Abstract Book

S86

ESTRO 37

to tumors in the liver even for large tumors. This study analyzed outcome after SBRT for CCC. Material and Methods Sixty-four patients with 82 lesions of inoperable CCC from a retrospective multicenter analysis of practice and outcome of the working group ‘Stereotactic Radiotherapy’ of the German Society for Radiation Oncology were analyzed. They were treated with SBRT between 1999 and September 2016 and data were entered into a centralized database. Prior to SBRT, the median tumor size and planning target volume were 4.4 cm and 114 cc, range 1.0 to 18 cm and 5 to 1876 cc, respectively. Twenty patients (24%) had chemotherapy before SBRT. Available parameters were analysed for local control (LC), overall survival (OS) and toxicity. Results Median follow-up time for patients alive was 35 months (range 7-91 months). Median overall survival (OS) time was 15 months; 2-year and 3-year OS rates were 32% and 21%. Prescribed biological effective radiation dose (BED) was median of 67.2 Gy (range 36-115 Gy; standard deviation 20 Gy) in median 8 fractions (3 – 17; 95% CI: 3- 12) and radiation dose was the only prognostic factor for local tumor control and OS. Of note, treatment volume was neither predictive nor prognostic. Higher doses correlated with better local control (LC) rate and OS. Patients receiving BED max greater than 91 Gy had a median OS time of 24 months versus 13 months for those receiving lower doses (p=0.015). Local control rates at 12 and 24 months were 91% and 80% after BED max greater than 91 Gy versus 66% and 39% (p=0.005) after lower doses. Tolerance of the treatment was good with 13% of grade 1 and no higher grade gastroduodenitis.

resection, transplantation, ablation, radioembolization, stereotactic body radiation therapy (SBRT), single-agent chemotherapy, and multi-agent chemotherapy. Sites were dichotomized at the median as using <= four vs >= five of the listed treatment modalities. Univariate and multivariate analyses were performed using Cox regression analysis to determine factors associated with improved OS. Sensitivity analyses were performed to determine the effect of changing the cut point used in dichotomization and to determine the effect of removing patients who were treated with less frequently used modalities (defined as any modality used in < 10% of patients from the entire cohort). Results There were a total of 117,777 patients w ith non- metastatic HCC included in the analysis. Of a tot al of 1230 sites, 830 (67.5 %) used four or fewer modalities. Most sites used single-agent chemotherapy, surgical resection and multi-agent chemotherapy. The distribution of treatment modalities by site modality number is shown in the table below. Average survival for patients treated at centers using four or fewer modalities was 11.6 months and 22.8 months for those treated at centers with greater than 4 modalities (HR = 0.52; p < 0.001). After adjusting for center volume, tumor size, single vs multiple tumors, invasion into a major vessel, T stage, age, gender, year of diagnosis, race, insurance status, academic vs community center, location, Charlson-Deyo comorbidity score, income, education, level of fibrosis, AFP, creatinine, bilirubin, and INR, treatment at a multi-modality center still provided a survival advantage (continuous HR = 0.83, p < 0.001; dichotomous HR = 0.69, p < 0.001). Sensitivity analysis varying the cut point from one to six showed that the HRs did not vary much (ranging from 0.88 (p <0.001) for a cut off of six to 0.65 (p < 0.001) for a cut off of two). Less frequently used treatments were multi-agent chemotherapy (9.5%), ablation (8.2%), transplant (7.8%), radioembolization (1.6%), and SBRT (0.7%). Excluding patients who received these treatments showed that treatment at a multi-modality center continued to provide a survival advantage (adjusted continuous HR = 0.85, p < 0.001; adjusted dichotomized HR = 0.66, p < 0.001). Conclusion Institutions that offered increasing number of modalities for the treatment of HCC were associated with improvement in OS for patients with non-metastatic HCC. This remained true after controlling for center volume and in sensitivity analysis after removing patients who received the less commonly used treatment modalities. OC-0168 Patterns of recurrence in the CRITICS gastric cancer phase III trial M. Verheij 1 , R. Van Amelsfoort 1 , K. Sikorska 2 , E. Jansen 1 , A. Cats 3 , N. Van Grieken 4 , H. Boot 3 , P. Lind 5 , E. Meershoek-Klein Kranenbarg 6 , M. Nordsmark 7 , H. Hartgrink 6 , H. Putter 6 , A. Trip 1 , J. Van Sandick 8 , H. Van Tinteren 2 , Y. Claassen 6 , J. Braak 6 , C. Van de Velde 6 1 Netherlands Cancer Institute, Radiation Oncology, Amsterdam, The Netherlands 2 Netherlands Cancer Institute, Biostatistics, Amsterdam, The Netherlands 3 Netherlands Cancer Institute, Gastrointestinal Oncology, Amsterdam, The Netherlands 4 VU University Medical Center, Pathologu, Amsterdam, The Netherlands 5 Karolinska Institutet, Oncology and Pathology, Stockholm, Sweden 6 Leiden University Medical Center, Surgical Oncology, Leiden, The Netherlands 7 Aarhus University Hospital, Oncology, Aarhus, Denmark 8 Netherlands Cancer Institute, Surgical Oncology, Amsterdam, The Netherlands

Conclusion This is the largest reported series on SBRT in cholangiocarcinoma. Overall survival and local control were improved after higher doses (BED) and tolerance was excellent. OC-0167 Multi-modality centers associated with improved survival for patients with hepatocellular carcinoma J. Jiang 1 , N. Ohri 1 , A. Kaubisch 1 , M. Kinkhabwala 1 , C. Guha 1 , R. Kabarriti 1 1 Montefiore Medical Center/Albert Einstein College of Medicine, Radiation Oncology, Bronx, USA Purpose or Objective To assess if patients treated at centers with increasing number of treatment modalities for HCC have improved This is a retrospective analysis of data from the Commission on Cancer’s National Cancer Database (NCDB) from 2004-2014 on patients with non-metastatic HCC. The treatment modalities assessed were surgical overall survival (OS). Material and Methods