Volume49, 49,Suppl. Suppl. 1, March 2013 Volume 4, November 2013 ISSN 0959-8049 EJC EUROPEAN JOURNAL OF CANCER 5th Asian Oncology Summit and 9th Annual MarkersConference in Cancer: of the Organisation for A joint meeting by ASCO, EORTC and NCI Oncology and Translational Research Abstract Book March 22–24, 2013, Bangkok, Thailand 7–9 November 2013 Abstract Book Brussels, Belgium THE OFFICIAL JOURNAL OF European Journal of Cancer Editor-in-Chief: Editors: Basic and Preclinical Research: Drug Development: Early Breast Cancer: Advanced Breast Cancer: Gastrointestinal Cancers: Genitourinary Cancers: Lung Cancer: Gynaecological Cancers: Head and Neck Cancer: Sarcomas: Melanoma: Neuro-oncology: Epidemiology and Prevention: Paediatric Oncology: Founding Editor: Past Editors: Editorial Office: Alexander M.M. Eggermont Institut Gustave Roussy Villejuif, France Richard Marais, Manchester, UK Giorgio Parmiani, Milan, Italy Jean-Charles Soria, Villejuif, France Kathleen I. Pritchard, Toronto, Canada David Cameron, Edinburgh, UK Eric Van Cutsem, Leuven, Belgium Michel Ducreux, Villejuif, France Cora Sternberg, Rome, Italy Mary O’Brien, London, UK Ignace Vergote, Leuven, Belgium Kevin Harrington, London, UK Jean-Yves Blay, Lyon, France Dirk Schadendorf, Essen, Germany Roger Stupp, Zurich, Switzerland Jan Willem Coebergh, Rotterdam, The Netherlands Rob Pieters, Rotterdam, The Netherlands Henri Tagnon Michael Peckham, London, UK; Hans-Jörg Senn, St Gallen, Switzerland; John Smyth, Edinburgh, UK Elsevier, The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK Tel: +44 (0) 1865 843590, Email: ejcancer@elsevier.com EDITORIAL BOARD CLINICAL ONCOLOGY J.-P. Armand (France) A. Ayhan (Japan) R. Blamey (UK) M. Bolla (France) J. Boyages (Australia) N. Brünner (Denmark) F. Cardoso (Portugal) J. Cassidy (UK) M. Castiglione (Switzerland) L. Cataliotti (Italy) L. Cheng (USA) H. Cody (USA) R. Coleman (UK) A. Costa (Italy) J. De Bono (UK) M.J.A. De Jong (The Netherlands) E. de Vries (The Netherlands) A. Dicker (USA) R. Dummer (Switzerland) F. Eisinger (France) S. Erridge (UK) G. Ferrandina (Italy) H. Gabra (UK) H. Gelderblom (The Netherlands) B. Hasan (Belgium) J.C. Horiot (Switzerland) C. Huber (Germany) R. Jakesz (Austria) J. Jassem (Poland) D. Jodrell (UK) V.C. Jordan (USA) A. Katz (Brazil) M. Kaufmann (Germany) I. Kunkler (UK) L. Lindner (Germany) P.E. Lønning (Norway) P. Lorigan (UK) K. McDonald (Australia) R. Mertelsmann (UK) F. Meunier (Belgium) T. Mok (Hong Kong) D. Nam (Korea) P. O’Dwyer (USA) J. Overgaard (Denmark) N. Pavlidis (Greece) J. Perry (Canada) P. Price (UK) D. Raghavan (USA) J. Ringash (Canada) J. Robert (France) A. Rody (Germany) D. Sargent (USA) M. Schmidinger (Austria) S. Sleijfer (The Netherlands) P. Sonneveld (The Netherlands) A. Sparreboom (USA) M. van den Bent (The Netherlands) M. Van Glabbeke (Belgium) G. Velikova (UK) U. Veronesi (Italy) A. Vincent-Salomon (France) A. Voogd (The Netherlands) E. Winquist (Canada) BASIC, PRECLINICAL AND TRANSLATIONAL RESEARCH A. Albini (Italy) P. Allavena (Italy) F. Balkwill (UK) M. Barbacid (Spain) M. Broggini (Italy) C. Catapano (Switzerland) J. Collard (The Netherlands) E. Garattini (Italy) A. Gescher (UK) R. Giavazzi (Italy) I. Hart (UK) W. Keith (UK) L.A. Kiemeney (The Netherlands) J. Lunec (UK) D.R. Newell (UK) G.J. Peters (The Netherlands) A. Puisieux (France) V. Rotter (Israel) M. Schmitt (Germany) C.G.J. Sweep (The Netherlands) G. Taraboletti (Italy) P. Vineis (UK) N. Zaffaroni (Italy) D. Forman (France) A. Green (Australia) K. Hemminki (Germany) C. Johansen (Denmark) L.A. Kiemeney (The Netherlands) M. Maynadié (France) H. Møller (UK) P. Peeters (The Netherlands) A.G. Renehan (UK) S. Sanjose (Spain) M.K. Schmidt (The Netherlands) H. Storm (Denmark) L.V. van de Poll-Franse (The Netherlands) H.M. Verkooijen (The Netherlands) R. Zanetti (Italy) G. Chantada (Argentina) F. Doz (France) A. Ferrari (Italy) M.A. Grootenhuis (The Netherlands) K. Pritchard-Jones (UK) L. Sung (Canada) M. van den Heuvel-Eibrink (The Netherlands) M. van Noesel (The Netherlands) EPIDEMIOLOGY AND PREVENTION B. Armstrong (Australia) P. Autier (France) J.M. Borras (Spain) C. Bosetti (Italy) H. Brenner (Germany) L.E.M. Duijm (The Netherlands) J. Faivre (France) S. Franceschi (France) PAEDIATRIC ONCOLOGY C. Bergeron (France) A. Biondi (Italy) E. Bouffet (Canada) M. Cairo (USA) H. Caron (The Netherlands) Volume 49, Supplement 4 November 2013 ISSN 0959-8049 European Journal of Cancer Editor-in-Chief Alexander M.M. Eggermont Markers in Cancer: A joint meeting by ASCO, EORTC and NCI Abstract Book 7–9 November 2013 Brussels, Belgium Amsterdam • Boston • London • New York • Oxford • Paris • Philadelphia • San Diego • St Louis European Journal of Cancer Aims and Scope The European Journal of Cancer (EJC) is an international multidisciplinary oncology journal, which publishes original research, reviews, and editorial comments on basic and preclinical cancer research, translational oncology, clinical oncology – including medical oncology, paediatric oncology, radiation oncology, and surgical oncology, and cancer epidemiology and prevention. The EJC is the official journal of the European Organisation for Research and Treatment of Cancer (EORTC), the European CanCer Organisation (ECCO), European Association for Cancer Research (EACR) and the European Society of Breast Cancer Specialists (EUSOMA). For a full and complete Guide for Authors, please go to http://www.ejcancer.com Advertising information. Advertising orders and inquiries can be sent to: USA, Canada and South America: Pat Hampton Advertising Department, Elsevier Inc., 360 Park Avenue South, New York, NY 10010-1710, USA; phone: (+1) (212) 633 3181; fax: (+1) (212) 633 3820; e-mail: p.hampton@elsevier.com. Europe and ROW: Advertising Sales: Elsevier Pharma Solutions; 32 Jamestown Road, London NW1 7BY, UK; Tel.: +44 (0) 20 7424 4259; fax: +44 (0) 20 7424 4433; e-mail: elsevierpharma.uk@elsevier.com. Publication information: European Journal of Cancer (ISSN 0959-8049). For 2013, volume 49 (18 issues) is scheduled for publication. Subscription prices are available upon request from the Publisher or from the Elsevier Customer Service Department nearest you or from this journal’s website (http://www.elsevier.com/locate/ejca). Further information is available on this journal and other Elsevier products through Elsevier’s website (http://www.elsevier.com). Subscriptions are accepted on a