ESTRO 2020 Abstract book

S672 ESTRO 2020

The gene signature in the present study is a powerful predictor and risk factor for BCR after radical prostatectomy. PO-1188 Optimal radiotherapy strategy as risk-group in non-metastatic prostate cancer patients (KROG 18-15) S.H. Choi 1 , Y.S. Kim 2 , J. Yu 2 , T. Nam 3 , J. Kim 4 , B. Jang 4 , J.H. Kim 5 , Y. Kim 6 , B.K. Jung 7 , A.R. Chang 8 , Y. Park 8 , S.U. Lee 9 , K.H. Cho 9 , J.H. Kim 10 , H. Kim 11 , Y. Choi 12 , Y.J. Kim 13 , D.S. Lee 14 , Y.J. Shin 15 , S.J. Shim 16 , W. Park 17 , J. Cho 1 1 Yonsei Cancer Center- Yonsei University College of Medicine, Department of radiation oncology, Seoul, Korea Republic of ; 2 Asan Medical Center, Department of Radiation Oncology, Seoul, Korea Republic of ; 3 Chonnam National University Hwasun Hospital, Department of radiation oncology, Hwasun, Korea Republic of ; 4 Seoul National University Bundang Hospital, Department of radiation oncology, Bundang, Korea Republic of ; 5 Seoul National University Hospital, Department of radiation oncology, Seoul, Korea Republic of ; 6 Kyung Hee University Hospital, Department of radiation oncology, Seoul, Korea Republic of ; 7 School of Medicine- Gyeongsang National University, Department of radiation oncology, Chang won, Korea Republic of ; 8 Soonchunhyang University Seoul Hospital, Department of radiation oncology, Seoul, Korea Republic of ; 9 The Proton Therapy Center- Research Institute and Hospital- National Cancer Center, Department of radiation oncology, Ilsan, Korea Republic of ; 10 Keimyung University Dongsan Medical Center- Keimyung University School of Medicine, Department of radiation oncology, Daegu, Korea Republic of ; 11 Inha University Hospital- Inha University School of Medicine, Department of radiation oncology, Incheon, Korea Republic of ; 12 Dong-A University Hospital- Dong-A University School of Medicine, Department of radiation oncology, Pusan, Korea Republic of ; 13 Kangwon National University Hospital, Department of radiation oncology, Chuncheon, Korea Republic of ; 14 The Catholic University of Korea- Uijeongbu, Department of radiation oncology, Uijeongbu, Korea Republic of ; 15 Inje University Sanggye Paik Hospital, Department of radiation oncology, Inje, Korea Republic of ; 16 Eulji Hospital- Eulji University School of Medicine, Department of radiation oncology, Seoul, Korea Republic of ; 17 Samsung Medical Center- Sungkyunkwan University School of Medicine, Department of radiation oncology, Seoul, Korea Republic of Purpose or Objective External beam radiotherapy (EBRT) is evolving with numerous strategies incorporated to improve outcomes, and several studies showed an improvement in outcome by increasing the radiation dose in prostate cancer. Nevertheless, the optimum dose and fractionation scheme for each NCCN risk group is still unclear. The aim of Korean Radiation Oncology Group (KROG) 18-15 study was to validate the NCCN classification in multi-institutional large cohort and find the optimal EBRT strategy in each risk group. Material and Methods Non-metastatic prostate cancer patients who underwent definitive RT between 2001 and 2015 were eligible for this study. A total of 1,854 patients in 17 institutions were included and all were re-grouped in accordance with the NCCN Version 2.2019 guideline risk groups. Biochemical failure (BCF) was defined as nadir +2 ng/mL or the initiation of ADT for increasing PSA regardless of its value. The BCF-free survival (BCFFS), overall survival (OS) rates were estimated and radiation-related toxicities were evaluated according to the RTOG criteria. Results The numbers of each risk group patients were as follows: low 181 (10%), intermediate 425 (23%), high 799 (43%),

prostate cancer (PCa) to achieve good local control and low rectal toxicity. As rectal gas and rectal volume are known to have a relevant effect on prostate motion, this study aims to reduce these parameters by using a Low FODMAP Diet (LFD) and to show feasibility of this intervention. Material and Methods We compared a prospective intervention group (IG, n = 25) which underwent RT for PCa and whose patients were asked to follow a LFD during RT with a retrospective control group (CG, n = 25) which did not get any dietary advice. In the planning CT scan and all available cone beam CT scans rectal gas was classified based on a semiquantitative score (scale from 1 to 5) and rectal volume was measured. Furthermore, patients’ compliance was evaluated by a self-assessment questionnaire. Results Clinical and treatment characteristics were well balanced between both groups. A total of 266 (CG, 10.6 per patient) and 280 CT scans (IG, 11.2 per patient), respectively, were analysed. The frequency distribution of gas scores differed significantly from each other ( p < .001) with the IG having lower gas scores. Rectal volume was smaller in the IG (64.28 cm 3 , 95% CI 60.92 – 67.65 cm 3 , SD 28.64 cm 3 ) than in the CG (71.40 cm 3 , 95% CI 66.47 – 76.32 cm 3 , SD 40.80 cm 3 ) ( p = .02). Mean intrapatient standard deviation as a measure for the variability of rectal volume was 22 cm 3 in the IG and 23 cm 3 in the CG ( p = .81). Patients’ compliance and contentment were satisfying. Conclusion The use of a LFD significantly decreased rectal gas and rectal volume. LFD was feasible with an excellent patients’ compliance. However, prospective trials with a larger number of patients and a standardized evaluation of gastrointestinal toxicity and quality of life are reasonable. PO-1187 A gene signature-based model to predict BCR- free survival after radical prostatectomy M. Li 1 , R. SHI 1 , B. Claus 1 1 University Hospital- LMU Munich, Department of Radiation Oncology, Munich, Germany Purpose or Objective More precise methods to identify high-risk prostate cancer (PCa) patients who may suffer biochemical recurrence (BCR) after radical prostatectomy is a critical issue. We sought to construct an integrated model based on a molecular signature and clinicopathological features to optimize the risk stratification for BCR and improve biochemical recurrence-free survival (BCRFS) prediction in PCa patients after radical prostatectomy. Material and Methods Using high-throughput screening and least absolute shrinkage and selection operator (LASSO) in the training set, a gene signature for BCRFS was established and its prognostic value was validated in other five independent datasets. Multivariate Cox regression analysis was performed to evaluate the importance of risk for BCR. Time-dependent receiver operating characteristic (tROC) was used to evaluate the predictive power. Combined with traditional clinicopathological variables (Gleason score, pathological stage and surgical margin status), a decision tree was built to improve the risk stratification. Bioinformatic analyses were performed to explore potential biological processes underlying the gene signature. Results The gene signature exhibited powerful capacity in discriminating high-risk BCR patients, and multivariate Cox regression analysis demonstrated that the gene signature consistently acted as a risk factor for BCR in all six cohorts. The decision tree was able to identify high-risk subgroup powerfully. Bioinformatic analyses suggested that the gene signature was related with cell cycle process in PCa. Conclusion