ESTRO 38 Abstract book

S12 ESTRO 38

[3] Cunha M, Monini C, Testa E and Beuve M. NanOx, a new model to predict cell survival in the context of particle therapy. Phys Med Biol 62, 1248-68 (2017).

molecular tissue response, and on the signalling pathways. Studies indicate a differential DNA damage response (DDR) following high LET irradiation, with the double strand breaks DSB preferentially repaired by homologous recombination (HR), leading to an increased level of unrepaired damage, which is reflected in the RBE. This change in DDR may also lead to a differential response in cellular processes downstream of the DNA repair. SP-0037 Implementation of nanodosimetric based radiobiological models in treatment planning systems F. Villegas 1 1 Uppsala University, Immunology- Genetics and Pathology, Uppsala, Swedeny Abstract text Modelling for prediction of relative biological effectiveness (RBE) is a fast-growing research topic within hadron therapy. Carbon ion treatment centers have been at the forefront in the use of RBE models in treatment planning systems (TPS). However, the current exponential expansion of proton therapy seen in the last decades implies questioning of the convention of applying a constant RBE of 1.1 in clinical practice. Several planning studies have shown that proton treatments considering the expected biological outcome per tissue type and particle energy could reduce the side effect risk and increase the probability for tumour control. Unfortunately, the prediction accuracy of the radiobiological models used for optimization has been limited as correlations between simple macroscopic characteristics of the radiation (e.g. physical dose or linear energy transfer) and biological endpoints (e.g. cell survival) are not univocal. A key to strengthen the prediction power is to use nanoscopic properties of the radiation which characterize the textures of the radiation´s energy deposition discrete events at the scale of biomolecules (e.g. DNA) directly causing the biological damage. Advancements in experimental and computational nanodosimetry have pushed forward the search for an operational and measurable nanodosimetric quantity capable of RBE prediction. The frequency of cluster size is a promising property upon which published phenomenological models have been based; although the cluster scoring is still a matter of debate whether it be inside a predefined volume [1] or be free from volume [2]. Multiscale energy deposition distributions have also been used for building mechanistic models such as the NanOx model [3] which elegantly incorporates the chemical stages of DNA damage. Regardless of the model, issues concerning benchmarking and uncertainty assessments of Monte Carlo simulations involved need to be addressed first. Validated computational tools can then simulate clinical beams to fully characterize them down to a nanometer level. Such dosimetric charts are the key to bypass the need for LET mapping used by existing TPS optimization engines. In parallel, another set of challenges need to be undertaken to produce a consistent and reproducible biological experimental database to aid the fine-tuning of both mechanistic and/or phenomenological models. Indeed, the fundamental parts towards an RBE optimized treatment planning are already in place thus the biggest challenge is to connect them through standardization processes which will not only make data transferable between facilities but also will help gain the trust of clinical community. [1] Selva A, Conte V and Colautti P. A Monte Carlo tool for multi-target nanodosimetry. Radiat Prot Dosim 180, 182- 6 (2018). [2] Villegas F, Bäckström G, Tilly N and Ahnesjö A. Energy deposition clustering as a functional radiation quality descriptor for modelling relative biological effectiveness. Med Phys 43, 6322 (2016).

Symposium: Quality in an IGRT

SP-0038 Continuous Quality Improvement Strategies to Support Volumetric IGRT W. Li 1 1 Princess Margaret Cancer Center, Department of Radiation Oncology, Toronto, Canada Abstract text Technological advances in the field of Radiation Therapy enables online image guidance of patient setup, increasing the precision and accuracy of treatment delivery. Maintaining sustained excellence in an IGRT environment require various continuous quality improvement strategies ranging from education and staff training to development of departmental protocols, procedures, and processes. This presentation will focus on supporting RTT education and training at a large urban cancer centre in Ontario, from initial training strategies and current annual refresher approach (team based and eLearning module approach), to development and implementation of RTT led IGRT rounds, where challenging clinical cases are shared and discussed. SP-0039 Auditits in IGRT - Development of standardized image guidance registration documents and workflows 1 H.De Boer 1 University Medical Center Utrecht, Department of Radiation Oncology, Utrecht, The Netherlands Image guided radiotherapy (IGRT) is essential in modern radiotherapy. The primary function of image guidance is to confirm tumour positioning to accurately deliver the planned dose to the PTV. Here we present evidence of the further usefulness of these routine images in the calculation of delivered dose to the organs at risk (OARs). Due to daily organ deformation and anatomical changes, the delivered dose to OARs can be different to planned dose. However, provided that imaging protocol thresholds are met, this is generally not accounted for. Deviations from planned dose induce differences to the normal tissue complication probability (NTCP) since toxicity relates to dose actually delivered to the OAR. The hypothesis of the CRUK VoxTox research programme (UK CRN ID 13716) is that delivered dose can be a better predictor of toxicity than planned dose. Tracking motion- inclusive daily delivered dose allows us to build a more accurate picture of the total accumulated dose received by the OAR. By constructing NTCP models based on delivered dose, the ultimate aim is to identify patients on- treatment who are at increased risk of toxicity, for selectively targeted adaptive radiotherapy. Material and Methods Daily MVCT scans were acquired for patients undergoing TomoTherapy for prostate cancer (n = 528), head & neck (H&N) cancers (n = 319), and cancers of the CNS (n = 26). Here we focus on the former two cohorts. Automation was vital for large scale processing of over 26,000 MVCT scans, and multidisciplinary expertise spanned engineering, Abstract not received SP-0040 Exploiting IGRT to calculate delivered dose for normal tissue sparing L. Shelley 1 1 University of Cambridge, Department of Engineering- JDG, Cambridge, United Kingdom Abstract text Introduction