ESTRO 38 Abstract book

S779 ESTRO 38

evaluation of HCC. The aim of this study was to evaluate the interobserver agreement in GTV of HCC in a multicenter panel and compare MRI and CT in GTV delineation. Material and Methods After the institutional review boards approved the study we analyzed anonymous, multiphasic-planning CT, and MRI obtained from five patients with HCC. Eight radiation oncologist in our center using CT and MRI to delineate GTVs of hepatocellular carcinomas. Liver locations were segments II, IV, V, VI and VIII. The cases show different difficulty: low acquisition contrast, previously treated with transarterial-chemoembolization, arterial thrombosis, or inhomogeneous liver parenchyma due to cirrhosis. In both CT and MRI, the GTV volumes and distance between the centers of mass (dCOM) were compared. The index of agreement was evaluated according to Landis and Koch. Results The GTV volume as defined on CT was in all cases larger or at least as large as the GTV volume on PET. The median GTV volume on MRI was 2 mL (0,4–85,6 mL), compared to 3,55mL on CT (0,2–64 mL) (p=0.09). Variance between observers in CT was minor that MRI (259,3 vs 332,4). The index of agreement was low (0,029) between CT and MRI in all patients. Conclusion In cases with well-defined tumor CT is more easy and reproductible. In cases with no defined tumor in CT we need other tools as MRI. The interobserver variability in target delineation of HCC in this study was noteworthy. EP-1434 Neoadjuvant chemoradiotherapy in patients with esopageal or esophageal gastric junction cancer N. Slim 1 , R. Tummineri 1 , P. Parise 2 , E. Mazza 3 , L. Albarello 4 , F. Puccetti 2 , A.M. Deli 1 , E. Incerti 5 , M. Azizi 1 , I. Dell'oca 6 , A. Cossu 2 , M. Reni 3 , R. Rosati 2 , P. Passoni 1 , N. Di Muzio 1 1 San Raffaele Scientific Institute- Milano- Italy, Radiation Oncology, Milan, Italy ; 2 San Raffaele Scientific Institute- Milano- Italy, Surgery, Milan, Italy ; 3 San Raffaele Scientific Institute- Milano- Italy, Clinical Oncology, Milan, Italy ; 4 San Raffaele Scientific Institute- Milano- Italy, Pathological Anatomy, Milan, Italy ; 5 San Raffaele Scientific Institute- Milano- Italy, Nuclear Medecine, Milan, Italy ; 6 San Raffaele Scientific Institute- Milano- Italy, Radiation Oncology, Milano, Italy Purpose or Objective to report our experience in patients (pts) with esophageal (EC) or esophageal gastric junction (EGJ) cancer treated with neoadjuvant chemoradiotherapy IG-IMRT PET based. Material and Methods from April 2014 to April 2018, 56 pts (m: 43, f: 13), with histologically proven EC or EGJ were treated according to CROSS study. Median age at diagnosis: 56 years (28-80), median KPS: 90 (80-100). Twenty-eight pts had adenocarcinoma (50%), 27 pts had squamous cell carcinoma (48.2%), 1 pt had adeno-squamous carcinoma (1.7%). All pts underwent c-e CT and PET simulation, repeated for restaging. Radiotherapy (RT) consisted in 41.4 Gy in 23 fractions combined to chemotherapy (ChT) with carboplatin and paclitaxel. Results clinical stage was: T1: 1pt (2%), T2: 14 pts (25%), T3. 38 pts (68%), T4: 3 pts (5%), N0: 11 pts (20%), N+: 45 pts (80%). The site of tumor was proximal/middle third in 4 pts (8%), middle/middle-distal third in 34 pts (60%), distal third/distal-EGJ/ EGJ in 18 pts (32%). Median tumor length was 4 cm (0.8-15 cm). RT was delivered by Tomotherapy in 37 pts (66%) and by VMAT in 19 pts (34%). All pts completed RT. Median cycles of ChT was 5 (2-6 cycles), 70% pts received a full dose of ChT. G3 acute haemathological toxicity was: neutropenia in 3.5 % (2pts), lymphopenia in 67.8% (38 pts), anemia in 1.7% (1pt). G3

included. To improve the comparability of the different Re-I regimens and to assess the relationship between RT dose and toxicity, when not reported in the study, equivalent dose in 2 Gy fractions was calculated according to the linear-quadratic model. Results Nine studies met the inclusion criteria, published between 2002 and 2017. Only 2 were prospective trials. Overall, 203 patients were definitely re-irradiated within the abdomen. Patients presented recurrence of pancreatic (100 patients), liver (38 patients) and gastro-esophageal (14 patients) cancer. Median follow-up from Re-I ranged from 5.9 to 28 months. Previous RT was delivered with a median dose of 50.4 Gy (45 to 74.5 Gy) using conventional fractionation; in 1 study a stereotactic RT (SBRT) boost was delivered after conventional external beam RT. The mean time relapsed since previous RT ranged from 2 to 32 months (median time=18 months). Technique used for Re- I were SBRT (5 studies), intensity modulated RT (1 study), 125 I seed interstitial brachytherapy (1 study) and proton therapy (2 studies). Re-I prescription doses were variable (22.5 Gy in 3 fractions to 126,5 Gy with brachytherapy with 125 I). Regarding to toxicities, 29 episodes of G3 toxicity were reported: 15 episodes were acute (transient RILD, abdominal pain and bleeding, anorexia and fatigue), whereas 14 were late (RILD, bowel obstruction, gastric perforation). Relationship between cumulative dose and toxicities rate (≥G3) is reported in Table 1. The 1‐year overall survival rate was reported in 8 studies and resulted as 48.2% (95%CI:38.1–61%). The 1‐year local recurrence free survival rate obtained from 7 studies was 70.7% (95% CI: 62.8–79.7%). Improvements in pain was evaluated in 5 studies, with 72.8% of patients experiencing pain relief.

Conclusion Few studies, most of them retrospective, are actually available in the literature able to reach our requisites. Based on the results of our analysis, abdominal re- irradiation seems to be safe feasible in terms of severe toxicities with good local control and symptoms palliation. Prospective studies or large data collections seem to be necessary to define patients selection criteria to ensure maximal benefit of Re-I treatment approach. EP-1433 GTV contouring in hepatocellular carcinoma: a comparison between two imaging techniques C. De la Pinta Alonso 1 , J.D. García 2 , D. Sevillano 2 , R. García 3 , M. Martín Martín 1 , M. Martín Sánchez 1 , T. Muñóz 1 , E. Fernández-Lizarbe 1 , R. Hernánz 1 , J.A. Domínguez 1 , C. Vallejo 1 , S. Sancho 1 1 Hospital Ramon y Cajal, Radiation Oncology, Madrid, Spain ; 2 Hospital Ramon y Cajal, Medical Physics, Madrid, Spain ; 3 Hospital Ramon y Cajal, Radiology Department, Madrid, Spain Purpose or Objective Definition of gross tumor volume (GTV) in hepatocellular carcinoma (HCC) requires dedicated imaging in multiple contrast medium phases. MRI has an important role in