ESTRO 38 Abstract book

S186 ESTRO 38

avenues for potentially safe dose escalation in difficult to treat anatomical locations such as pancreatic cancers, it also creates increased demand and time pressures on clinical care teams. Here we discuss examples from clinical practice highlighting the benefits, challenges, and practical solutions for clinical use of real time MR-guided adaptive radiation. SP-0373 Fractionation and breast cancer: towards more efficient schedules J.R. Yarnold 1 1 The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, Radiotherapy & Imaging Department, Sutton, United Kingdom Abstract text In 1952, Lionel Cohen, reported a retrospective analysis of fractionation parameters associated with local control in women treated in Johannesburg for locally advanced/recurrent breast cancer. He found that breast cancer is sensitive to fraction size, an interpretation confirmed in 1984 when Bruce Douglas re-analysed Cohen’s data using the new L-Q model. This historical observation simulated a series of UK randomised trials that have generated direct estimates of alpha/beta (a/b) for breast cancer and late reacting normal tissues in patients irradiated after primary surgery. An a/b value of around 3 Gy has been generated for both endpoints. So, conventional 2 Gy fractions are equally sparing of cancer and the late reacting normal tissue and have no clinical advantages. 40 Gy in 15 fractions of 2.7 Gy in 3 weeks has become an international standard, but it may not represent the safe and effective limit of hypofractionation. The FAST randomised trial controlled for time by testing 5 fractions delivered once-weekly for 5 weeks against 50 Gy in 25 fractions in 5 weeks. The results informed the FAST forward randomised trial testing two dose levels of a 5-fraction regimen delivered in 1 week against 40 Gy in 15 fractions in 3 weeks and 3-yr late normal tissue responses have been presented. The UK trial designs, incorporating 2 test dose levels, enable interpolation, where needed, to identify test schedules iso-effective with the Control regimen for tumour and normal tissues without distinguishing fraction size and time-related effects. OC-0374 GWAS identifies new susceptibility loci for late toxicity following prostate cancer radiotherapy C. West 1 , S. Kerns 2 , L. Dorling 3 , G. Barnett 4 , D. Dearnaley 5 , L. Fachal 6 , L. Veldeman 7 , M. Parliament 8 , A. Vega 9 , A. Dunning 10 , B. Rosenstein 11 1 The University of Manchester, Translational Radiobiology, Manchester, United Kingdom ; 2 University of Rochester Medical Center, Department of Radiation Oncology, Rochester, USA; 3 University of Cambridge, Department of Public Health & Primary Care, Cambridge, United Kingdom ; 4 Addenbrookes Hospital, Department of Oncology, Cambridge, United Kingdom; 5 Institute of Cancer Research, Academic Unit of Radiotherapy, Sutton, United Kingdom ; 6 University of Cambridge, Department of Public Health and Primary Care, Cambridge, United Kingdom ; 7 Universiteit Gent, Department of Radiation Oncology & Experimental Cancer Research, Gent, United Kingdom; 8 University of Alberta, Department of Oncology, Alberta, Canada ; 9 Fundación Pública Galega Medicina Xenómica, Complejo Hospitalario Universitario de Santiago, Santiago de Compostela, Spain ; 10 University of Cambridge, Centre for Cancer Genetic Epidemiology, Cambridge, United Proffered Papers: RB 4: Normal tissue effects of radiotherapy

Kingdom ; 11 Mount Sinai, Icahn School of Medicine, New York, USA Purpose or Objective There is a risk of long-term side-effects following radiotherapy for cancer. Identification of common genetic variants can provide information on susceptibility, and may lead to improve risk prediction and inform functional work to better understand biological mechanisms. The purpose of this Radiogenomics Consortium study was to expand our previous meta-analysis from ~1,600 patients to 4,000 patients in an attempt to identify new loci associated with risk of late toxicity following radiotherapy. Material and Methods An individual patient data meta-analysis was carried out in six European-ancestry genome-wide association studies (n=3,871) of radiotherapy-treated prostate cancer survivors. Radiotherapy side-effects were graded prospectively in all studies. Cox proportional hazards regression tested associations of ~6 million genotyped or imputed variants with urinary and rectal toxicity endpoints (time to first ≥grade 2 event). Results Individual patient data meta-analysis of the European ancestry cohorts identified three signals with genome- wide significance and low Bayesian false discovery probability (<2%): rs17055178 (p=6.2x10 -10 ), rs10969913 (2.9x10 -10 ), rs11122573 (p=1.8x10 -8 ), associated with rectal bleeding, decreased urinary stream, and haematuria radiation induced risk, respectively. Fine mapping the regions containing these signals identified an additional independent signal associated with haematuria. Credible causal variants at the four signals lie in gene- regulatory regions and some modulate expression of nearby genes. All variants retained significance in multivariable models including clinical and dosimetric risk factors. Previously identified variants showed a consistent association effect in the new cohorts ( KDM3B, DHAH5, ATM ). Two new and two previously identified signals showed a similar effect in two Japanese ancestry cohorts. Conclusion Our study highlights the importance of common genetic variants as a determinant of long-term side-effects following radiotherapy, increases understanding of the architecture of common genetic variants affecting risk of late radiotherapy side-effects, and shows further multi- national studies in larger cohorts are worthwhile. OC-0375 Improving lung cancer outcome by reducing normal lung tissue toxicity L. Giuranno 1 , E. Roig Moreno 1 , R. Iannone 1 , M. Vooijs 1 1 Maastricht Radiation Oncology Maastro, Radiotherapy, Maastricht, The Netherlands Purpose or Objective Lung cancer is the leading cause of cancer death in western countries. The current standard of care includes surgery , chemotherapy and radiotherapy. While significant progress has been made in terms of treatment modality radiotherapy is limited by dose-limiting side-effects which negatively affect tumour control and patient's quality of life. Reducing side-effects may improve tumor control by dose-escalation and treatment-time. The Notch signaling pathway plays an important role in lung cell differentiation and regeneration of the airway epithelium and its deregulation is associated with several cancers including lung cancer. Notch signaling pathway alteration leads to poor outcome and treatment resistance in patients and in preclinical models suggesting Notch signaling as a novel therapeutic target. However, the mechanism through which Notch inhibition integrates with airway repair and cellular differentiation is not fully understood. What is currently lacking are primary human lung tissue models that enable robust evaluation of normal