ESTRO 38 Abstract book

S256 ESTRO 38

hormonotherapy was administered to 91 pts (44.4%). With a median follow-up of 38 months (179 pts presented a minimum follow-up of 36 months), acute and late grade ≥2 pt+ was observed in 95 pts (46.3%) and 60 pts (29.3%), respectively. No correlation was observed between values of RILA and acute pt+. A decreased incidence of late grade ≥2 pt+ was observed for increasing values of RILA (p = 0.04). The area under the ROC curve was 0.60. For cut-off values of RILA ≥24% and <14%, sensitivity and specificity were 75% and 37%, 45% and 74%, respectively. Negative predictive value for grade ≥2 pt+ was equal to 76% for RILA ≥24% and positive predictive value was equal to 45% for RILA <14% where the overall prevalence of grade ≥2 pt+ was estimated at 29%. A significant increase in the risk of grade ≥2 pt+ was found if patients had pre-treatment urinary symptoms (sHR = 1.90, p = 0.016) and decreased if pts presented a RILA ≥14% (sHR = 0.58, p = 0.04). Conclusion RILA significantly predicts the risk of pt+ in case of prostate radiotherapy associated with pre-treatment clinical symptoms. This study validates the use of RILA in a predictive multiparametric nomogram as a rapid screening test before RT delivery and may guide physicians in personalizing radiation therapy scheme. OC-0499 Neutrophilia as prognostic factor for outcome in the CAO/ARO/AIO-04 phase 3 rectal cancer trial M. Diefenhardt 1 , R. Hofheinz 2 , T. Beissbarth 3 , D. Arnold 4 , J. Müller von den Grün 1 , T. Liersch 5 , P. Ströbel 6 , G. Grabenbauer 7 , R. Fietkau 8 , J. Weitz 9 , M. Ghadimi 5 , C. Rödel 1 , E. Fokas 1 1 Klinikum der Johann Wolfgang Goethe Univ, Department of Radiotherapy and Oncology, Frankfurt, Germany ; 2 University Hospital Mannheim, Department of Medical Oncology, Mannheim, Germany ; 3 University Medical Center Göttingen, Department of Statistics-, Göttingen, Germany ; 4 Helios Klinikum Hamburg, Department of Medical Oncology, Hamburg, Germany ; 5 University Medical Center Göttingen, Department of General- Visceral and Pediatric Surgery, Göttingen, Germany ; 6 University Medical Center Göttingen, Institute of Pathology, Göttingen, Germany ; 7 DiaCura & Klinikum Coburg, Department of Radiation Oncology and Radiotherapy, Coburg, Germany ; 8 University of Erlangen-Nürnberg, Department of Radiation Therapy, Erlangen, Germany ; 9 University of Dresden, Department of General and Visceral and Pediatric Surgery, Dresden, Germany Purpose or Objective Peripheral blood leukocytosis and neutrophilia reflect cancer inflammation and have been proposed as prognostic immunological biomarkers in various malignancies. However, previous studies were limited by their retrospective nature and small patient numbers. Material and Methods The CAO/ARO/AIO-04 randomized phase 3 trial accrued 1236 patients with locally-advanced rectal adenocarcinoma. Patients were randomized to receive either fluorouracil-based preoperative chemoradiotherapy alone (5-FU CRT) or with oxaliplatin (5-FU/Oxaliplatin CRT) followed by surgery and adjuvant chemotherapy.The baseline level of peripheral blood leukocytes, neutrophils, hemoglobin, platelets, lactate dehydrogenase (LDH) and carcinoembryonic antigen (CEA) was correlated with clinicopathologic characteristics and clinical outcome. Multivariable analyses adjusted for treatment arm, pathologic stage, resection status and blood biomarkers were performed using Cox regression

