose-effect relationships for morbidity endpoints is presented Case examples will illustrate dose accumulation between BT and EBRT using image registration that highlight the application of advanced techniques in clinical practice. Furthermore, novel alternative approaches and future outlooks in dose summation are discussed. These include dose accumulation on hollow organ walls utilizing dose surface maps and the potential of deep-learning approaches for deformable image registration and dose accumulation between EBRT and BT This presentation aims to provide a comprehensive understanding of the challenges associated with dose summation in combined EBRT and BT treatments for gynecological cancer. By exploring practical solutions and emerging methodologies, it seeks to facilitate advancements in treatment planning and optimization, ultimately improving patient outcomes. Invited Speaker

3280

Accumulation for treatment monitoring

Lena Nenoff

OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Cal Gustav Carus, TUD Dresden University of Technology, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany. Helmholtz-Zentrum Dresden-Rossendorf, OncoRay, Dresden, Germany

Abstract:

Dose accumulation can merge the doses calculated on different images at different time points within or between fractions or even between different treatments. To combine doses, the underlying anatomical images have to be registered and the registration result is used to map the doses. In some anatomical areas this can be done with rigid image registration, but in most areas this requires deformable image registration (DIR). The mapped doses can be added up on the reference image. While DIR has a major impact on the accumulated dose, other effects such as biological effects, equivalent dose (EQD) or biological effective dose (BED) calculation, energy conservation and interpolation are relevant secondary effects that contribute to the overall dose accumulation uncertainty. In clinical practice, dose accumulation is applied on different time scales to monitor the treatment progress in different treatment phases. Depending on the time difference the dose accumulation uncertainty varies. Additionally, the impact of the dose accumulation depends on its application (Figure 1). The fastest time scale is a simple 4D dose calculation for tumors affected by intrafraction breathing motion. The dose is calculated in different breathing phases, mapped to a reference phase, and added up. This can be used to evaluate interplay effects [1,2] or the impact of the starting phase [3]. The uncertainty of 4D accumulated doses is rarely quantified, but likely highest in the area of the dose gradient and sliding surfaces, such as lung-rib cage [4]. Between fractions, dose accumulation is used to add up the summed treatment dose. With widely available daily 3D images the daily delivered dose can be reconstructed. Differences between planned and delivered doses can be evaluated and plan adaptation can be indicated [5]. Inter-fraction dose accumulation can be challenging especially in the presence of large changes or during adaptive radiotherapy [6]. The uncertainty the DIR and therefore the

S5 ESTRO 2024 accumulated dose can be increased compared to intrafraction dose accumulation, for example due to different imaging modalities (e.g. CT to CBCT or CT to MRI). Also mass changes of the patient and/or tumor can lead to an increased uncertainty of the DIR or of the dose mapping/interpolation method [7]. Additionally, the mapped dose to the filling of hollow organs such as bladder or bowel is not meaningful, since the filling was exchanged from one day to the other. Especially during adaptive radiotherapy with substantial changes in the target or relevant organs at risk (OARs) the accumulated doses between adaptive treatments are hard to interpret, since the treatment goal might have changed for some voxels. Combined treatments between different treatment modalities do require dose accumulation. The corresponding dose accumulation uncertainties varies. For example, if proton, photon and electron doses are optimized on the same image, the dose accumulation uncertainty is small. It can however be large if different patient anatomies are enforced, as in the combined radiotherapy and brachytherapy. This can extent the common regularization parameters implemented in most DIRs [8]. Different BEDs between different modalities can further increase the dose accumulation uncertainty. This can be even more extreme in the case of reirradiations. With months or even years between the treatments, anatomical changes can be large, for example due to surgical resection or fibrosis [9,10]. A dose mapping to the new anatomy can be important to account for previously irradiated dose in individualized prescriptions and reduced dose tolerances to OARs and normal tissue [11]. Recovery patterns of normal tissues after irradiation are rarely investigated, but will gain importance with increasing numbers of reirradiations [9,12]. The mapped dose from the primary treatment can also be correlated to places of local failure and side effects visible in follow up images to the previous delivered dose [13]. Outcome modeling employs inter-patient DIR to map the planned (or delivered) dose to a reference anatomy. Side effects are correlated to specific voxel groups [14]. This method does contain large DIR uncertainties, due to large inter-patient differences but has the advantage that it allows for substructure analysis. In summary, dose accumulation did arrive in clinical practice. Its uncertainties are however not well quantified, depending on the application accumulated doses should be interpreted with care [15]. Invited Speaker

