Abstract Book

S78

ESTRO 37

OC-0154 REQUITE Big Data Resource for Validating Predictive Models and Biomarkers of Radiotherapy Toxicity C. West 1 , R. Elliott 1 , C. Talbot 2 , A. Webb 2 , P. Seibold 3 , D. Azria 4 , D. De Ruysscher 5 , R. Symonds 6 , L. Veldeman 7 , B. Rosenstein 8 , P. Lambin 5 , T. Burr 9 , S. Gutiérrez Enríquez 10 , T. Rancati 11 , A. Vega 12 , J. Chang-Claude 3 1 The University of Manchester, Division of Cancer Sciences, Manchester, United Kingdom 2 University of Leicester, Department of Genetics, Leicester, United Kingdom 3 German Cancer Research Center DKFZ, Division of Cancer Epidemiology, Heidelberg, Germany 4 Montpellier Cancer Instiute, Department of Radiation Oncology, Montpellier, France 5 Maastricht University Medical centre, Department of Radiation Oncology, Maastricht, The Netherlands 6 University of Leicester, Department of Cancer Studies, Leicester, United Kingdom 7 Gent University Hospital, Radiation Oncology Department, Ghent, Belgium 8 Icahn School of Medicine at Mount Sinai-, Department of Radiation Oncology, New York, USA 9 University of Nottingham, Source Bioscience, Nottingham, United Kingdom 10 Vall d’Hebron Institute of Oncology-VHIO, Oncogenetics Group, Barcelona, Spain 11 Fondazione IRCCS Istituto Nazionale dei Tumori, Prostate Cancer Program, Milan, Italy 12 Fundacion Publica Galega de Medicina Xenomica- SERGAS, Grupo de Medicina Xenomica-USC, Santiago de Compostela, Spain Purpose or Objective Many models and biomarkers were reported to have potential to predict a cancer patients’ risk of toxicity following radiotherapy (RT), but the challenge is to validate them for clinical application. Validation requires access to well-annotated big datasets. The European Union funded FP7 REQUITE project was established with the aim of carrying out a prospective, longitudinal, multi- centre study to compile a large centralised, standardised dataset and biorepository for validating models and biomarkers that predict a cancer patient’s risk of RT toxicity. Material and Methods Case record forms were produced to collect standardised epidemiology, treatment, side effect and quality-of-life data from patients recruited in 12 main centres in Belgium, France, Germany, Italy, Spain, The Netherlands, UK and USA. Questionnaires for collecting patient reported Common Toxicity Criteria for Adverse Events were translated into multiple languages and validated for use. Data are collected both prior to RT and post-RT and at years 1 and 2. A centralised biorepository stores DNA and PAXgene tubes. The radiation induced lymphocyte assay (RILA) was carried out in three centres (France, Germany, UK). Results 4,440 patients were recruited prospectively and an additional 383 lung cases identified and included to bring the total in the resource to 4,823 (91% of the planned 5,300). Prospective recruitment over the time possible during the funding period was highest for breast (2,071 of 2,100 planned) then prostate (1,810 of 2,100 planned) and lung (559 of 1,100 planned). The quality controlled centralised database for electronic data capture and storage contains a considerable amount of data (>80,000 completed case record forms) and includes dose volume histograms for 4,125 (93% of total) and DICOM files for 4,170 (94% of total) patients. Breast photos taken at baseline and two years to score change in appearance are also stored: 2,039 photos at baseline (98%) and currently 704 photos at two years. RILA data were collected for 1,322 patients. A centralised biobank repository contains

DNA from 4,450 patients and 3,128 PAXgene tubes, which are available for future studies. Infinium OncoArray-500K SNP genotyping is underway for all samples, and the genomics data will be available for export via a data portal. A process was established to provide access to the resource via submission of a concept form, and a public discovery platform (Café Variome) will be available in 2018 enabling researchers to query the REQUITE database. Conclusion REQUITE has developed a standardised approach for collecting and curating RT data linked with a biorepository. The outputs of REQUITE are already serving as resource for REQUITE partners and the wider radiobiology community. OC-0155 The effect of accelerated partial breast irradiation on quality of life. D.H.M. Jacobs 1 , N. Horeweg 1 , M. Straver 2 , E.M.A. Roeloffzen 3 , G. Speijer 4 , J. Merkus 5 , J. Van der Sijp 2 , M.E. Mast 6 , U. Fisscher 6 , A.L. Petoukhova 6 , A.G. Zwanenburg 3 , C.A.M. Marijnen 1 , P.C.M. Koper 6 1 Leiden University Medical Center LUMC, Radiotherapy, Leiden, The Netherlands 2 Haaglanden Medical Center, Surgery, The Hague, The Netherlands 3 Isala, Radiotherapy, Zwolle, The Netherlands 4 Haga Hospital, Radiotherapy, The Hague, The Netherlands 5 Haga Hospital, Surgery, The Hague, The Netherlands 6 Haaglanden Medical Center, Radiotherapy, The Hague, The Netherlands Purpose or Objective To investigate health related quality of life (HRQL) in elderly breast cancer patients after two types of Accelerated Partial Breast Irradiation: intraoperative radiotherapy (IORT) and external beam APBI (EB-APBI). Material and Methods Between 2011 and 2016 women ≥60 years with breast tumours of ≤30 mm undergoing breast conserving therapy were included in a prospective multi-centre cohort study. Patients were treated with electron IORT (1x23.3 Gy prescribed at 100% isodose) (n=268) or photon EB-APBI (10x3.85 Gy daily approximately 4 weeks after surgery) (n=207). All patients received EORTC QLQ C30 and BR23 questionnaires before surgery (baseline) and at several time points up till 1 year, and were eligible for this analysis if a baseline and at least one of the follow up questionnaires were available. A linear mixed model was used to analyse HRQL across the first year after treatment between groups, with time points baseline, 3, 6 and 12 months. Additionally, EORTC scales were compared cross- sectional postoperative, post-treatment and at 1 year. The post-treatment time point was postoperative for IORT (mean 2 weeks after surgery) and after radiotherapy for EB-APBI patients (mean 7 weeks after surgery). Treatment effect and predictors for global health status (GHS) at 1 year were analysed using multivariable analysis. A p-value of ≤0.01 was deemed significant for repeated measures and multiple testing. For EORTC scales, differences of ≥10 points between groups were considered clinically relevant. Results For this analysis 203 IORT and 154 EB-APBI patients were included. Linear mixed model analyses demonstrated that emotional functioning and future perspective differed significant between groups, but not over time (Figure 1). All other scales showed no difference between groups. Proffered Papers: CL 3: Breast