ESTRO 37 Abstract book

ESTRO 37

S398

1 Proton Medical Research Center and Tsukuba university, Radiation oncology, Tsukuba, Japan Purpose or Objective To investigate the clinical outcome of concurrent chemoradiotherapy (CCRT) using photon and proton beams for locally advanced pancreatic cancer (LAPC). Material and Methods A total of 75 patients of LAPC who were treated in our institute using CCRT between November 2005 and March 2016 was retrospectively reviewed. This study included 41 men and 34 women, whose age ranged from 36 to 84 years (median 65 years). 39 patients were treated by chemotherapy (0.2 - 24.8 months, median 3.0 months) before coming to our institute. At the time of CCRT clinical stage of IB/IIA/IIB/III was 1/3/7/64 according to UICC TNM staging system, 7th edition (0/1/2/22 in photon group, 1/2/5/42 in proton group). Of all patients, 25 patients were enrolled in using photon beams (photon group) : (50.0-50.4 Gy with 25-28 fractions, median 50.0 Gy with 25 fractions) and 50 patients in using proton beams (proton group) : (45.0-60.0 GyE / 25-33 fraction: 1.8 - 2.0 GyE / fraction in 29 with median 54.0GyE / 27 fractions, and 67.5 GyE / 25 fractions: 2.7 GyE / fraction in 21 cases). For chemotherapy, gemcitabine was used in 66, S-1 in 8, gemcitabine with S-1 in 1 patients. Hyperthermia was concurrently performed in 39 patients (12 in photon group, 27 in proton group). We examined overall survival (OS) and local control (LC) from the start of CCRT, and severe toxicity rates were compared between photon and proton groups. Results A total of 72 patients completed the treatment as planed and 3 patients were discontinued due to the newly developed liver metastases during the treatment course (all in the proton: 3). Follow-up period was 1.3-37.8 (median: 12.8) months. For photon group and proton group, the 1/2-year OS rates were 67.1/11% and 72.4/52.4%, and the median OS was 15.7 and 25.6 months (p=0.06, respectively), the 1/2 years LC rates were 49.2/0% and 79.5/66.1%, and median LC was 10.6 and more than 36 months (p=0.003, respectively). Regarding to the prognostic factor, total irradiation dose was the only significant in both of univariate and multivariate analyses in OS (p=0.002 and 0.016, respectively). In LC, type of radiotherapy and total irradiation dose were significant in univariate analysis (p=0.003, <0.001, respectively), however, total irradiation dose was only significant in multivariate analyses (p=0.024, respectively). Chemotherapy and hyperthermia was not significant prognostic factor both OS and LC. Acute adverse events of grade 3 (CTCAE, version 4.0) were observed in 5 patients in the photon group (leukopenia: 4, bleeding from tumor: 1) and 8 patients in the proton group (leukopenia: 8). Late adverse events of grade 3 were observed in 1 patient in the photon group (duodenal stenosis), and 1 patient in the proton group (gastric ulcer). No patients more than grade 4 was not observed. Conclusion OS of CCRT using proton beams was satisfactory. High dose irradiation in CCRT can improve LC and OS in LAPC. PO-0771 Fractionated SBRT for adrenal metastasis: contributing to local tumor control with modest toxicity S. Knippen 1 , K. Burjakow 1 , F. Putz 1 , A. Knippen 1 , N. Achterberg 1 , S. Lettmaier 1 , R. Fietkau 1 1 University Clinic Erlangen, Radiation Oncology, Erlangen, Germany Purpose or Objective To report on the Erlangen experience with LINAC SBRT for adrenal metastasis from various primary tumors that were treated from 2003 to 2015.

Material and Methods 33 pts were treated. Primary sites included lung (n = 19), melanoma (n=8), colorectal (n=2), hepatocellular (n=1), esophageal (n=2) and breast cancer (n=1). 14 patients were treated to avoid local tumor related symptoms, 19 patients were treated with local curative intent. Radiation treatment Treatment planning was done based on an exhale and mid ventilation CT. Further planning CTs were done on fx 1 and 5 on a regular basis to check for the correctness of the breathing pattern. Irradiation was performed using a NOVALIS linear accelerator. From 2012 on, isocenter was verified before each treatment session using the Varian ExacTrac® system to minimize setup errors. Dose was prescribed to target volume (ITV) surrounding 90% isodose. Follow-up Depending on their overall performance status and prognosis, patients received clinical check-ups and radiological imaging. Median follow-up was 11 months. Statistical analysis IBM SPSS v. 24 was used for univariate analysis (log-rank test) using Kaplan-Meier curves, non-parametric Kruskal- Wallis test, and the Chi-square test for frequency distributions. Toxicity was graded according to NCI CTCAE v4.0. Depending on radiologic imaging, patients were classified as stable disease (progression in the RT volume), regression (if volume did show some shrinkage) and progression. Results Median survival was 11 mos, median PFS was 5 mos. Median local failure free survival was 21 mos. 16 pts lived more than 12 mos and local control after 1 year was 56.3%. 10 pts lived more than 24 months and in these, local control was still 50%. Pts that were treated in a curative intent did show a significant better survival curve (p<0.0001, see fig. 1) and PFS (p=0.004) than those treated in palliative intention. BED was calculated (a/b 10). Min. BED was 42Gy, max. BED was 108.8Gy and median BED was 67.2Gy.Three groups were formed (up to 50.7Gy BED, 12.1%, 56 – 72.8Gy, 72.7%, 73.2Gy up to 108.8Gy, 15.2%). We found significant differences between the overall survival curves (p=0.046, see fig. 2), favoring the high dose group. In PFS there was a trend (p=0.066) to improved survival. 28 pts had documented treatment toxicity, 21 pts. were free from any adverse events or discomfort. In 7 cases a toxicity grade I was noted, there was no case of documented adrenal dysfunction during follow up. There was no reported long term toxicity.