ESTRO 2020 Abstract book

S148 ESTRO 2020

provide guidance. The focus of the talk will be auto- segmentation algorithms aimed at defining the target. I will first touch on previous efforts for auto-delineation of the clinical and gross disease in HN before getting into more recent approaches that involve deep learning algorithms. A significant amount of work has also been performed in auto-delineation of organs at risk in the HN region and we will be briefly discussing some recent methods involving deep learning algorithms as well as workflows that aim at introducing them in clinical practice. SP-0272 The role of tumor growth models for CTV definition J. Unkelbach 1 , B. Pouymayou 1 , R. Ludwig 1 , M. Guckenberger 1 , P. Balermpas 1 1 universitätsspital Zürich, Radiation Oncology, Zürich, Switzerland Abstract text CTV definition is particularly complex because the CTV, which accounts for microscopic extensions of the tumor beyond the GTV, is per definition not visible in radiological images. While GTV delineation can be thought of as an image segmentation problem, in which an abnormal appearing tumor mass is delineated on radiological images, CTV delineation requires knowledge of the patterns of microscopic tumor progression. This talk will illustrate how mathematical models of tumor growth can contribute to the CTV delineation problem with the goal of automating current guidelines, improving consistency in delineation, and ultimately improving definition of the CTV. Two examples will be considered. The first example considers GTV-to-CTV extension for glioblastoma. Glioblastoma are known to infiltrate the normal appearing brain far beyond the GTV visible on MRI. However, microscopic tumor invasion is not isotropic but is influenced by anatomical barriers such as the falx and ventricles. In addition, tumor cells primarily spread within white matter and infiltrate grey matter to a lesser degree. Consequently, major sulci also represent barriers for microscopic tumor progression. Previously, phenomenological tumor growth models based on reaction-diffusion equations were developed to model these growth patterns. These models effectively define a geodesic distance from the GTV, i.e. a distance measure that accounts for anatomical barriers and anisotropy in tumor invasion [1]. In combination with automated segmentation of relevant brain anatomy, automated GTV- to-CTV expansion that is consistent with the complex neuroanatomy can be achieved. The second example investigates elective nodal CTV definition (CTV-N) for head & neck cancer. Head & neck cancers spread through the lymphatic system and form metastases in regional lymph nodes in the neck. In addition to a primary tumor CTV to account for tumor infiltration of adjacent tissues (CTV-T), the CTV-N includes electively irradiated lymph node levels that are at risk of harboring occult metastases despite the absence of macroscopic metastases visible on radiological images. We present recent work on the development of a statistical model of lymphatic tumor progression [2]. The model estimates the probability of microscopic involvement of lymph node levels given the individual patient's state of disease progression, i.e. the location of primary tumor and macroscopic lymph node metastases. The model is based on the methodology of Bayesian networks. The graph of the network represents the anatomically defined lymph drainage patterns; the parameters of the network, which

quantitatively describe the probabilities of tumor spread to and between lymph node levels, are learned from a dataset of lymphatic progression patterns in previously treated patients. The model may allow for further personalization of CTV definition in the future. In current practice, most of the lymph drainage region on both sides of the neck is irradiated for the majority of patients, at least for advanced stages. The model may identify lymph node levels with very low risk of involvement in individual patients, which may be excluded from the CTV in order to reduce side effects. [1] Unkelbach J, Menze B, Konukoglu E, Dittmann F, Le M, Ayache N, Shih H. Radiotherapy planning for glioblastoma based on a tumor growth model: improving target volume delineation. Phys. Med. Biol., 59(3):747-770, 2014 [2] Pouymayou B, Balermpas P, Riesterer O, Guckenberger M, Unkelbach J. A Bayesian network model of lymphatic tumor progression for personalized elective CTV definition in head and neck cancer. Phys. Med. Biol., 64:165003, 2019 SP-0273 How to get patients involved in RTT education A. Stewart-Lord 1 1 London South Bank University, School of Health and Social Care, London, United Kingdom Abstract text Introduction : The value of involving patients (service users) and carers in the education of health and social care professionals has long been recognised. Evidence suggests that patient involvement in health education helps create a workforce who appreciate the importance of patient perspectives and engagement, not only in education but also in the development and monitoring of healthcare delivery and quality improvement. Patient and carer involvement in Higher Education : Infrastructure is required to co-ordinate and celebrate the contribution people with lived experience of radiotherapy bring to RTT education. One of the most successful approaches is that of co-production. This approach enables students to build a skill set of person centredness, compassion, empathy and resilience. Co- production in RTT education is achieved at various levels throughout the educational journey. Starting with RTT curriculum and course design; followed by student recruitment; teaching, learning and assessment. Facilitators and Barriers: Universities have been challenged by the complexities associated with patient involvement in education, even though professional bodies and education regulators require the involvement of patients in health education there is limited guidance on how this should be achieved. Successful working relationships between the University and patient groups or service users are often associated with a clear engagement strategy allowing mutual benefit for all parties involved. This can be achieved by providing a single entity provision for engaging with people who have lived experience of health and social care. In RTT education the involvement of patients ensure that their preferences, needs and values guide clinical decisions, and that care is respectful of and responsive to their needs. During this working relationship patients also have access to training programmes on recruitment, communication, interview skills and research methods. Sessions can be delivered through online learning or face to face. The People’s Academy provides Symposium: Patient involvement in radiation oncology