prepaid basis only and are entered on a calendar year basis. Issues are sent by standard mail (surface within Europe, air delivery outside Europe). Priority rates are available upon request. Claims for missing issues should be made within six months of the date of despatch. Orders, claims, and journal inquiries: please contact the Elsevier Customer Service Department nearest you: St. Louis: Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA; phone: (800) 6542452 [toll free within the USA]; (+1) (314) 4478871 [outside the USA]; fax: (+1) (314) 4478029; e-mail: JournalsCustomerService-usa@elsevier.com. Oxford: Elsevier Customer Service Department, The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK; phone: (+44) (1865) 843434; fax: (+44) (1865) 843970; e-mail: JournalsCustomerServiceEMEA@elsevier.com. Tokyo: Elsevier Customer Service Department, 4F Higashi-Azabu, 1-Chome Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; phone: (+81) (3) 5561 5037; fax: (+81) (3) 5561 5037; e-mail: JournalsCustomerServiceJapan@elsevier.com. Singapore: Elsevier Customer Service Department, 3 Killiney Road, #08-01 Winsland House I, Singapore 239519; phone: (+65) 6349 0222; fax: (+65) 6733 1510; e-mail: JournalsCustomerServiceAPAC@elsevier.com. Author inquiries For inquiries relating to the submission of articles (including electronic submission) please visit this journal’s homepage at http:// www.elsevier.com/locate/ejca. For detailed instructions on the preparation of electronic artwork, please visit http://www.elsevier. com/artworkinstructions. Contact details for questions arising after acceptance of an article, especially those relating to proofs, will be provided by the publisher. You can track accepted articles at http://www.elsevier.com/trackarticle. You can also check our Author FAQs at http://www.elsevier.com/authorFAQ and/or contact Customer Support via http://support.elsevier.com. Language services. Please write your text in good English (American or British usage is accepted, but not a mixture of these). Authors who feel their English language manuscript may require editing to eliminate possible grammatical or spelling errors and to conform to correct scientific English may wish to use the English Language Editing service available from Elsevier’s WebShop http:// webshop.elsevier.com/languageediting/ or visit our customer support site http://support.elsevier.com for more information. Funding body agreements and policies. Elsevier has established agreements and developed policies to allow authors whose articles appear in journals published by Elsevier, to comply with potential manuscript archiving requirements as specified as conditions of their grant awards. To learn more about existing agreements and policies please visit http://www.elsevier.com/fundingbodies. USA mailing notice: European Journal of Cancer (ISSN 0959-8049) is published monthly with extra issues in January, March, May, July, September, November by Elsevier Ltd. (The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK). Periodicals postage paid at Jamaica, NY 11431 and additional mailing offices. USA POSTMASTER: Send change of address to European Journal of Cancer, Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA. AIRFREIGHT AND MAILING in USA by Air Business Ltd., c/o Worldnet Shipping Inc., 156-15, 146th Avenue, 2nd Floor, Jamaica, NY 11434, USA. Ó 2013 Elsevier Ltd. All rights reserved This journal and the individual contributions contained in it are protected under copyright by Elsevier Ltd and the following terms and conditions apply to their use: Photocopying. Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. For information on how to seek permission visit www.elsevier.com/permissions or call (+44) 1865 843830 (UK)/(+1) 215 239 3804 (USA). Derivative works. Subscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions. Permission of the Publisher is required for resale or distribution outside the institution. Permission of the Publisher is required for all other derivative works, including compilations and translations (please consult www.elsevier.com/permissions). Electronic storage or usage. Permission of the Publisher is required to store or use electronically any material contained in this journal, including any article or part of an article (please consult www.elsevier.com/permissions). Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior permission of the Publisher. Notice. No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality or value of such product or of the claims made of it by its manufacturer. The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Peer Review Policy for the European Journal of Cancer (EJC) The practice of peer review is to ensure that only good science is published. It is an objective process at the heart of good scholarly publishing and is carried out by all reputable scientific journals. Our reviewers therefore play a vital role in maintaining the high standards of the European Journal of Cancer (EJC) and all manuscripts are peer reviewed following the procedure outlined below. Initial manuscript evaluation The Editors first evaluate all manuscripts. In some circumstances it is entirely feasible for an exceptional manuscript to be accepted at this stage. Those rejected at this stage are insufficiently original, have serious scientific flaws, have poor grammar or English language, or are outside the aims and scope of the journal. Those that meet the minimum criteria are passed on to experts for review. Authors of manuscripts rejected at this stage will be informed within 2 weeks of receipt. Type of Peer Review The EJC employs single blind review, where the reviewer remains anonymous to the authors throughout the process. How the reviewer is selected Reviewers are matched to the paper according to their expertise. Our reviewer database contains reviewer contact details together with their subject areas of interest, and this is constantly being updated. Reviewer reports Reviewers are asked to evaluate whether the manuscript: – – – – – Is original Is methodologically sound Follows appropriate ethical guidelines Has results which are clearly presented and support the conclusions Correctly references previous relevant work Reviewers are not expected to correct or copyedit manuscripts. Language correction is not part of the peer review process. Reviewers are requested to refrain from giving their personal opinion in the ‘‘Reviewer blind comments to Author’’ section of their review on whether or not the paper should be published. Personal opinions can be expressed in the ‘‘Reviewer confidential comments to Editor’’ section. How long does the peer review process take? Typically the manuscript will be reviewed within 2-8 weeks. Should the reviewers’ reports contradict one another or a report is unnecessarily delayed a further expert opinion will be sought. Revised manuscripts are usually returned to the Editors within 3 weeks and the Editors may request further advice from the reviewers at this time. The Editors may request more than one revision of a manuscript. Final report A final decision to accept or reject the manuscript will be sent to the author along with any recommendations made by the reviewers, and may include verbatim comments by the reviewers. Editor’s Decision is final Reviewers advise the Editors, who are responsible for the final decision to accept or reject the article. Special Issues / Conference Proceedings Special issues and/or conference proceedings may have different peer review procedures involving, for example, Guest Editors, conference organisers or scientific committees. Authors contributing to these projects may receive full details of the peer review process on request from the editorial office. Becoming a Reviewer for the EJC If you are not currently a reviewer for the EJC but would like to be considered as a reviewer for this Journal, please contact the editorial office by e-mail at ejcancer@elsevier.com, and provide your contact details. If your request is approved and you are added to the online reviewer database you will receive a confirmatory email, asking you to add details on your field of expertise, in the format of subject classifications. iv Acknowledgements With the support of the Minister of Scientific Research of the Brussels Capital Region Pharmaceutical Industries Media partners for their valuable assistance in improving the visibility of this meeting v Table of Contents Acknowledgements iv Foreword vii Faculty List viii Societies’ Profiles x Scientific Program Scientific Program – Overview xiii Scientific Program – Details xiv Faculty’s Biosketches xix Speakers’ Presentations S1 Topic 1: New trial concept and challenges for the future? S1 Topic 2: Biology-driven clinical trials Topic 3: Progress and implications of novel technology for clinical trials S2 S3 Topic 4: Regulatory issues Topic 5: Circulating / Imaging biomarkers S4 S5 Topic 6: Predictive markers for immunotherapy Topic 7: Pathway driven approaches S5 S6 Topic 8: Genomics driven approaches An EORTC Workshop on Biospecimen Pre-Analytical Stability and Diagnostics S7 S8 Six Best Poster Abstracts – Oral Presentations S11 Poster Presentations S13 Author Index S39 Disclosures S43 Abstracts Index S46 As a significant number of cancers in Europe can be attributed to smoking a strict no-smoking policy will be enforced within all areas used by the conference. Save the Date Markers in Cancer Diagnostic Development Tutorial May 5-6, 2014 Bethesda North Marriott Hotel and Conference Center Bethesda, Maryland, USA The 1 ½ day Diagnostic Development Tutorial will address issues critical to understanding the advances, limitations, development, and validation of molecular markers. This limited-attendance event will feature didactic presentations and hands-on discussions of real-life challenges. Working in small teams, participants will create development plans for assays that will aid clinical decisions related to choice of therapy. This format will encourage interaction and reflective learning and provide ample opportunity for discussion and faculty feedback. Attendance is limited to early-career oncologists. This live activity has been approved for AMA PRA Category 1 Credit ™. markersincancer.org vii Markers in Cancer: A joint meeting by ASCO, EORTC and NCI 7–9 November 2013, Brussels, Belgium “Be ready for the next generation of cancer treatments” Foreword Dear Colleagues, We cordially invite you to attend Markers in Cancer, a joint meeting by ASCO, EORTC & NCI which will be held 7–9 November 2013 in Brussels. Next generation sequencing and high throughput screening technologies are revolutionizing our current molecular understanding of cancer. In addition, molecular tumor profiles can now be established at affordable cost within a timeframe compatible with clinical practice. It is anticipated, that in the coming decade biomarker-based patient selection will move to the forefront of effective cancer treatment for the majority of patients. This year’s program will address the challenges we face, the opportunities which await us, and focus on the relevance and the feasibility of implementing emerging technologies in clinical practice. You will learn the ins and outs of current biomarker research including next generation biomarker discovery as well as the development and quality control necessary to turn next generation biomarkers into reliable diagnostics. Furthermore, it is likely that clinical trial design involving multiplex biomarker testing and molecular imaging will have to be adapted to provide evidence of which (combination) treatment is most effective for the heterogeneous and genetically diverse cancer patient population presenting in the clinic. The meeting brings together clinicians, laboratory scientists, bio-informaticians, pathologists, statisticians, industry representatives, healthcare providers, regulatory agencies, and patient advocates who together with an outstanding faculty will deal with these emerging issues. We are also pleased to announce that, for the first time, we will host a Workshop on “Biospecimen Pre-Analytical Stability and Diagnostics: A joint NCI BRN/SPIDIA Workshop”! A Diagnostic Development Tutorial will be held prior to the Main Meeting (5–7 November 2013) in order to address issues critical to understanding the advances, development, limitations, and validation of molecular markers. This Diagnostic Development Tutorial, “From Hypothesis to Product”, is open to a limited audience and will feature didactic presentations and hands on discussion of real life challenges. We look forward to welcoming you in Brussels for what promises to be a captivating and intellectually stimulating meeting! After the meeting you will be ready for the new era of cancer care based on next generation guided therapy selection. Christophe LE