ESTRO 37 Abstract book

ESTRO 37

S404

Purpose or Objective An accurate diagnosis followed by multimodality therapy is essential to achieve long-term survival patients with pancreatic ductal adenocarcinoma (PDAC). We have proposed a model to select better patients for neoadjuvant treatments. Material and Methods From 2008 through 2017 84 patients (F:41; M:43) with borderline resectable or unresectable PDAC were accrued in three consecutive prospective studies of radiochemotherapy (RCT) with or without induction chemotherapy (IC). In all cases an accurate pre- treatment staging including CT scan, EUS, FDG-PET/CT and laparoscopy with peritoneal washing was performed. After IC and approximately 4 weeks after the completion of RCT, a re-evaluation was performed regarding tumour response and respectability with clinical examination, laboratory test, tumor markers, CT scan, FDG-PET/CT scan and laparoscopy. Results The median age of patients was 68 years (range 36–75 years). According to the results of the pre-treatment workup, twenty patients (23.8%) had metastatic disease and were therefore excluded from the protocols. Patients who underwent IC (Gem-Ox or FOLFIRINOX schemes) were evaluated for clinical response after 2 months by using CT scan and PET-CT scan. Nine patients experienced disease progression, three patients refused to continue the protocols. Fifty-two patients received concomitant RCT and were evaluated for clinical response. The median follow-up for all patients was 25 months (range, 6.1 to 81 months). Twenty-four patients (46%) underwent surgical radical resection, with negative margins. No patient died due to perioperative complications. For the whole group, the median PFS was 18.1 months. Two-year and three- year LC were 79% and 62%, respectively. Median OS was 16.2 months. One-year OS, two-year OS and three-year OS were 74%, 35% and 26%, respectively. Patients who underwent resection had a significantly longer median OS compared with non resected patients (37.6 months vs 13 months, p = 0.02). The median PFS for resected patients was 33 months compared with 11.2 months for non resected patients (p = 0.005). The median PFS for patients treated by upfront RCT was 10 months compared with 20 months for patients treated by IC followed by RCT (p = 0.001). Conclusion Radiochemotherapy confirms his role in the management of borderline resectable and locally advanced pancreatic cancer. Our diagnostic workup, including CT scan, FDG- PET/CT scan and laparoscopy could be a model useful to clinicians who treat pancreatic cancer to improve patients' selection and clinical outcomes. PO-0781 A nomogram for predicting survival of gastric cancer patients received resection and chemoradiation M.L. Zhou 1,2 , R. Hu 1,2 , W. Yang 1,2 , Y. Wang 1,2 , S. Zhang 3 , X. Yang 3,4 , W.J. Deng 1,2 , G.C. Li 1,2 , X. Wang 3 , Z. Zhang 1,2 1 Fudan University Shanghai Cancer Center, Department of Radiation Oncology, Shanghai, China 2 Shanghai Medical College- Fudan University, Department of Oncology, Shanghai, China 3 Cancer Center- West China Hospital- Sichuan University, Department of Abdominal Oncology, Chengdu, China 4 The First Hospital of Ziyang- Sichuan Province, Department of Oncology, Ziyang, China Purpose or Objective The aim of the current study was to find clinicopathologic variables associated with overall survival (OS) for patients with gastric cancer (GC) receiving postoperative chemoradiation after curative resection, and create a nomogram for individual risk prediction.

Material and Methods The clinicopathological and survival data of 667 patients between July 2004 and October 2015 treated at Fudan University Shanghai Cancer Center (FUSCC) and West China Hospital of Sichuan University (WCHSCU) were reviewed. Among patients in the FUSCC cohort (n=488), 90% of them were randomly assigned into the training set (n=439) and the other were assigned to the internal validation set (n=49). Patients in the WCHSCU cohort (n=179) were assigned to the external validation set. Multivariate analysis using the Cox proportional hazard regression model was performed. Discrimination and calibration of the nomogram were tested by c-index and calibration plots. Results Using the backward selection with Akaike information criteria (AIC), sex, age, tumor site, tumor grade, depth of invasion, number of positive lymph nodes and number of negative lymph nodes were selected as factors predictive of OS, and incorporated in the construction of the nomogram. C-index for the training set and internal validation set were 0.677 (95% CI: 0.671-0.684) and 0.655 (95% CI: 0.606-0.704), respectively. C-index for the external validation set was 0.664 (95% CI: 0.600-0.727). Conclusion Independent predictors of death following radical surgery and postoperative chemoradiation for primary GC were used to create a nomogram to predict OS. The nomogram was able to stratify patients into prognostic groups, and performed stably in internal and external validations. PO-0782 Borderline pancreatic adenocarcinomas: is there a place for a neoadjuvant treatment ? M. Terlizzi 1 , C. Pouypoudat 1 , E. Buscail 2 , S. Adgié 1 , N. Leduc 1 , E. Terrebonne 3 , D. Smith 3 , J.F. Blanc 3 , C. Cassinotto 4 , C. Laurent 2 , L. Chiche 2 , G. Belleannée 5 , R. Trouette 1 , V. Vendrely 1 1 CHU Bordeaux, Radiotherapy, Bordeaux, France Purpose or Objective To assess the R0 resection rate, overall and disease-free survivals of patients (pts) with borderline (BR) pancreatic cancer treated by upfront surgery (US) or neoadjuvant treatment followed by surgery (NTS). Material and Methods This retrospective analysis of a prospective observational cohort included all consecutive pts treated by surgery from 2010 to 2016 for non-metastatic BR pancreatic adenocarcinomas. Treatment strategy consisted in US or NTS in case of R1 resection risk according to the multidisciplinary tumor board decision. The primary endpoint was the R0 resection rate, secondary endpoints were overall and disease-free survivals. Kaplan-Meier method was used to estimate OS and DFS. Differences between groups were evaluated using Fisher’s exact test for categorical variables. Results Fifty-five pts (male: 26, 47.3 %, female: 29, 52.7%; median-age 64.5, range 47-79 years) with BR tumors according to NCCN 2017 classification were included. Thirty-four patients were treated by NTS consisting in induction chemotherapy (CT) with FOLFIRINOX regimen for 28 pts (82.3%) followed by concomitant chemoradiotherapy (mean dose: 56Gy, range 50-59Gy) and surgery for 30 pts (88.2%). Twenty-one pts had US. There was a significant difference in pre-treatment tumor characteristics with 35 % of NTS pts presenting with mesenteric arterial abutment versus 5% for US pts (p=0.01). Furthermore, a mesenteric veinous abutment was present for 91 % of NTS pts versus 57 % of US pts 2 CHU Bordeaux, Surgery, Bordeaux, France 3 CHU Bordeaux, Oncology, Bordeaux, France 4 CHU Bordeaux, Radiology, Bordeaux, France 5 CHU Bordeaux, Pathology, Bordeaux, France