[1]

https://doi.org/10.1016/j.phro.2023.100465.

S6

Invited Speaker

ESTRO 2024

[2]

https://doi.org/10.1016/j.radonc.2020.07.055.

[3]

https://doi.org/10.1016/j.phro.2023.100473.

[4] https://doi.org/10.1016/j.semradonc.2019.02.007.

[5] https://doi.org/10.1016/j.semradonc.2019.02.011.

[6]

https://doi.org/10.3389/fonc.2022.1086258.

[7]

https://doi.org/10.1016/j.radonc.2023.109527.

[8]

https://doi.org/10.1118/1.4903300.

[9]

https://doi.org/10.1016/j.radonc.2023.109585.

[10]

https://doi.org/10.1080/0284186X.2021.1982145.

[11]

https://doi.org/10.1016/S1470-2045(22)00447-8.

[12]

https://doi.org/10.1016/j.ijrobp.2010.08.021.

[13]

https://doi.org/10.1038/s41598-017-07586-x.

[14]

https://doi.org/10.1016/j.radonc.2023.109868.

[15]

https://doi.org/10.1088/1361-6560/ad0d8a.

3281

Addressing late effects of head and neck cancer patients

Sabina Khan

University College London Hospitals NHS Foundation Trust, H&N Oncology, Radiotherapy & Proton Beam Therapy, London, United Kingdom

Abstract:

The incidence of Head and Neck cancer (HNC) in the UK has increased by 34% since the 1990's whilst survival rates have increased due to earlier diagnosis and improved forms of treatment. Radiotherapy can play a vital role in the radical/ curative treatment of these cancers as either primary or adjuvant treatment. Late effects of treatment can manifest months or years after the completion of radiotherapy and are delayed responses to treatment. As survival rates increase, the consequence of treatment becomes more apparent. Whilst innovation in radiotherapy, such as Intensity Modulated Radiotherapy and Proton Beam Therapy, have contributed to the reduction of acute and chronic side effects, patients are still at risk of late effects that can significant effect their quality of life - physically, psychologically and socially.

There is a common misconception that the cancer journey ends when patients are cancer free- this is simply false. Patients experiencing late effects would argue that cancer needs to be seen as a chronic condition - where side

S7 ESTRO 2024 effects in the long term are often under and misdiagnosed. Often these need to be managed by different members of the team. This talk will highlight the possible late effects patients treated for HNC can experience, what survivorship means for this cohort of patients and the challenges in diagnosing late effects. We will explore the use of patient reported outcomes to address what is most important to the patient, not what the clinician thinks is important. Invited Speaker

3283

Advanced treatment concepts for upright radiotherapy

Thomas Bortfeld

Massachusetts General Hospital, Radiation Oncology, Boston, USA

Abstract:

This talk will shed light on the promise of upright treatment systems to democratize particle therapy by eliminating the need for the large, heavy and costly gantry. Upright treatments are particularly promising in particle therapy because the proton and heavy ion gantries are much bulkier than electron and photon beam gantries. In addition, particle therapy is much less dependent on flexibility of beam placement than standard IMRT/VMAT. It is the combination of this greater advantage of upright treatment in particle therapy, which can potentially facilitate the installation of proton therapy in conventional treatment rooms, and the smaller compromise of not having a gantry that has led to a resurgence of interest in upright fixed-beam treatments. We will first review the potentially reduced flexibility of beam placement in gantry-less upright treatment paradigms and its impact on the ability to shape dose distributions. An analysis of over 4000 patient treatments at our center has shown that the full flexibility of the gantries is not utilized. In fact, there are only a very limited number of gantry angles (primarily the "cardinal" angles), and couch angles at 0, 90, and 270 degrees that have been used far more frequently than others. While the current trend is toward the use of proton arc therapy (PAT), it has earlier been shown that the power of intensity-modulated proton therapy (IMPT) facilitates highly conformal dose delivery from a "small angle" approach. In fact, unlike IMRT/VMAT, which relies on spreading beam angles out over very large angles to achieve the desired level of conformality (which is due to the theory of the Radon transform), IMPT can achieve dose conformality from a single beam. However, for the purpose of skin sparing, spreading out beams is required to some degree. These findings facilitate upright treatment: while arc therapy can be performed on a chair, there is no need for wide arcs, and a single-sided, e.g. anterior, approach is usually sufficient, for example in the case of treatments of the breast. Non-coplanar beams have been shown to be beneficial in some situations. They pose a slightly larger but not unsurmountable challenge for upright treatments: the chair can be tilted backward, forward, or sideways by a limited amount, in addition the beam can be bent out of the horizontal plane. Moving the patient on a chair around a fixed beam requires new strategies in image guidance and real-time adaptation. Other presentations in this session have addressed some of these challenges. New opportunities arise because image guidance systems can be more easily combined with a fixed beam than with gantry-based beam delivery. MRI-guided proton therapy is becoming a possibility. We will report on our very early experience with an ultra-low field MRI system for breast imaging and image guidance. A conceptual design of an image-guided gantry less system integrated into a standard treatment room will be presented.

S8

Invited Speaker

ESTRO 2024

3284

AI for automation of contouring in brachytherapy

Matteo Maspero

UMC Utrecht, Radiotherapy, Utrecht, Netherlands

Abstract:

Radiotherapy holds a central role in cancer treatment, with the objective of delivering precise radiation doses to tumor tissues while safeguarding adjacent healthy structures. Automatic segmentation of target and organs at risk structures is indispensable, enhancing efficiency, accuracy, and consistency in treatment planning. This paper offers a thorough overview of the integration of deep learning techniques into brachytherapy for automatic segmentation. The necessity for automatic segmentation arises from the intricate and variable nature of anatomical structures, necessitating meticulous contouring for treatment planning. Pre-artificail intelligence segmentation techniques, such as manual contouring and atlas-based algorithms, exhibit limitations in coping with the escalating complexity of contemporary treatment plans. Deep learning, a subset of artificial intelligence, has emerged as a potent tool for automatic segmentation, providing improved accuracy, efficiency, and speed. The implementation of these technologies in clinical practice is examined, emphasizing the necessity for rigorous validation and seamless integration into existing workflows. In the context of brachytherapy, this paper scrutinizes the principal clinical applications and the current status of autocontouring. Addressing challenges and opportunities in applying deep learning to brachytherapy, the paper underscores the importance of tailored solutions for this distinctive treatment modality. The conclusion encourages thinking beyond prevailing applications, advocating for the exploration of AI in novel areas within radiotherapy. As technology advances, contemplating unconventional applications of deep learning, such as treatment response prediction and adaptive planning, may unlock new possibilities for optimizing cancer care. Various approaches to autocontouring using deep learning are explored, encompassing a review of available commercial solutions. The capabilities and potential benefits of these solutions in clinical settings are underscored.