TOURNEAU Co-chair John MARTENS Co-chair viii Faculty List Chairs: Christophe Le Tourneau & John Martens Organizing Committee ASCO EORTC NCI James Abbruzzese MD Anderson Cancer Center Houston (US) Christophe Le Tourneau Institut Curie Paris (FR) Tracy Lively National Cancer Institute Rockville (US) Lisa Carey UNC – Chapel Hill Chapel Hill (US) John Martens Erasmus MC Rotterdam (NL) Magdalena Thurin National Cancer Institute Rockville (US) Scientific Committee ASCO EORTC NCI Janet Dancey Ontario Institute for Cancer Research Kingston (CA) Nandita deSouza Institute of Cancer Research & Royal Marsden NHS Foundation Trust Sutton (GB) Shivaani Kummar National Cancer Institute Bethesda (US) Francisco Esteva NYU Cancer Institute New York (US) Patricia LoRusso Karmanos Cancer Institute Detroit (US) David Sidransky Johns Hopkins University Baltimore (US) Walter Stadler University of Chicago Chicago (US) Christian Dittrich Ludwig Boltzmann Institute for Applied Cancer Research Vienna (AT) Jacqueline Hall EORTC Headquarters Brussels (BE) Michail Ignatiadis Institut Jules Bordet Brussels (BE) Roberto Salgado Institut Jules Bordet Brussels (BE) Frank Lin National Cancer Institute Rockville (US) Helen Moore National Cancer Institute Bethesda (US) Eric Polley National Cancer Institute Bethesda (US) John Welch National Cancer Institute Rockville (US) Tutorial Planning Committee ASCO EORTC NCI Susan Hilsenbeck Baylor College of Medicine Houston (US) Fred Sweep UMC Radboud Nijmegen (NL) Mei Polley National Cancer Institute Bethesda (US) Edward Kim Levine Cancer Institute Carolinas Healthcare System Charlotte (US) Sabine Tejpar University Hospital Gasthuisberg Leuven (BE) Paul Williams Frederick National Laboratory for Cancer Research Frederick (US) Faculty List ix Other Faculty Members Tommy Alain McGill University and Goodman Cancer Research Centre Montreal (CA) Udai Banerji The Institute of Cancer Research & The Royal Marsden Sutton (GB) Robert Becker U.S. Food and Drug Administration Silver Spring (US) Jan Bogaerts EORTC Headquarters Brussels (BE) David Chelsky Caprion Montreal (CA) Nai-Kong Cheung Memorial Sloan-Kettering Cancer Center New York (US) Elisabeth de Vries University Medical Center Groningen Groningen (NL) Angela DeMichele University of Pennsylvania School of Medicine Philadelphia (US) Mitchell Dowsett Royal Marsden Hospital London (GB) David Eberhard University of North Carolina Chapel Hill (US) Iordanis Gravanis European Medicines Agency London (GB) Kelly Oliner Amgen, Inc. Newbury Park (US) Richard Grose Barts Cancer Institute London (GB) Xavier Paoletti Institut Curie Paris (FR) Juhl Hartmut Indivumed GmbH Hamburg (DE) Mario Pazzagli University of Florence Florence (IT) Monika Hegi University Hospital Lausanne Lausanne (CH) Michael Postow Memorial Sloan-Kettering Cancer Center New York (US) Manuel Hidalgo Centro Integral Oncologico Clara Campal Madrid (ES) Jordi Rodon Vall d’Hebron Institute of Oncology Barcelona (ES) Stefanie Jeffrey Stanford University Standford (US) Richard Kaplan University College London, UCL Hospital & NCRN Coordinating Centre London (GB) Shirin Khambata-Ford Novartis Pharmaceuticals Corp East Hanover (US) Denis Lacombe EORTC Headquarters Brussels (BE) Ultan McDermott Wellcome Trust Sanger Institute Cambridge (GB) Philip Febbo (US) Federico Monzon Baylor College of Medicine Houston (US) Thierry Gorlia EORTC Headquarters Brussels (BE) Uwe Oelmueller QIAGEN GmbH Hilden (DE) Nicolas Servant Institut Curie Paris (FR) Steven Shak Genomic Health Inc Redwood City (US) Frazer Symmans UT MD Anderson Cancer Center Houston (US) Geraldine Thomas Imperial College London London (GB) Fernando Ulloa Montoya GSK Vaccines Rixensart (BE) Kurt Zatloukal Institute of Pathology Graz (AT) x Societies’ Profiles The ASCO Mission and Vision The American Society of Clinical Oncology (ASCO) is a professional oncology society committed to conquering cancer through research, education, prevention and delivery of high quality patient care. Vision Statements: • All cancer patients will have lifelong access to high quality, effective, affordable and compassionate care; • The most accurate cancer information will be available so that patients and physicians can make informed decisions about cancer prevention and treatment; • Information we learn from every patient will be used to accelerate progress against cancer; • Resources will exist to attract the best clinicians and investigators to provide optimal patient care and to conduct transformative research; • ASCO will be recognized as the most trusted source of cancer information worldwide. Membership Our diverse network of more than 30,000 oncology professionals recognizes ASCO’s dedication to provide the highest quality resources in education, policy, the pioneering of clinical research and above all, advancing the care for patients with cancer. ASCO is unique in that we are the only organization that encompasses all oncology subspecialties, allowing our members to grow from the professional and personal expertise of their colleagues worldwide and across disciplines. International members make up approximately 30% of the Society’s total membership and represent more than 120 countries. The ASCO Annual Meeting The ASCO Annual Meeting brings together more than 30,000 oncology professionals from a broad range of specialties to share the latest cancer research in the areas of basic and clinical science. The abstracts presented each year at the Annual Meeting reflect the foremost research and strategies in oncology that will directly impact patient care. To learn more, visit www.asco.org Societies’ Profiles xi The European Organisation for Research and Treatment of Cancer (EORTC) brings together European cancer clinical research experts from all disciplines to improve cancer care. Both multinational and multidisciplinary, the EORTC Network comprises more than 2,500 collaborators from all disciplines involved in cancer research and treatment in more than 300 hospitals in over 30 countries. EORTC Headquarters, a unique pan European clinical research infrastructure based in Brussels, Belgium, handles some 30 protocols that are permanently open to patient entry, over 50,000 patients who are in follow-up, and a database of more than 180,000 patients. Through translational and clinical research, the EORTC offers an integrated approach to drug development, drug evaluation programs and medical practices. EORTC