Abstract Book

S337

ESTRO 37

Results After a median follow-up of 6.49 years, 135 (55%) and 40 (13%) men experienced biochemical and metastatic relapses, respectively. IDC/CA subpathologies were identified in 160 (53%) of cases. Presence of IDC/CA predicted for inferior bRFR (HR 2.6, 95% CI 1.8-3.2, p <0.001) and mFR (HR 3.1, 95% CI 1.5-6.4, p = 0.0014). Of the 104 cases with GC scores, 22 (21%) were stratified as low-, 30 (29%) as intermediate-, and 52 (50%) as high- risk; a low GC score was associated with superior bRFR (HR 0.25, 95% CI 0.1-0.5, p <0.001) and mFR (HR 0.15, 95% CI 0.03-0.8, p = 0.025). On multivariable analyses, IDC/CA and GC independently predicted for bRFR, after adjustment for clinicopathologic prognostic indices, corresponding to an improved discrimination (C-statistics = 0.737). Conclusion IDC/CA subpathologies and GC are independent predictors of biochemical and metastatic relapses beyond conventional clinicopathologic prognostic features in the PORT setting. Patients harbouring IDC/CA subpathologies are at risk of systemic relapses, and HT-PORT intensification strategies ought to be considered in this unfavourable subgroup. Conversely, for men with low GC scores, de-intensification strategies (i.e. early salvage, PORT alone) should be explored. SP-0634 Navigation in the publication ethics landscape: Publish or perish or...? D. Zips 1 1 University Hospital Tübingen, Radiation Oncology, Tübingen, Germany Abstract text In this lecture I will reflect on the ethics of publication in biomedical research based on my own experience as clinician scientist and journal editor. For example I will discuss the dilemma of increasing pressure to publish as a key performance indicator in the academic setting on one side and on the other side keeping the highest quality standards in science. My talk will include suggestions for best practice which may be useful especially for young scientists. SP-0635 Risk Management: Quality and Safety New EU legislation and implementation M. Coffey 1 1 Coffey Mary, Discipline of Radiation Therapy- Trinity College Dublin, Dublin, Ireland Abstract text There are two components to the legal framework. One is the legislation or the statutes and laws enacted by the national legislative body and the more flexible system of regulation. Together they establish the basic framework of the national infrastructure and the regulations. Enabling legislation should be consistent with the national situation and reflect the practice in the country. Legislation is usually in place long-term and is not readily amended. Where technological developments and Teaching Lecture: Risk Management: Quality and Safety New EU legislation and implementation ESTRO 37 Tuesday 24 April 2018 Teaching Lecture: Navigation in the publication ethics landscape: Publish or perish or...?

Conclusion This is the first OE-MRI clinical study to evaluate biomarker repeatability and show that OE-MRI is sensitive to tumor hypoxia changes induced by radiotherapy. Our data suggest that OE-MRI is well-tolerated, feasible and repeatable and this provides a strong rationale for incorporating this imaging method into hypoxia-targeted therapy trials and radiotherapy dose painting studies in NSCLC. OC-0633 Subpathologies and genomic classifier for individualized post-prostatectomy radiotherapy A. Berlin 1 , M. Chua 1 , H. Truong 2 , A. Hosni 1 , M. Pintilie 1 , E. Davicioni 3 , A. Dicker 2 , T. Van der Kwast 4 , R. Bristow 1 , R. Den 2 1 Princess Margaret Cancer Centre - University Health Network, Department of Radiation Oncology - University of Toronto, Toronto, Canada 2 Sidney Kimmel Medical College at Thomas Jefferson University, Department of Radiation Oncology, Philadelphia, USA 3 GenomeDx Biosciences, Inc, Vancouver, Canada 4 University Health Network, Department of Pathology and Laboratory Medicine, Toronto, Canada Purpose or Objective Clinical stratification for post-prostatectomy radio- therapy (PORT) using conventional clinicopathologic prognostic factors (i.e. post operative PSA, pT-category, Gleason score [GS], and surgical margins status) is imprecise, and accounts for substantial over- and under- treatment. Current recommendations include the combination of PORT with hormonal therapy (HT) to improve salvage rates. Therefore, better individualized patient selection could spare unnecessary toxicities and improve outcomes. Herein, we investigated the prognostic utility of the unfavourable subpathologies intraductal carcinoma and cribriform architecture (IDC/CA), and a 22-gene genomic classifier (GC) in prostate cancer (PCa) patients receiving PORT. Material and Methods We utilized a cohort of 302 men who received PORT from two academic institutions. Conventional clinicopathologic prognostic factors were: median pre-radiotherapy PSA 0.17 (IQR 0.07-0.47) ng/ml, 22% GS 8-10, 74% extraprostatic extension (EPE), and 64% surgical margins involved. PORT was delivered as adjuvant or salvage in 38% and 62% of cases, respectively; 20% received HT- PORT . Pathological specimens were centrally reviewed. In 104 cases, GC scores were determined from the prostatectomy specimen by Decipher prostate cancer classifier assay (GenomeDx Biosciences, San Diego, CA). Study endpoints were biochemical relapse-free (bRFR) and metastasis-free (mFR) rates.