3285

AI in brachytherapy for dose optimisation

Shirin Abbasinejad Enger

McGill University, Medical Physics Unit, Department of Oncology, Montreal, Canada

Abstract:

Artificial intelligence (AI), particularly deep learning, promises to streamline and increase various aspects of the brachytherapy workflow. This includes tasks such as tumour and organs at risk segmentation, catheter digitization, dose prediction, treatment plan optimization, quality control and assurance of treatment plans. This will improve

S9 ESTRO 2024 efficiency and accuracy by reducing human errors, minimizing inter-observer differences in segmentation and planning, and saving time. This presentation provides a comprehensive overview of recent advancements and applications of AI technologies in various aspects of brachytherapy, highlighting challenges, opportunities, and future directions. Invited Speaker

3287

Applicability and usage of dose mapping/accumulation in radiotherapy

Martina Murr 1 , Kristy K. Brock 2 , Marco Fusella 3 , Nicholas Hardcastle 4 , Mohammad Hussein 5 , Michael G. Jameson 6 , Isak Wahlstedt 7 , Johnson Yuen 8 , Jamie R. McClelland 9 , Eliana Vasquez Osorio 10 1 Section for Biomedical Physics, Department of Radiation Oncology, Tübingen, Germany. 2 The University of Texas MD Anderson Cancer Center, Department of Imaging Physics, Texas, USA. 3 Terme Hospital, Department of Radiation Oncology, Abano, Italy. 4 Peter MacCallum Cancer Centre & Sir Peter MacCallum Department of Oncology, University of Melbourne, Melbourne, Australia. 5 Metrology for Medical Physics Centre, National Physical Laboratory, Teddington, United Kingdom. 6 GenesisCare New South Wales, Faculty of Medicine and Health, New South Wales, Australia. 7 Technical University of Denmark, Department of Health Technology, Kongens Lyngby, Denmark. 8 St George Hospital, Cancer Care Centre, New South Wales, Australia. 9 Centre for Medical Image Computing and Wellcome and EPSRC Centre for Interventional and Surgical Sciences, Dept of Medical Physics and Biomedical Engineering,, London, United Kingdom. 10 Division of Cancer Sciences, Faculty of Biology,, Medicine and Health, Manchester, United Kingdom Dose mapping/accumulation (DMA) has been a topic in radiotherapy (RT) for years but has not yet found its widespread way into clinical RT routine. During the ESTRO Physics workshop 2021 on ‘‘commissioning and quality assurance of deformable image registration (DIR) for current and future RT applications”, we built a working group on DMA from which we present the results of our discussions in this symposium based on the article [1]. This article aims to shed light on the current DMA situation in RT and highlight the issues that hinder its conscious integration into clinical RT routine. As a first outcome of our discussions, we present a scheme (cf. Figure 1) where representative RT use cases are positioned, considering expected anatomical variations and the impact of dose mapping uncertainties on patient safety, which we have named the DMA landscape (DMAL). This tool is helpful for future reference when DMA applications get closer to clinical day-to-day use. Secondly, we discussed current challenges, lightly touching on first-order effects (related to the impact of DIR uncertainties in dose mapping) and focusing in detail on second-order effects often dismissed in the current literature (such as resampling and interpolation, quality assurance considerations, and radiobiological issues). Finally, we developed recommendations and guidelines for vendors and users. Our main points include: • Striving for context-driven DIR (considering their impact on clinical decisions/judgments) rather than perfect DIR. • Being conscious of the limitations of the implemented DIR algorithm. • Consider when dose mapping (with properly quantified uncertainties) is a better alternative than no mapping. Abstract:

S10

Invited Speaker

ESTRO 2024

Figure 1. DMAL presents the current landscape of use cases. The span of each box represents the typical ranges in the anatomical variation and the expected impact of dose mapping uncertainties for a given use case.

[1] Murr M, Brock KK, Fusella M, Hardcastle N, Hussein M, Jameson MG, et al. Applicability and usage of dose mapping/accumulation in radiotherapy. Radiother Oncol 2023;182. https://doi.org/10.1016/j.radonc.2023.109527.

3288

Are guidelines followed in the real world?