studies have contributed to success stories in drugs development including the registration of several drugs by the United States Food and Drug Administration and the European Medicines Agency. The EORTC has a proven track record in establishing new standards, e.g. RECIST, QOL, etc., and in changing clinical practice. Full clinical, scientific, operational, quality assurance, and regulatory support for clinical and translational research projects is provided by the EORTC Headquarters staff who augment this support with strong expertise in biostatistics, clinical study design and methodology, endpoint definition and analysis, and new initiatives in imaging and long term survivorship. Through its scientific strategy, the EORTC aims to define the future of cancer therapy. It recognizes that the role of pragmatic data and related methodology need to be developed within multi-stakeholder platforms, and with this goal in mind, the EORTC participates in projects such as those put forth by the Innovative Medicines Initiative. Such activities make the EORTC one of Europe’s leading players in transforming experimental discoveries into new treatments with a significant clinical impact for the benefit of patients. www.eortc.org xii Societies’ Profiles The National Cancer Institute (NCI) is a component of the National Institutes of Health (NIH), one of eight agencies that compose the Public Health Service (PHS) in the Department of Health and Human Services (DHHS). The NCI, established under the National Cancer Act of 1937, is the Federal Government’s principal agency for cancer research and training. The National Cancer Act of 1971 broadened the scope and responsibilities of the NCI and created the National Cancer Program. Over the years, legislative amendments have maintained the NCI authorities and responsibilities and added new information dissemination mandates as well as a requirement to assess the incorporation of state-of-the-art cancer treatments into clinical practice. The National Cancer Institute coordinates the National Cancer Program, which conducts and supports research, training, health information dissemination, and other programs with respect to the cause, diagnosis, prevention, and treatment of cancer, rehabilitation from cancer, and the continuing care of cancer patients and the families of cancer patients. Specifically, the Institute: • Supports and coordinates research projects conducted by universities, hospitals, research foundations, and businesses throughout this country and abroad through research grants and cooperative agreements. • Conducts research in its own laboratories and clinics. • Supports education and training in fundamental sciences and clinical disciplines for participation in basic and clinical research programs and treatment programs relating to cancer through career awards, training grants, and fellowships. • Supports research projects in cancer control. • Supports a national network of cancer centers. • Collaborates with voluntary organizations and other national and foreign institutions engaged in cancer research and training activities. xiii Scientific Program – Overview Posters can be viewed in the Grand Hall 1 (Level-2): From Thursday 7 November (12h30) to Saturday 9 November (11h00) SESSION A – Thursday 7 November 2013 (Afternoon) Personalized Medicine 12.30–13.30 13.30–15.15 Welcome Coffee TOPIC 1: New Trial Concepts and Challenges for the Future 15.15–15.45 15.45–18.00 Coffee Break TOPIC 2: Biology-Driven Clinical Trials 18.00–20.30 Posters viewing & discussion (Buffet dinner) SESSION B – Friday 8 November 2013 (Morning) Challenges of Next Generation Cancer Therapy 08.00–10.00 10.00–10.30 TOPIC 3: Progress and Implications of Novel Technology for Clinical Trials Coffee Break 10.30–12.30 TOPIC 4: Regulatory Issues 12.30–13.30 Lunch SESSION C – Friday 8 November 2013 (Afternoon) NonInvasive Biomarkers 13.30–15.30 TOPIC 5: Circulating / Imaging Biomarkers 15.30–16.00 16.00–17.45 Coffee Break TOPIC 6: Predictive Markers for Immunotherapy SESSION D – Saturday 9 November 2013 (Morning) Markers for Cancer Therapy 08.00–10.00 TOPIC 7: Pathway Driven Approaches 10.00–10.30 10.30–12.15 Coffee Break TOPIC 8: Genomics Driven Approaches 12.15–13.30 Lunch An EORTC Workshop on Biospecimen Pre-Analytical Stability and Diagnostics Saturday 9 November 2013 (Afternoon) 13.30–14.30 14.30–15.30 Opening and Introduction SESSION I: Preanalytical Variation Affecting Analysis of Proteins, RNA, and Free Circulating 15.30–16.00 DNA in blood Coffee Break 16.00–17.30 SESSION II: Preanalytical Variation Affecting Analysis of Proteins, RNA, and Morphology in Post-Surgical Tissues 17.30–18.30 SESSION III: Development of Evidence-Based Biospecimen Standards xiv Scientific Program – Details SESSION A Personalized Medicine Thursday 7 November 2013 (Afternoon) Chairs: Christophe Le Tourneau (EU) & Tracy Lively (US) TOPIC 1 (13h30 – 15h15) NEW TRIAL CONCEPT AND CHALLENGES FOR THE FUTURE Biostatistical challenges for the design of biology-driven trials (X. Paoletti) Bioinformatics analysis for real-time applications (N. Servant) Pathology challenges for biology-driven trials: The Ki67 experience (M. Dowsett) Selected abstract for oral presentation (abstract # MC13-0049): Two-stage adaptive cutoff design for building and validating a prognostic biomarker signature (M. Polley) TOPIC 2 (15h45 – 18h00) BIOLOGY-DRIVEN CLINICAL TRIALS Rationale of the design of the WINTHER trial (J. Rodon) Rationale of the design of the I-SPY trial (A. DeMichele) Rationale of the design of the SHIVA trial (C. Le Tourneau) Molecular profiling based assignment of cancer therapy (MPACT) (S. Kummar) Selected abstract for oral presentation (abstract # MC13-0075): Rapid assessment of TORC1 suppression as a functional biomarker predicting responsiveness to RAF and MEK inhibitors in BRAF-mutant melanoma patients (R.B. Corcoran) Scientific Program – Details xv SESSION B Challenges of Next Generation Cancer Therapy Friday 8 November 2013 (Morning) Chairs: John Martens (EU) & Lisa Carey (US) TOPIC 3 (08h00 – 10h00) PROGRESS AND IMPLICATIONS OF NOVEL TECHNOLOGY FOR CLINICAL TRIALS Clinical reporting of NGS data (D. Eberhard) Selected abstract for oral presentation (abstract # MC13-0060): Analytical validation of the MPACT assay, a targeted next generation sequencing clinical assay for cancer patient treatment selection (C. Lih) Are we ready for genomes in the