ESTRO 38 Abstract book

S784 ESTRO 38

(p=0.039), and chemotherapy after PBT (p=0.003) were significant factor for favorable survival. The MST for curative and palliative group was 25 and 7 months, respectively. No severe (grade 3) acute toxicities were observed. As for late toxicity, 3 patients experienced grade 3 biliary tract infection, but the causal relationship with irradiation was unclear. Conclusion The PBT could deliver high doses to the tumor safely, and may contribute to improvement of local tumor control and survival for patients with unresectable ICC. EP-1445 Gastroduodenal toxicity in patients having bile duct brachytherapy for perihilar cholangiocarcinoma N. Mcdonnell 1 , C. Deufel 1 , K. Merrell 1 , M. Neben Wittich 1 , T. Whitaker 1 , W. Harmsen 2 , K. Fruth 2 , M. Haddock 1 , C. Hallemeier 1 1 Mayo Clinic, Radiation Oncology, Rochester, USA ; 2 Mayo Clinic, Biomedical Statistics and Informatics, Rochester, USA Purpose or Objective For patients with localized, unresectable perihilar cholangiocarcinoma (pCCA), an approach of neoadjuvant chemoradiotherapy (CRT), bile duct brachytherapy (BT), and orthotopic liver transplantation has shown favourable survival. This study aims to determine if there is a correlation between BT gastroduodenal (GD) dose and GD This was a single institution retrospective review of all patients with localized, unresectable pCCA who received CRT and biliary high dose rate (HDR) BT as planned neoadjuvant therapy before liver transplant. Patients received external beam RT (EBRT, planned dose of 45 Gy in 30 fx, twice per day over 3 weeks) with concurrent 5- fluorouracil. Between 0 and 4 weeks after completion of EBRT, HDR BT (Iridium-192 remote afterloading system) was administered through 1-2 nasobiliary or percutaneous transhepatic catheter(s). CT-based BT treatment planning was utilized, with the target volume typically including a 1 cm radial expansion from the centre of the involved perihilar bile ducts. The duodenum/stomach was contoured and dose was converted to biologically equivalent dose (BED) using alpha/beta=3. Treatment was 1-4 fx administered over 1-2 days. Following completion of BT, patients received maintenance capecitabine until liver transplant. GD AEs were graded using NCI-CTCAE, v5.0. Cumulative incidence of grade 2+ and 3+ GD AEs was determined from date of BT, with death a competing risk. Univariable regression models were generated to examine for potential correlation between patient, disease, and treatment variables and GD AEs. Results The study included 120 patients treated from 2009-2018. Median age was 54 years (range 24-73) and 73% were male. 61% had primary sclerosing cholangitis and 45% had a history inflammatory bowel disease (IBD). Delivered EBRT dose was 45 Gy in 95% (range, 42-50 Gy). HDR BT schedule was 9.3 Gy/1 fx (74%), 16 Gy/4 fx (15%), 12.4 Gy/2 fx (6%), or other (5%). The median GD D0.1cc was 4.27 Gy and the median BED 3 was 12.67 Gy. Median follow up after BT was 14 months (range 1-99 months). Grade 2+ and 3+ GD AEs occurred in 38% (n=46) and 28% (n=34). Grade 3+ GD AEs were haemorrhage (n=17), stricture (n=11), ulcer (n=5), gastritis/duodenitis (n=4), and perforation (n=1). Median time from BT to first grade 3+ AE was 5.8 months (range 0.1-64). At 12 months, the cumulative incidence of G2+ and G3+ GD AEs were 33% (95% CI: 24-42%) and 23% (95% CI: 14-31%), respectively. On univariable analysis, there was no association between age, gender, PSC, history of IBD, or EBRT dose and risk of grade 2+ or 3+ GD AEs (all P>0.05). BT GD D0.1cc (minimum dose to the 0.1 cc adverse events (AEs). Material and Methods

– 0.4; range, 0.1 – 0.5 mm; p=0.37) in all 35 SBRT session among all patients. Inter-fractional motion measurements were based on more than 25,000 registries for each patient. Pancreatic movement was scored in the left-right (X), superior- inferior (Y), and anterior-posterior (Z) axes in real time using Calypso´s transponders tracking mode. There was a major pancreatic movement in the Y-axis: 10+/-1.2mm, compared to the X-axis: 4+/-0.4mm and Z-axis: 5+/- 0.3mm.Mean intra-fractional motion was also found to be larger for the Y-axis: 5+/-3.6mm than X-axis: 3+/-2.3 mm and Z-axis: 4+/-3.1mm. Conclusion Our results suggest that percutaneous US-guided implantation is feasible with limited side effects or risk of migration. Significant motion of the tumor was reported during and between treatments. Craneal-caudal movement (Y-axis) ranged 10mm representing a challenge for adequate tumor coverage. Standardized PTV margins for pancreatic SBRT should take this movement into consideration. Real-time tumor tracking may allow for a more accurate delivery of radiation therapy in pancreatic cancer. EP-1444 Clinical results of proton beam therapy for unresectable intrahepatic cholangiocarcinoma S. Shimizu 1 , T. Okumura 1 , N. Mizoguchi 1 , H. Numajiri 1 , K. Murofushi 1 , K. Onishi 1 , Y. Oshiro 1 , M. Mizumoto 1 , T. Nonaka 1 , H. Ishikawa 1 , H. Sakurai 1 1 Proton Beam Therapy Center-University of Tsukuba Hospital, radiation oncology, Tsukuba city, Japan Purpose or Objective Surgery is considered the only curative treatment for intrahepatic cholangiocarcinoma (ICC). However, most patients with ICC present with unresectable disease at the time of diagnosis, only approximately 30% of the patients with ICC have operable disease. The standard of therapy for unresectable ICC is chemotherapy, but the median survival time (MST) is less than 1 year. Recent progress of radiotherapy has made it possible to deliver high doses to the tumor, and some studies have suggested efficacy of radiotherapy for unresectable ICC. Proton beam therapy (PBT) can deliver much higher doses to the tumor while keeping dose constraint to surrounding normal tissues. In this study we evaluated the clinical outcome and prognostic factors of PBT for patients with unresectable ICC. Material and Methods From October 2001 to March 2017, 37 patients with unresectable ICC (Stage I, II, IVa, IVb: 4, 4, 19, 10) were treated with PBT. The reason of unresectable disease were medically inoperable (old age, poor performance status (PS)), tumor progression, and both for 5, 22, and 10 patients, respectively. The patients group included 22 male and 15 female, and the median age was 68 years old (range: 32-87 years old). Twelve, 19, and 6 patients had PS of 0, 1, and 2, respectively. Child-Pugh score was A for 27 patients, and B for 10 patients. Ten patients had multiple tumors, and 10patients had distant metastases. Jaundice was observed in 11 patients. Patients were classified into curative group (25 patients) and palliative group (12 patients), depending on whether the planning target volume covered all the macroscopic tumors or not. The treatment dose was 66Gy in 10 fractions, 72.6 GyE in 22 fractions, and 74.0 Gy in 37 fractions for 1, 21, and 5 patients, respectively. Sixteen patients received concurrent chemotherapy. Results Median follow-up time for patients alive at time of analysis was 37.5 months. The MST for 37 patients was 15 months, and overall survival time was 60%, 41%, 31% at 1, 2, and 3 years, respectively. One- and 2- year local control rates were 94% and 69%, respectively. According to multivariate analysis, curative treatment (p=0.008), lack of jaundice