Lasse Refsgaard

Aarhus University Hospital, Danish Center for Particle Therapy, Aarhus, Denmark. Aarhus University, Department of Clincal Medicine, Aarhus, Denmark

Abstract:

When practise-changing radiotherapy trials provide high level evidence, the radiotherapy community has an intent to change treatment practices. This intent is crystallised in changes of clinical guidelines and by extension, hopefully the way treatments are carried out. Following this logic, the adherence to clinical guidelines plays a major role in assuring that patients benefit from new evidence. Several studies have shown that adhering to guidelines in radiotherapy is essential for maintaining the quality and consistency of treatments. This is important in clinical trials, but also fundamental in a non-trial patient setting where the majority (approximately 95%) of treatments take place. What good are guidelines if we do not follow them? The positive impact of guidelines is enhanced by the number of centres subscribing to them; however, this also makes the corresponding evaluation of adherence more complicated. Nevertheless, with the recent advances in data collection automation and processing in radiotherapy, it is now possible to collect and analyse the actual treatment data of all patients in a multicentre setting, allowing for a comprehensive investigation into guideline adherence on

S11 ESTRO 2024 complete real-world data. This offers an alternative method to manual registration of treatment characteristics, which oftens show limitations in completeness and interpretability. In Denmark, the Danish Breast Cancer Group (DBCG), has conducted the DBCG RT Nation study, in which the full radiotherapy data for 7448 high-risk breast cancer patients was collected. This represents all treatments conducted in accordance with DBCG guidelines in the period 2008-2016 for this patient group on a national level. From this data, methods for evaluating guidelines adherence were developed and demonstrated. Guidelines included target definition, introduction of respiratory gating, and dose coverage of the internal mammary lymph nodes. The results were compared to the manual registration data from the DBCG database. In this talk, the interdependent connection between trials, guidelines and treatments will be discussed. The methods, results, and lesson learned from the DBCG RT Nation study will be presented and used to enable a discussion on how we should evaluate guideline adherence in the age of data science and automation. Invited Speaker

3290

Assigning causality for treatment-intent modelling

Wouter A.C. van Amsterdam

University Medical Center Utrecht, Data Science and Biostatistics, Utrecht, Netherlands

Abstract:

Statistics and machine learning inform us to expect when passively observing the world, but as health care professionals and researchers we typically aim to improve health care. Improving health care, either at the individual patient level or at a system level, requires knowing the effects of interventions or treatments, preferably tailored to the unique patient or situation. The best way to learn the effects of treatments is with randomized controlled trials (RCTs), but RCTs are expensive, slow, and sometimes unethical. Causal Inference formally defines what a treatment effect is and how it may be estimated inside and outside of RCTs. Causal inference thereby broadens the range of causal questions we can realistically answer and datasets we can use to do so. In this talk, I introduce a ‘language’ of causal inference, using the Potential Outcome framework. Next, I discuss how directed acyclic graphs (DAGs) help designing causal inference studies. I’ll provide a template of a typical causal inference study and will finish with challenges and opportunities of causal inference typical to the field of oncology.

3291

Personalized nodal CTV definition

Jan Unkelbach, Roman Ludwig, Yoel Perez Haas, Esmee Looman, Panagiotis Balermpas

University Hospital Zurich, Radiation Oncology, Zurich, Switzerland

Abstract:

The presentation focuses on defining the elective nodal clinical target volume (CTV-N) for Head & Neck squamous cell carcinoma (HNSCC) patients. Following current guidelines for definitive (chemo)radiotherapy of HNSCC, large parts