clinic? (U. McDermott) Application of next generation sequence for retrospective biomarker analyses in tumors samples from Phase II/III clinical trials (S. Khambata-Ford) Discussion / Q&A (All) TOPIC 4 (10h30 – 12h30) REGULATORY ISSUES Translational scientists – Overview of process of taking an assay from the lab to the clinic (J. Martens) EMA perspective on approval issues (I. Gravanis) FDA view on approval issues and new guidelines (R. Becker) R&D’s view on obstacles to development of diagnostics for clinical development (S. Shak) Clinician’s view (TBC) Statistical issues on biomarker ascertainment and power to detect interactions (J. Bogaerts) Discussion / Q&A (All) xvi Scientific Program – Details SESSION C NonInvasive Biomarkers Friday 8 November 2013 (Afternoon) Chairs: Michail Ignatiadis (EU) & Magdalena Thurin (US) TOPIC 5 (13h30 – 15h30) CIRCULATING / IMAGING BIOMARKERS Circulating tumor cells (CTCs) and circulating DNA to monitor tumor heterogeneity in clinical trials (S. Jeffrey) PET imaging to monitor tumor heterogeneity in clinical trials (E. de Vries) Selected abstract for oral presentation (abstract # MC13-0026): KRAS mutated plasma DNA as predictor of outcome from irinotecan monotherapy in metastatic colorectal cancer (K. Spindler) Panel Discussion: What is the role of “liquid biopsy” and functional imaging to interrogate tumor evolution in clinical trials (All) TOPIC 6 (16h00 – 17h45) PREDICTIVE MARKERS FOR IMMUNOTHERAPY Monitoring response to anti CTLA-4 therapy (M. Postow) Predictive gene signature (GS) in MAGE-A3 antigen-specific cancer immunotherapy (F. Ulloa Montoya) Bone marrow minimal residual disease (MRD) was the strongest predictor of survival from high risk metastatic neuroblastoma (NB) following anti-GD2 immunotherapy, when tested in multivariate models that include FcR polymorphism and missing ligand for inhibitory killer-immunoglobulin-like receptor (KIR) (N. Cheung) Selected abstract for oral presentation (abstract # MC13-0071): Immune response against non-targeted tumor antigens after treatment with sipuleucel-T and its association with improved clinical outcome (D. Guhathakurta) Scientific Program – Details xvii SESSION D Markers for Cancer Therapy Saturday 9 November 2013 (Morning) Chairs: Christian Dittrich (EU) & Francisco Esteva (US) TOPIC 7 (08h00 – 10h00) PATHWAY DRIVEN APPROACHES Challenges and opportunities for molecular screening platforms: The EORTC SPECTA program (D. Lacombe) Pathways triggered by FGF and other growth factors (R. Grose) Prediction of the efficacy of mTOR targeted therapies (T. Alain) Genotype-based combinations of RAS/RAF and PI3K pathway inhibitors (U. Banerji) TOPIC 8 (10h30 – 12h15) GENOMICS DRIVEN APPROACHES Genomic wide biomarker discovery in personalized patient derived xenografts (M. Hidalgo) RAS mutations as markers of resistance for colorectal cancer patients treated with the anti-EGFR monoclonal antibody panitumumab (K. Oliner) Epigenetic markers in cancer (M. Hegi) Selected abstract for oral presentation (abstract # MC13-0079): Prospective mutational characterization of Japanese patients with non-small cell lung cancer using surgically resected tumor specimens by next-generation sequencing (Y. Koh) xviii Scientific Program – Details An EORTC Workshop on Biospecimen Pre-Analytical Stability and Diagnostics Saturday 9 November 2013 (Afternoon) Chairs: Roberto Salgado (EU) & Helen Moore (US) OPENING AND INTRODUCTION (13h30 – 14h30) Opening remarks from the co-chairs (H. Moore & R. Salgado) Keynote address: Overcoming the challenges of biospecimen collection for R&D, clinical trials and biobanking (G. Thomas) Overview of biospecimen science in the BRN and SPIDIA programs (H. Moore & U. Oelmueller) SESSION I (14h30 – 15h30) Preanalytical Variation Affecting Analysis of Proteins, RNA, and Free Circulating DNA in blood A controlled assessment of pre-analytical variables and their impact on the proteome in the collection and storage of plasma and serum (D. Chelsky) Evidence-based guidelines for the pre-analytical phase of DNA, RNA and cell-free DNA testing in blood samples (M. Pazzagli) SESSION II (16h00 – 17h30) Preanalytical Variation Affecting Analysis of Proteins, RNA, and Morphology in Post-Surgical Tissues Effects of intrasurgical and postsurgical variables on stability of signaling molecules in cancer tissues (J. Hartmut) Evaluation of novel alternatives to formalin fixation for companion diagnostics (K. Zatloukal) Preanalytical variation affecting detection of RNA in breast cancer tissues (F. Symmans) SESSION III (17h30 – 18h30) Development of Evidence-Based Biospecimen Standards NCI biospecimen evidence-based practices (H. Moore) SPIDIA – Dissemination of results into CEN technical specifications for biospecimen handling (U. Oelmueller) Elements of a Global Action Plan for Harmonization of Biospecimen Handling (R. Salgado) xix Faculty’s Biosketches James Abbruzzese James L. Abbruzzese, MD, FACP is the Waun Ki Hong Distinguished Chair and Chairman of the Department of Gastrointestinal Medical Oncology at the University of Texas M.D. Anderson Cancer Center in Houston, Texas. Dr. Abbruzzese has published over 400 peer-reviewed articles, numerous chapters and reviews. In 2001 Dr. Abbruzzese served as a co-chair for the American Association for Cancer Research Program Committee and he was the Program Chairman for the American Society of Clinical Oncology Annual Meeting in 2007. He has recently served as a member of the AACR Board of Directors and currently serves on the ASCO Board of Directors. He is a past member of the AACR Research Fellowships Committee, the ASCO Grant Awards and Nominating Committees, and currently serves as chairman of National Cancer Institute Clinical Trials and Translational Research Advisory Committee. He is a Deputy Editor of Clinical Cancer Research, and member of several other editorial boards including past service for the Journal of Clinical Oncology. His scholarly interests center on clinical and translational research for pancreatic cancer. Tommy Alain Tommy Alain received his PhD at the University of Calgary under the mentorship of Dr. Peter Forsyth and Dr. Patrick Lee. He trained in oncology, cell biology, and virology while studying the oncolytic potencies of reovirus, vesicular stomatitis virus, and myxoma virus in various cancers. During his post-doctoral studies at McGill University in the laboratory of Dr. Nahum Sonenberg, he pursued work on