only the clinical factors was larger outside than inside the surgical area. The data showed better discrimination for predicting resistance than sensitivity to developing late reaction. In univariate analysis of the upper RILA tertile versus the lower and middle tertiles, the positive predictive value (PPV) for resistance to fibrosis was 95.1% outside and 73.2% within the surgical area whereas the negative predictive values (NPV) were 21.0% and 16.0%, respectively. For resistance to telangiectasia, PPV was 97.5% and NPV was 12.7%. Conclusion CD4 + RILA predicted fibrosis and telangiectasia although it seemed better at predicting resistance. The higher PPV for resistance to fibrosis outside the surgical area supports that CD4 + RILA is indeed associated with risk of radiation- induced fibrosis whereas fibrosis within the surgical area could be partly explained by clinical factors. Although the patients in this study were recruited prospectively and had a long follow-up (>10y), the predictive modeling is limited by blood samples being taken at the time of follow-up rather than prospectively. Future prospective studies are needed to validate the present findings. OC-0498 Results of the prospective trial evaluating radiation-induced lymphocyte apoptosis and prostate RT D. Azria 1 , G. Crehange 2 , F. Castan 3 , E. Schwartz 4 , Y. Belkacemi 5 , J. Lagrange 6 , T. NGuyen 7 , O. Chapet 8 , F. Mornex 9 , G. Noel 10 , E. Lartigau 11 , D. Pasquier 11 , S. Clippe 12 , C. Hennequin 13 , S. Gourgou 14 , M. Brengues 15 , P. Fenoglietto 4 , M. Farcy-Jacquet 16 , E. Ozsahin 17 1 ICM-Val d'Aurelle, Radiation Oncology and Radiobiology Unit, Montpellier, France ; 2 Centre GF Leclerc, Radiation Oncology, Dijon, France ; 3 ICM Val d'Aurelle, Biometry Unit, Montpellier, France ; 4 ICM-Val d'Aurelle, Radiation Oncology, Montpellier, France ; 5 AP-HP Henri Mondor, Radiation Oncology, Creteil, France ; 6 AP-HP Henri Mondor, Radiation Oncology, Créteil, France; 7 Institut Jean Godinot, Radiation Oncology, Reims, France; 8 Centre Hospitalier Lyon Sud, Radiation Oncology, Pierre Bénite, France ; 9 Centre Hospitalier Lyon Sud, Radiation Oncology, Pierre Benite, France ; 10 Centre Paul Strauss, Radiation Oncology, Strasbourg, France ; 11 Centre Oscar Lambret, Radiation Oncology, Lille, France; 12 Centre Marie Curie, Radiation Oncology, Valence, France ; 13 AP- HP Saint Louis, Radiation Oncology, Paris, France ; 14 ICM- Val d'Aurelle, Biometry Unit, Montpellier, France ; 15 IRCM ICM-Val d'Aurelle, Radiobiology unit, Montpellier, France; 16 CHU Nimes, Radiation Oncology, Nimes, France ; 17 CHUV, Radiation Oncology, Lausanne, Switzerland Purpose or Objective Monocentric cohorts suggested that radiation-induced CD8 T-lymphocyte apoptosis (RILA) can predict late toxicity after curative intent radiotherapy (RT). We assessed the role of RILA as a predictor of pelvic toxicities (pt+) after localized prostate RT in a prospective multicentre trial. Material and Methods A total of 383 prostate-cancer patients (pts) treated by RT were recruited at eight centres. Before the amendment (2014), only favourable stage prostate cancers were included (n=205). The amendment allowed inclusions of prostate cancers treated with pelvic irradiation for higher risk disease or postoperative RT (n=179). RILA was assessed before RT by flow cytometry. Impact of RILA on pt+ (primary endpoint) or relapse was assessed using a competing risks method. Receiver-operator characteristic (ROC) curve analyses were also performed in intention to treat. This study is registered with ClinicalTrials.gov, number NCT00893035 and analyses of the 205 localized prostate-cancer pts are presented here. Results All pts received 80 Gy to the planning target volume defined as the prostate only plus a margin of 10 mm except in regard to the rectum (5 mm). Concomitant

models. Results

After a median follow-up of 50 months, neutrophilia remained an independent adverse prognostic factor for disease-free survival (DFS; HR 1.528; 95% CI 1.181-1.977;