ESTRO 36 Abstract Book

S587 ESTRO 36 _______________________________________________________________________________________________

RT history (n=2). We retrospectively reviewed records of 80 patients about clinical stage, pathologic characteristics, performing surgery or chemotherapy, aim of radiotherapy, radiation dose and technique, and clinical outcomes such as local recurrence, overall survival according to radiation groups. We analyzed prognostic factors of adjuvant RT such as stage, extent of surgery, resection margin, radiation dose, and chemotherapy. We also reviewed treatment related complication using CTCAE criteria version 4.0. All analyses were performed using SPSS, version 22. Results Median age at diagnosis was 51 years (21-82 years). Most common sites were salivary glands (n=35, 43.8%), oral cavity (n=14, 17.5%), and paranasal sinuses (n=12, 15.0%). Half of patients (n=41, 51.3%) had a locally advanced tumor at diagnosis (T3 : n=11, 13.8%), T4 : n=30, 37.5%). Sixty-nine patients underwent surgery. Detailed patient's characteristics according to RT aim were in table. Sixty- nine (86.2%) patients underwent adjuvant radiotherapy and 11 patients (13.8%) underwent definitive radiotherapy. Radiation dose were 50.4 – 76 Gy per 24-42 fx (median 64.8 Gy). With median follow-up of 114.3 months (9.7 – 236.3 months), local tumor progression was found in 21 patients (26.3%). 5 year overall surviva (OS)l rate was 82.4-91.4% in adjuvant arm and 72.7% in definitive arm. 5 year local recurrence free survival (LRFS) rate was 74.1-97.1 % in adjuvant arm, and 48.5% in definitive arm. Survival curves following treatment arms and stage were in graph. All patients tolerated the radiotherapy well. Conclusion Adjuvant radiotherapy to head and neck ACC seemed better clinical outcomes compared with definitive radiotherapy. However, in this report, all patients who received definitive radiation therapy were advanced stage (stage III :1, stage IV :10). Considering stage, 72.7% of 5YOS rate and 48.5% of 5YLRFS rate in definitive radiotherapy arm were not inferior results. Therefore definitive radiotherapy may be considered alternative treatment modality in patients with inoperable adenoid cystic carcinoma of head and neck. EP-1068 Hypoxic imaging obtained at 2-h postinjection in FMISO-PET M. Kawamura 1 , M. Yoshimura 1 , T. Katagiri 1 , T. Mitsuyoshi 1 , H. Inokuchi 2 , T. Ishimori 3 , Y. Nakamoto 3 , T. Mizowaki 1 , M. Hiraoka 2 1 Kyoto University- Graduate School of Medicine, Radiation Oncology and Image-Applied Therapy, Kyoto, Japan 2 Japanese Red Cross Wakayama Medical Center, Radiation Oncology, Wakayama, Japan 3 Kyoto University- Graduate School of Medicine, Diagnostic Imaging and Nuclear Medicine, Kyoto, Japan Purpose or Objective Positron emission tomography (PET) / computed tomography (CT) using 18F-fluoromisonidazole (FMISO) has been used as an imaging tool for tumor hypoxia. It has been reported that several quantitative values in FMISO- PET at 4-h postinjection were reproducible. However, it is controversial whether they are reproducible when scanning is performed in earlier time, e.g. 2-h postinjection. If quantitative values at 2-h postinjection are equivalent with those at 4-h postinjection, the total examination time in FMISO-PET can be shortened. The purpose of this study was to investigate the difference of quantitative values in FMISO-PET at 2-h and 4-h postinjection in patients with head and neck cancer. Material and Methods A total of 10 patients with untreated locally-advanced head and neck cancer who underwent FMISO-PET/CT scan

from August 2015 to October 2016 in our institute were analyzed. Image acquisition was performed twice, 2-h and 4-h after administration of FMISO using a combined PET/CT scanner (Discovery IQ, GE Healthcare). After taking a region of interest in a primary tumor, the maximum standardized uptake value (SUVmax), SUVmean, SUVpeak, tumor-to-blood ratio (TBR), tumor-to-muscle ratio (TMR), metabolic tumor volume (MTV) and total lesion hypoxia (TLH) were measured. We evaluated the Spearman's rank correlation coefficients of these quantitative values, and also calculated the percent difference defined as difference between the two values divided by average of two values. Results SUVmax (mean±SD) at 2-h and at 4-h were 2.4±0.8, and 2.6±1.0, respectively, in this population. The Spearman's rank correlation coefficients of SUVmax, SUVmean, SUVpeak, TBR, TMR, MTV, and TLH were 0.97, 0.97, 0.97, 0.93, 0.93, 0.95, and 0.95, respectively. The percent differences (mean±SD) of these values were 8.3±5.6%, 6.8±6.3%, 4.1±5.1%, 16.8±12.5%, 17.1±9.1%, 42.7±52.7%, and 41.7±58.1%, respectively. Quantitative values were highly correlated between the two scans; however, there were two cases in which %differences of MTV and TLH exceeded 50%, i.e. 62.1% and 166.7% in MTV and 57.1% and 180.0% in TLH.

Conclusion Our preliminary data demonstrate that quantitative values at 2-h and 4-h were highly reproducible. In a few patients, volumetric parameters had moderate percent difference,