S12 ESTRO 2024 of the neck's lymph drainage region are electively irradiated due to the risk of occult lymph node metastases. Current guidelines are mainly based on the prevalence of lymph node involvement in each lymph node level (LNL) for a given primary tumor location. The presentation will report on efforts to reduce and personalize the definition of the CTV-N based on the individual patient's state of tumor progression at the time of diagnosis, including the primary tumor's T-stage, location, and lateralization, as well as the location of clinically detected lymph node metastases. Three main aspects of the problem will be discussed. 1) We built a large multi-institutional database of 2756 HNSCC patients to quantify patterns of lymph node involvement per LNL depending on primary tumor characteristics. The web-based platform LyProX.org was developed to make the data publicly available and to allow interactive exploration and visualization of the data. 2) Based on the data, a statistical model is developed to estimate the individual patient's probability of occult lymph node metastases in each clinically negative LNL. The statistical model is based on Hidden Markov Models (HMM). Each LNL is described by a random variable that indicates the true involvement of the LNL including occult metastases. Lymphatic metastatic tumor progression is described via the transition matrix of the HMM, which is in turn parameterized via a directed graph representing the lymphatic drainage. The model parameters learned from the data are the probabilities of the tumor to spread to a LNL and between LNLs. 3) A phase II clinical trial on personalized volume-deescalated elective nodal irradiation in oropharyngeal SCC is currently in preparation. The primary endpoint of the study is the rate of N-site recurrences in unirradiated LNLs. The CTV-N will be defined such that the estimated risk of occult lymph node involvement in all unirradiated LNLs combined is <10%. Main features of the trial are that a) ipsilateral level IV is not irradiated if level III is clinically negative, b) contralateral levels III/IV are not electively irradiated unless the upstream levels II/III are involved, c) unilateral irradiation is performed in patients with lateralized tumors and clinically negative contralateral neck, d) level I and V are irradiated in fewer patients compared to current guidelines. Invited Speaker

3293

Automatic delineation: How should it be used and possible dangers or drawbacks

Ditte Sloth Møller 1,2 , Anne Ivalu Sander Holm 1 , Lise Bech Jellesmark Thorsen 1

1 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark. 2 Aarhus University, Department of Clinical Medicine, Aarhus, Denmark

Abstract:

The use of high-quality AI segmentation algorithms results in a higher degree of agreement among individual delineators and better adherence to delineation guidelines. It also saves time for the individual delineator, freeing up time for other tasks in increasingly strained healthcare systems. However, such automation comes with new risks and potential drawbacks. Algorithms may perform worse than anticipated, especially if they are applied in patient groups or settings they were not developed for. With the introduction of AI segmentation in the daily clinic, a sense of false security could develop, resulting in inadequate quality and correction of AI delineations.

S13 ESTRO 2024 In the talk, we will discuss ways to implement AI auto-segmentation algorithms in large-scale clinical practice. Since the availability of AI algorithms is rapidly increasing, we will focus on flexible and adaptable systems to be able to implement new and updated algorithms quickly. We will discuss procedures for validation before the introduction of an algorithm and how to incorporate the results of the validation into the daily use of the specific algorithm, quality assurance and verification of the complete treatment chain (End-to-end test), and continuous monitoring of AI-segmented structures after clinical implementation. Continuous monitoring has multiple purposes. Detecting the lack of editing of the automated contours is just as important as detecting systematic errors because it elucidates changes in the use of the algorithms. During the presentation, we will share our experiences with the introduction of AI algorithms and how it has changed work procedures with a focus on how to catch errors in an AI-derived structure. Finally, we will discuss educational needs and methods to ensure the continued ability of staff to identify correct as well as incorrect performance of AI algorithms. Invited Speaker

3296

Auto-segmentation QA

Liesbeth Vandewinckele

KU Leuven, Laboratory of Experimental Radiotherapy, Leuven, Belgium

Abstract:

Auto-segmentation is the most important application of Artificial Intelligence (AI) in radiotherapy. It is currently offered by a lot of companies and being introduced in a lot of radiotherapy departments. However, the quality assurance (QA) of auto-segmentation models remains challenging since the applied techniques appear as black boxes to the user. In the following, different types of QA are explained and recommendations are given of how to deal with them in clinical practice. Routine model QA is QA of the auto-segmentation model itself by evaluating the model output after a change in the clinical workflow. These changes can for example be software updates or changes in imaging device (CT, MRI, ...), imaging protocol (patient positioning, field of view, fixation aids), ... The evaluation of the model should in these cases be performed by an end-to-end test on a reference dataset that reflects the current clinical practice and has not been used to create the model. The obtained automatic segmentations should be compared in both a quantitative and qualitative manner to manually obtained segmentations. Quantitative metrics used for autosegmentation purposes are for example the DICE similarity coefficient, the Hausdorff distance or volume metrics. A qualitative evaluation should be performed by a physician by comparing both segmentations side by side. When the routine QA tests fail, re-commissioning of the autosegmentation model is necessary. Case-specific QA refers to the QA of the auto-segmented contours of an individual patient. Manual supervision of the obtained automatic segmentations is at this moment the most important tool for case-specific QA. The segmentations can in this way be adapted to the user's preference. Next to manual verifications, methods exist to flag outlier patients/segmentations upfront to the manual check. A first method is to perform a similarity check to compare the new patient's data to the ones of the training set since the behaviour of the model in situations it has not seen before is unknown. A second method consists of a statistical model that can detect outliers by evaluating volume, centroid and structure shape of the auto-segmentations. A third method is to simultaneously use an independent, secondary automatic segmentation methods that can reveal outliers if differences are present in both auto-segmentations.

S14 ESTRO 2024 The use of automated QA for auto-segmentation models is at this moment not yet established in clinical practice. However, it could be of great importance during further expansion of adaptive radiotherapy since Invited Speaker

3298

Brachy/SBRT rationale

Piotr Wojcieszek

National Research Institute of Oncology, Gliwice Branch, Brachytherapy, Gliwice, Poland

Abstract:

Ultrahypofractionation is an unprecedented concept in the modern radiation oncologist's arsenal to cure cancer. The evolution of imaging, planning algorithms and precision of LINACs has brought us to the moment when we can ask ourselves if conventional means still 2Gy per fraction. Although classic radiotherapy (i.e., hiperfractionation, conventional, mild hypofractionation, and altered schedules) has saved many lives and has pushed our understanding up to today, we live now in the era of precisely delivered high fractions in short overall treatment time. Nevertheless, there is one method, also more than 100 years old, in which aficionados were fearless in using doses per fraction higher than those recognized as ultra-hypofractionated. This lecture aims to show where we are now with high-dose-rate (HDR) brachytherapy (interventional radiotherapy) and stereotactic body radiotherapy (SBRT). Such knowledge inspires an understanding of different radiotherapy techniques in clinical settings.

3299

Breast brachytherapy: Can we match EBRT hypofractionation?

Jean-Michel Hannoun-Levi

Antoine Lacassagne Cancer Center, Radiation Oncology, Nice, France

Abstract:

Regarding radiation therapy regimens, since the publication of the results of the “Boost versus no boost” EORTC phase 3 randomized trial, one of the main subjects related to adjuvant breast irradiation has been the investigation of new hypofractioned and accelerated regimens with the goal of shortening significantly the total treatment time. Moderate hypofractionation was first investigated in the START B prospective phase 3 randomized trial, which confirmed that 40 Gy in 15 fractions over 3 weeks was non-inferior in terms of local control compared to 50 Gy in 25 fractions over 5 weeks. IMPORT LOW trial was the first phase 3 randomized trial to show that APBI was possible in 3 weeks. More recently, FAST-Forward phase 3 randomized trial reported non-inferior local control rates after a moderate hypofractionated regimen (40 Gy/15 fractions/3 weeks) compared to accelerated-hypofractionated whole breast irradiation (26 Gy/5 fractions/1 week). In the framework of accelerated-hypofractionated adjuvant breast irradiation, and specifically for low-risk tumors, multicatheter interstitial brachytherapy (MIBT) is another validated option since the publication of the GEC-ESTRO phase 3 randomized trial results comparing whole versus accelerated and partial breast irradiation (APBI). In this trial, more than 80% of the APBI cohort were treated with high-dose rate