ESTRO 36 Abstract Book

S536 ESTRO 36 2017 _______________________________________________________________________________________________

autophagy protein LC3-II and decreased levels of p62.

with pCR. This effect was most pronounced for the PC lipid species (Fig. 1) . Furthermore, preliminary results identified changes in plasma lipid species during CRT. A steeper increase in lipogenicity (PC, PE, Cer) was observed during CRT (time point 1 to 3) for patients achieving pCR in comparison to patients without pCR. Statistical analyses on the complete patient group are ongoing in order to validate our findings and to develop a discriminative marker panel with the most promising lipid markers.

Conclusion Plasma lipidomics is a novel field for biomarker development. Preliminary analyses show the potential of lipid profiling to discriminate rectal cancer patients with heterogeneous responses, treated with standard CRT. Further work will lead to the development of a predictive lipid plasma marker panel. Such a predictive panel might be used to stratify patients for an individualized treatment, thereby improving the quality of li fe for these patients. PO-0978 Potential predictive biomarkers to chemoradiotherapy response in rectal cancer: a lipidomic study. F. Perrotti 1 , P. Del Boccio 2 , D. Pieragostino 2 , L. Caravatta 1 , M. Di Tommaso 1 , C. Rosa 1 , M. Di Perna 1 , P. Sacchetta 2 , D. Genovesi 1 1 "SS Annunziata" Hospital, Radiotherapy, Chieti, Italy 2 "G. D'Annunzio" University, Medical Oral and Biotechnological Sciences, Chieti, Italy Purpose or Objective To highlight the lipid signature able to predict the tumor response to chemoradiotherapy (CRT), in patients with advanced rectal cancer (LARC), by using a Lipidomics approach. Material and Methods Between March 2013 and September 2014, 18 patients with LARC were treated with preoperative CRT at the Radiation Oncology Unit of SS Annunziata Hospital in Chieti – Italy. Sera were prospectively collected during routine chemistry tests before treatment (T0) and at day 14° (T14) and 28° (T28) of CRT. An open Liquid Chromatography tandem Mass Spectrometry (LC-MS/MS) analysis was performed to characterize lipid expression at T0. Differential lipids were validated by an independent targeted approach and studied during treatment. Results Sixty-five lipids significantly differentiated responder (RP) vs no-responder (NRP) patients; five lipids were validated as predictive of response to CRT: Sphingomyelin (SM, d18:2/18:1), Lysophosphatidylcholine (LPC, 16:0/0:0), Lysophosphatidylcholine (LPC, 15:1(9z)/0:0), Lysophosphatidylethanolamine (LPE, 22:5/0:0) and Phosphatidylcholine (PC, 40:2). Receiver Operator Characteristic curve (ROC curve), generated combining the pattern of the 5 validated lipids, showed an AUC of 0.95. Conclusion The prediction of response to neoadjuvant CRT in LARC allows to personalize treatments and to improve response rate and survival outcomes. In this study we focused on serum lipids to define a differential profile able to predict response. Our results showed a pattern of 5 lipids that differentiated RP and NRP before treatment. The ongoing confirmation of these results could provide a new insight

Conclusion Treatment with the HDACi SAHA resulted in induction of apoptosis and autophagy in both CRC cells and at a lesser extent in a relevant normal tissue model. Firstly, our results may contribute to explain adverse effects of SAHA on normal intestinal epithelial cells, and secondly, identify a therapeutic window where tumor radiosensitization can be achieved by SAHA. PO-0977 Plasma lipidomics for predictive biomarker analysis in rectal cancer. P. Bulens 1,2 , A. Debucquoy 2 , K. Bloch 2 , S. Fieuws 2 , J. Swinnen 2 , K. Haustermans 1,2 1 University Hospital Leuven, Radiation Oncology, Leuven, Belgium 2 KU Leuven - University of Leuven, Oncology, Leuven, Belgium Purpose or Objective Selection of patients with locally advanced rectal cancer, eligible for individualized treatment strategies, is hampered by the lack of reliable predictors of response. Plasma markers based on liquid biopsies would allow minimally invasive patient stratification. Most liquid biopsy approaches are based on the detection of free circulating DNA or tumor cells. Since the development and progression of cancer is associated with dramatic changes in lipid metabolism, we propose a radically different approach based on alterations in circulating lipids. Material and Methods From prospectively collected plasma samples of 85 rectal cancer patients at 3 time points (before chemoradiation (CRT), 2 weeks into CRT, end of CRT), lipids were extracted using a modified Bligh-Dyer protocol. Samples were subjected to mass spectrometry-based lipid profiling on a fully operational lipidomics platform. This approach allowed us to assess the concentration of approximately 200 different lipid species including phosphatidylcholine (PC), phosphatidylethanolamines (PE), phosphatidylinositol (PI), phosphatidylserine (PS) and ceramides (Cer). Based on the assessment of these species, discriminative lipid profiles of patients achieving a pathologic complete response (pCR) and patients lacking such response will be delineated using biostatistical approaches including PCA analysis followed by LDA and correction for false discovery due to multiple testing. Results 13 out of 85 patients achieved a pCR (15,3%). Preliminary analyses showed slightly less lipogenic profiles for patients