the control of translation initiation and the mammalian target of rapamycin (mTOR) signaling pathway for oncology research, in particular on the roles that the translation initiation factor eIF4E, and its repressors 4E-BPs, play in modulating the efficacy of anti-cancer therapeutics. Currently a research associate in the laboratory of Dr. Sonenberg at McGill University, he will begin in the new year his new position as a Scientist and Assistant Professor at the Children’s Hospital of Eastern Ontario and the University of Ottawa. Dr. Alain has expertise in cell based assays, generating cell models of cancer, and developing virology and translation strategies to target malignancies. Udai Banerji Dr. Udai Banerji obtained his MD in India at the University of Bombay in 1994. He started as a research fellow at the Institute of Cancer Research, obtaining his PhD in 2005 while completing his medical oncology training at The Royal Marsden. In April 2007, he was appointed as a Senior Lecturer and Honorary Consultant in Medical Oncology at The ICR and The Royal Marsden; he is now a Cancer Research UK Clinical Senior Lecturer. The portfolio of the Phase I unit at The Royal Marsden includes over 30 phase I trials. Dr. Banerji is the Deputy Director of the Drug Development Unit and is involved in running a number of these trials. He is interested in the discovery and applications of HSP90, PI3K, AKT and mTOR inhibitors and has been involved in phase I studies of HSP90 inhibitors such as 17-AAG, 17-DMAG, CNF204 and AUY922. In addition to Phase 1 trials, his independent laboratory research interests include the use of pharmacodynamic (PD) biomarkers; examples include the use of HSP70, CRAF and CDK4 as PD biomarkers and HSP90 inhibitors. Other interests include quantifying signalling output to design combinations of signalling agents to reverse resistance to anti-cancer drugs. Robert Becker Robert Becker is Chief Medical Officer for the Office of In Vitro Diagnostics and Radiological Health (OIR), Center for Devices and Radiological Health (CDRH), FDA, with special attention to inter-office coordination on regulation of newly emerging genetic/genomic IVD’s. Dr. Becker previously served as Director, Division of Hematology and Immunology Devices, in OIR (then named OIVD). He is experienced in regulation of IVD’s aimed at cell- and tissue-based specimens (e.g. classical hematology, flow cytometry, cytology, histopathology), plus blood coagulation tests, and immunoserologic tests. Dr. Becker earned his MD and PhD in Immunology at Duke University, and he is board certified in anatomic and clinical pathology. He served in the United States Air Force as a pathologist at the Armed Forces Institute of Pathology, Washington, DC from 1983 to 2004, specializing in urologic pathology and with research and clinical service applying image analysis and flow cytometry to diagnostic pathology. Jan Bogaerts Jan Bogaerts earned his degree (1986) and PhD (1993) in mathematics at the Free University of Brussels (VUB). In 1988 he earned a degree in management (VUB). From 1986 to 1993 he worked as assistant in mathematics and statistics at the faculties of Economic, Social and Political Sciences at VUB. In 1993 he joined BMS as statistician, and was promoted Associate Director Statistics. He worked on the development of several drugs in oncology, including FDA and EMEA submissions. In 2004 he joined the EORTC as statistician of the EORTC Breast Cancer Group. xx In 2006 he was appointed senior statistician and in 2010 Head of the Statistics department. In 2012 he was appointed Methodology Vice Director. He contributed to the development of version 1.1 of RECIST and is on the RECIST Steering Committee. He represents EORTC as course director of the Flims workshop. He is also the statistician of the EORTC MINDACT trial. Statistical interests include the use of methodological issues around Progression Free Survival, alternative ways to use changes in tumor measurements as predictive markers, and the correct evaluation of the contribution of new markers to existing prognostic risk evaluation. Lisa Carey Lisa A. Carey, MD is the Richardson and Marilyn Jacobs Preyer Distinguished Professor of Breast Cancer Research in the UNC Department of Medicine, Division of Hematology/Oncology. In 2012 she became the Division Chief of Hematology/Oncology, as well as the Physicianin-Chief of the North Carolina Cancer Hospital. At UNCLinberger she is the Medical Director fo the Breast Center, and in 2010 became the Cancer Center’s Associate Director for Clinical Research. Dr. Carey has a longstanding research interest in the clinical applications of laboratory findings in breast cancer, with a particular interest in the clinical implications of different molecular subtypes of breast cancer. She designs and leads clinical trials of novel drugs and approaches, and is a close collaborator with several laboratory investigators and epidemiologists.She was awarded a Doris Duke Clinician Scientist Award in 1999, a Career Development Award from the NCI in 2000, and was inducted into the Johns Hopkins Society of Scholars in 2008. In 2011, Dr. Carey was awarded the NCI Director’s Service Award. David Chelsky No information received Nai-Kong Cheung Nai-Kong V. Cheung received his MD and PhD (Immunology) from Harvard Medical School. Following specialty training in Pediatric Hematology/Oncology at Stanford University, he became interested in the immunotherapy of cancer, developing monoclonal antibodies targeting tumor antigens in metastatic neuroblastoma, including antibody 3F8 against ganglioside GD2 and 8H9 against B7-H3. These antibodies were successfully tested in the setting of minimal residual disease (MRD) where myeloid effectors and natural killer cells play critical roles. In order to measure MRD accurately, he exploited genome-wide screens to develop panels of molecular markers. He holds the Enid A. Haupt Chair in Pediatric Oncology, heads the neuroblastoma program and directs the Robert Steel Laboratory at Memorial Sloan-Kettering Cancer Center in New York. There he has established an active translational program Faculty’s Biosketches to explore novel genetically engineered antibodies and to bring them to pediatric patients diagnosed with metastatic cancers. http://www.mskcc.org/prg/prg/bios/203.cfm and http:// www.mskcc.org/mskcc/html/55349.cfm Janet Dancey Dr. Dancey is Program Leader for High