Abstract Book

S339

ESTRO 37

disqualify the use of MRI in RT, not even as a replacement for CT. However, it is important to understand these properties and how to ensure that the MR images have sufficient quality to be used as planning material for dose delivery. The aim of this talk is to help you gaining an understanding of what to look out for when using MR in RT and what you can do minimize risks. The image quality in general is obviously very important considering that the target definitions and OAR delineations heavily depend on it and inconsistent image quality will lead to uncertainties. Detecting subtle changes with the MRI equipment is therefore crucial and implementation of QA procedures to detect these changes will be the first topic discussed. One major concern with the use of MR images in RT is the geometric accuracy. There are several sources of geometric distortion; including susceptibility effects, gradient non-linearity, concomitant fields and water-fat shift. How to minimize and QA the distortions, when they may be significant and what impact they have on dose calculation will be discussed. Using an MR-only workflow requires ED maps to be generated from MR images. This is a complex procedure typically involving models learned from paired MR and CT data or non-rigid registration of an MR image to a CT atlas. Errors can enter the ED maps in several ways that sometimes are difficult to anticipate. This will be discussed in the context of an example and possible methods to detect issues with MR based ED maps will also part of this final topic. SP-0640 Photon dosimetry protocols: what is new? H. Palmans 1 1 Palmans Hugo, Medical Physics, Teddington, United Kingdom Abstract text The development of Codes of Practice or Protocols for clinical reference dosimetry over the last five decades has in general been a slow process in which major steps of improvement have always been in response to new technical advancements of measurement equipment, computing infrastructure or radiotherapy delivery. A historic overview of the main improvements in reference dosimetry for external photon beams will be given in this lecture and new recommendations published in the past decade will be discussed in more detail as well as current activities and future initiatives. From the recommendations published in the last five years the Addendum to AAPM TG-51 [1] was the first one to use Monte Carlo calculated beam quality correction factors based on simulation of the entire detector geometry, bypassing the common approach of factorising the effect of various perturbations that had been commonly used before that. It was also the first Protocol to make specific recommendations for flattening-filter- free (FFF) photon beams in response to the commercialisation of new treatment machines such as TomoTherapy and CyberKnife and the development of FFF beams on existing machine types by other linac vendors. The last two decades has seen an enormous growth in the use of complex radiotherapy for which dosimetric concepts used for broad-field photon beams have been put into question. These include the definition of field size, reference conditions, beam quality and the use of ionization chambers for field output factors. It has, for example, been shown that water-to-air stopping power ratios at the basis of ionization chamber dosimetry do not Teaching Lecture: Photon dosimetry protocols: what is new?

Teaching Lecture: State of the art in treatment of oesophageal cancer

SP-0638 State of the art in treatment of oesophageal cancer F. Cellini 1 1 Università Cattolica del Sacro Cuore, Radiation Oncology - Gemelli ART, Rome, Italy Abstract text Esophageal cancer is one of the most unknown and deadliest cancers worldwide, mainly because of its extremely aggressive nature and poor survival rate. Esophageal cancer is the 6 th leading cause of death from cancer and the 8 th most common cancer in the world. The 5-year survival is around 15%-25%. Although some International guidelines agreed on the preferable treatment option for the different presentations, still some controversies are pending, and other issues are arisen by the most recent evidences from literature. Thus, sometime in the clinical practice surgeons, medical oncologist and radiation oncologist struggle to select the most efficient treatment schedule and best timing to be offered to their patients. The differences in the therapeutic management of locally advanced and unresectable esophageal cancer will be outlined. Moreover the issue of dealing with gastroesophageal junction lesions and adenocarcinomas will be addressed. Finally, the role of modern technology opportunities in the modern radiotherapy era and in the next future, will be addressed. SP-0639 MRI based radiotherapy – what can go wrong and how to QA? Garpebring 1 1 University Umea Norrlands Universitetssjukhus, Department of Radiation Sciences, Umea, Sweden Abstract text In radiotherapy (RT) one tries to optimize the tumor control probability for a given acceptable level of normal tissue complication probability. Typically, a small change in the delivered dose at this optimum can have a quite large effect on the outcome. It is therefore very important to ensure that the tumor will be given the exact prescribed dose and in this process one must take uncertainties in tumor position and extent as well as dose delivery into account. A key tool here is the imaging that is used for target and organs at risk (OAR) localization, patient positioning and dose calculation. Traditionally computed tomography (CT) is the work horse here, in a large extent due to its excellent geometric accuracy and simple relation between the Hounsfield units and the tissue electron density (ED). However, in the last decade, interest in including magnetic resonance imaging (MRI) in the RT workflow has increased, either as an additional modality or as a replacement for CT. The reason is the superior soft tissue contrast of MRI and the availability to use functional imaging techniques. Or put in other words, MRI offers superior images for delineating the tumors and the OARs and is hence a tool for reducing uncertainties. Unfortunately, MRI is not quantitative (absolute intensity values have no meaning), does not have the same geometric accuracy as CT and there is no simple relationship between the intensities in an MR image and the ED of the tissue. These properties of MRI does not Teaching Lecture: MRI based radiotherapy – What can go wrong and how to QA?