ESTRO 2020 Abstract Book

S926 ESTRO 2020

increase of mean ADC-levels from week 2 with 934·10 -6 mm 2 /s (513-1310 mm 2 /s) to week 7 with 976·10 -6 mm 2 /s (661-1376 mm 2 /s). NeoADT and IGRT led to a 1.4-fold increase of the initial ADC-level. Conclusion In this prospective study we were able to demonstrate feasibility of an ambitious imaging protocol with MR-based adaptive IGRT. DWI measurements indicated an increase of mean ADC-levels during neoADT and IGRT with the most prominent increase during RT. Whether DWI could become a surrogate biomarker predictive for response will be evaluated using clinical factors and long-term follow-up. PO-1683 A Novel Integration of RT Dose and SPECT- MPI: A Segment Model of Left Ventricle for Breast Cancer H. Lin 1,2 , K. Lin 1,2 , C. Yu 1,2 , C. Hsu 1,2 , S. Wang 1,3 , P. Shueng 2,4 , T. Wu 1 1 National Yang-Ming University, Department of Biomedical Imaging and Radiological Sciences, Taipei City, Taiwan ; 2 Far Eastern Memorial Hospital, Department of Radiation Oncology, New Taipei City, Taiwan ; 3 Far Eastern Memorial Hospital, Department of Nuclear Medicine Center, New Taipei City, Taiwan ; 4 National Yang-Ming University, Department of Medicine, Taipei City, Taiwan Purpose or Objective An increased risk of heart disease after radiotherapy is well established in breast cancer. However, we would like to further evaluate radiation doses of the heart or different segments of the left ventricle in the era of precision medicine. This study aims to establish a novel integration and visualization of cardiac radiation dose and myocardial function using a segment model of left ventricle for breast cancer. Material and Methods We established a phantom and simulate computed tomography (CT) scans for the radiotherapy of breast cancer. We performed single photon emission computed tomography - myocardial perfusion imaging (SPECT-MPI) scans and inject Tl-201 into the myocardium of the phantom to simulate cardiac perfusion defect images. We performed image acquisition, fusion and registration of simulation CT and SPECT images in the radiation therapy planning (RTP). The cardiac structure including left ventricle, 4 rings, 20 segments according to AHA 20 segment model were delineated. We simulated and import a RTP for left breast cancer. The dosimetry of left ventricle, 4 rings, 20 segments and myocardial function were analyzed. We established a novel integration and visualization of cardiac radiation dose and myocardial function using a segment model of left ventricle for breast This study successfully integrated the simulation CT and MPI images in the radiation treatment planning system, and discrete subvolumes of the left ventricle defined by the AHA 20-segment model. In the dosimetric analysis, apical cap received the highest mean dose (2.45 Gy) in 4 rings and 20 segment received the highest mean dose (3.49 Gy) in 20 subvolumes, followed by 19 segment (3.13 Gy), 17 segment (2.66 Gy), 16 segment (2.50 Gy), and 18 segment (1.87 Gy) (Table 1). We integrated and visualized cardiac radiation dose and myocardial function using a segment model of left ventricle for breast cancer. Table 1 Subvolumes of the left ventricle dose and MPI cancer. Results

score.

Conclusion This study successfully provides a novel interface to evaluate cardiac RT doses and SPECT-MPI of patients receiving radiotherapy. PO-1684 Serial MRI assessment of salivary glands in patients with Oropharyngeal SCC receiving radical RT C. Paterson 1 , O. Kemp 2 , L. Devlin 3 , P. McLoone 1 , A. Duffton 3 1 Beatson WoSCC, Clinical Oncology, Glasgow, United Kingdom ; 2 University of Glasgow, Faculty of Medicine, Glasgow, United Kingdom ; 3 Beatson WoSCC, Therapy Radiography, Glasgow, United Kingdom Purpose or Objective Identifying changes in salivary gland characteristics as patients undergo radical RT may allow prediction of xerostomia. The aim of this study was to examine the feasibility of obtaining volumetric and functional measurements in the major salivary glands (SG) using anatomical and DW MRI during H&N RT. Material and Methods Patients with intermediate-high risk OPSCC participating in the MeRInO study were included. MRI scans (T1W post contrast FS and DW-MRI) were carried out prior to fractions 1 and 11 of RT on a GE Signa 1.5T HDxt (GE, Crawley, UK). Bilateral parotid (PG), submandibular (SM) and sublingual (SL) SGs were delineated on axial slices in Eclipse V15.5 (Varian medical systems Palo Alto). Anatomical images were used to define structures prior to transfer onto DW b=0 sequence. Adjustments were made to contours, accounting for visual assessment of the structure as well as movement and artefact, before transfer to ADC map. Volumes from anatomical and functional sequences and ADC mean for each SG were measured at each time point. Results 31 patients (124 scans) were evaluated. Bilateral SM and SL delineations were successful on all 124 scans. 8 PGs (either right or left) were not fully included on baseline scans (3 on T1PCFS and 5 on ADC map) and 15 PGS were incomplete on scans in week 3 (8 on T1PCFS and 7 on ADC map) as the field of view was prioritised for gross tumour volume in line with the primary objective of the MeRInO study. Salivary gland characteristics are shown in table 1