ESTRO 2020 Abstract book

S413 ESTRO 2020

over the total patient group. The clinical relevance of these differences is unknown. Applying an individualised strategy could be beneficial. If large peak-to-peak amplitudes in the Z-direction are observed in the free breathing scan we advise to examine if a BH technique could reduce CTV-PTV margins.

including doses and volumes, after which it is essential to apply modern radiation therapy techniques on anatomically defined target volumes. To facilitate this, validation and education of the target volume guidelines will be continued, as well as the development of facilitating tools including atlas-based and deep-learning- based autocontouring software.

Award Lecture: K. Breur Award Lecture

Award Lecture: Donal Hollywood Award

SP-0678 Eternal mammary nodes P. Poortmans 1 1 Paris Sciences et Lettres Université, Marie Curie Professor, Herentals, Belgium Abstract text Breast cancer management progressed tremendously over more than a century thanks to the efforts and dedication of a large number of giants in oncology. Nevertheless, many new questions continue to arise in this eternally evolving field, including several for the radiation oncologist. Many of those are related to the interactions between the multitude of patient-, tumour- and treatment-related parameters. Target volumes for radiation therapy were traditionally defined using conceptual bony and soft tissue landmarks like, for example in EORTC protocol 22922 for the internal mammary nodes: “The lateral border lies at the middle of the clavicula. The upper border is 3 cm above the head of the clavicula. The medial border lies in general 1 cm across the midline but has also to be defined clinically at the supraclavicular region by the radiotherapist on the patient and not only during fluoroscopy on the simulator. The lateral field border of the IM part of the field lies 5 cm lateral from the midline. The lower field border is defined by the lower border of the 4th rib but depends also on the location of the primary tumour and on the specific anatomy of the individual patient. A block is shielding the larynx if needed. After breast conserving surgery, the lateral field edge of the IM field is matched to the tangential fields irradiating the breast tissue.” Since 2015, consensus-based ESTRO guidelines aim to radically replace the old “same field for everybody” approach to anatomy- based individualised target volumes based on the knowledge that the lymphatic vessels (and thereby nodes) follow the veins: “This volume includes the lymph nodes alongside the internal thoracic veins, which are always positioned medially to the corresponding arteries. On the right side, the internal thoracic vein drains into the brachiocephalic vein, while the internal thoracic artery originates from the subclavian artery, with up to 1–2 cm distance in cranio-caudal direction between these vessels dorsal to the clavicular head. On the left side the internal thoracic vessels are connected to the subclavian artery and the brachiocephalic vein with less distance in- between. In the most cranial part, where only the artery is present, a margin of 5 mm is added to the artery.” While this anatomy-based approach was broadly endorsed and subsequently validated in several series investigating the sites of regional recurrences, unfortunately most patients continue to be treated using the field- based treatment set-up that is derived from the 2D simulator-based practises from the 80 ties and 90 ties from the 20 th century, just polished-up with a CT-scan to perform fancy-looking dosimetric calculations. The only way to further improve radiation therapy for our future breast cancer patients is to select for each individual patient the right indications and prescriptions

OC-0679 On-line MRI-based proton beam range verification: first experimental proof-of-concept S. Schellhammer 1,2 , S. Gantz 1,2 , L. Karsch 1,2 , E. Van der Kraaij 3 , J. Smeets 3 , A. Serra 4 , J. Pawelke 1,2 , A. Hoffmann 1,2,5 1 OncoRay - National Center for Radiation Research in Oncology, Medical Radiation Physics Section, Dresden, Germany ; 2 Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiooncology - OncoRay, Dresden, Germany ; 3 Ion Beam Applications SA, Research & Development, Louvain-la-Neuve, Belgium ; 4 ASG Superconductors, Paramed MRI Unit, Genua, Italy ; 5 Faculty of Medicine and University Hospital Carl Gustav Carus at the Technische Universität Dresden, Department of Radiotherapy and Radiation Oncology, Dresden, Germany Purpose or Objective Previous studies have shown the feasibility of using MRI to determine the proton beam end of range by radiation- induced biological effects occurring in-vivo using an off- line retrospective approach. The successful integration of MRI and proton therapy that has recently been achieved with in-beam MRI at a fixed proton research beam line opened up the perspective to investigate the potential of in-beam MRI to capture proton beam signatures on-line during dose delivery. This study aimed to investigate the feasibility to verify the proton beam range in phantom materials with on-line MRI. Material and Methods A cylindrical plastic bottle filled with deionised water was placed in the magnetic isocentre of a 0.22 T in-beam MR scanner (MrJ2200, ASG Superconductors) and irradiated with a centrally impinging proton beam of different energies (190 MeV, 200 MeV, 210 MeV, 225 MeV) and different beam currents (1 nA, 3 nA, 9 nA, and 27 nA, corresponding to dose rates of 1.7 Gy/s, 5 Gy/s, 15 Gy/s, and 45 Gy/s, respectively). A 17 cm thickness block of PMMA was placed 45 mm in front of the phantom as range modulator. During irradiation, 5 mm thickness single-slice MR images were acquired in a coronal plane through the centre of the beam using T 2 *-weighted gradient echo (GE) and inversion recovery gradient echo (IRGE) pulse sequences with respective acquisition times of 18 s and 35 s. Immediately after irradiation, three consecutive MR images were acquired to assess transient effects due to the beam turn-off. To test the effect of material composition and viscosity, the experiments were repeated with other fluids (ethanol, petroleum), highly viscose materials (sugar syrup, mayonnaise, gelatine) and solid tissue-mimicking material (pork chop). Results For both the GE and IRGE sequences, beam current- and energy-dependent signatures were observed in the MR images of the fluid-filled phantom only (Figure 1). The GE sequence produced a hyperintense, vertical, central line artefact which, upon irradiation, exhibited a wobble at the expected range. The IRGE sequence produced a