Impact Clinical Trials, Ontario Institute for Cancer Research, Director, Clinical Translational Research and Physician Coordinator, Melanoma Disease Site Committee for NCIC Clinical Trials Group, Chair, Cancer Care Ontario Experimental Therapeutics Network and Professor, Department of Oncology, Queen’s University. Previously, she was Associate Chief in the Investigational Drug Branch of the Cancer Therapy Evaluation Program of the National Cancer Institute. She completed medical school at the University of Ottawa in 1988. She received certifications in internal medicine and medical oncology. In her current position Dr. Dancey is responsible for development of strong translational research programs within NCIC CTG clinical trials and OICR High Impact Clinical Trial Program. Her clinical focus is on melanoma and gastrointestinal malignancies. Key accomplishments include establishment of the High Impact Clinical Trial Program following successful international peer review, and development of novel trials to evaluate investigational drugs in rare tumour settings, to evaluate next-generation sequencing technologies in cancer patient management. Elisabeth de Vries Prof. Dr. E.G. Elisabeth de Vries, MD, PhD is Professor of Medical Oncology, and head of the Department of Medical Oncology at the University Medical Center Groningen, Groningen, the Netherlands. She is involved in patient care, teaching, and research. Her research lines are aimed at increasing the sensitivity of tumors to anticancer drugs, and she uses imaging techniques to support this. Apart from laboratory studies, she performs and coordinates clinical trials. She has received numerous grants and is PI of CTMM (Center for Translational and Molecular Medicine) grant MAMMOTH ,of ERC advanced grant OnQview and of Alpe d’HuZes grant IMPACT. She is currently chairperson of the committee for the new RECIST 2.0 version on behalf of the EORTC. In 2002, she was appointed as a member of the Royal Academy of Arts and Sciences (KNAW). She received the European Society of Medical Oncology (ESMO) award in 2009. She is Fellow of the European Academy of Cancer Sciences and member of the Governing Body European Academy of Cancer Sciences since 2010. She was awarded a Royal Netherlands Academy of Sciences professorship in 2011. Faculty’s Biosketches Angela DeMichele Dr. DeMichele is an Associate Professor of Medicine and Epidemiology at the University of Pennsylvania and Senior Scholar in the Center for Clinical Epidemiology and Biostatistics. She received a BS in Biochemistry from Brown University, and MD from Washington University and an MS in Clinical Epidemiology from the University of Pennsylvania, where she also received clinical training in Internal Medicine and Hematology/Oncology. She is currently Co-Leader of the Breast Cancer Research Program and Co-Director of the 2-PREVENT Breast Cancer Translational Center of Excellence at the Abramson Cancer Center. Her research focuses on developing breast cancer biomarkers and novel therapeutics through numerous clinical trials and epidemiologic studies supported by the NIH, Pharma and Komen Foundation. She is a past recipient of an ASCO YIA, a Career Development Award from the American Cancer Society and an NIH K23 award. She directed Penn’s Doris Duke Clinical Research Fellowship Program and has served on the American Board of Internal Medicine Oncology Subspecialty Board, the ASCO Education Committee and numerous editorial boards, including the Journal of Clinical Oncology. Nandita deSouza Nandita M. deSouza, Professor of Translational Imaging and Co-Director of the MRI Unit, Institute of Cancer Research, UK. Main interests: Oncological imaging with particular emphasis on gynaecological, prostate and breast tumours, using functional imaging techniques to understand biology, improve staging and monitor treatment response. Nandita holds a Cancer Research UK Imaging Centre grant in MRI as Co-Principal Investigator and has several project and studentship grants. She currently chairs the EORTC Imaging Group. Publications: ∼130 peer-reviewed articles, several book chapters and editor of a multi-author book on Endocavitary MRI of the pelvis. Christian Dittrich Christian Dittrich, MD, Professor of Medicine at the Vienna University School of Medicine, is head of the 3rd Medical Department Centre for Oncology and Hematology at the Kaiser Franz Josef-Spital in Vienna and director of the Ludwig Boltzmann Institute for Applied Cancer Research (LBI-ACR VIEnna). Recently, he became coordinator of the ESMO Faculty for Principles of Clinical Trials and Systemic Therapy. He serves as member of the EORTC Protocol Review Committee (PRC) and Scientific Audit Committee (SAC). He is a representative of the EORTC Network of Core Institutions (NOCI) and serves actually as chair of the EORTC New Drug Advisory Committee (NDAC) and as member of the EORTC Board. Representing ESMO, he is faculty member of the Clinical Trials Workshops of Flims and Saudi Arabia. He has been running a CME-initiative in trials’ methodology in German xxi language under the patronage of ESO-d/DESO since 1998. He has research interests in clinical trials’ methodology, new drug development and translational research, focused on solid tumors. He has been acting as principal or coinvestigator in numerous clinical trials from phase I (first in human) to phase III. Mitchell Dowsett No information received. David Eberhard David A. Eberhard MD, PhD is Associate Professor in the Departments of Pathology and Pharmacology at the University of North Carolina at Chapel Hill. Areas of research interest include: molecular pathology and genomics of solid tumors; oncology companion diagnostics; preclinical and clinical development of novel assays and therapeutics for personalized medicine in oncology; digital pathology and image analysis of solid tumors. Dr. Eberhard directs the Pre-Clinical Molecular Pathology Laboratory in the Lineberger Cancer Center at UNC, providing molecular pathology support for translational research efforts in cancer genomics including Next-Generation Sequencing of tumor and non-tumor human biosamples. Previously, he served as Director of Clinical Trials Pathology Services at LabCorp, as a pathologist-scientist at Genentech Inc. and as Assistant Research Professor at the University of Virginia. Francisco Esteva Dr. Francisco J. Esteva is Professor of Medicine at New York University School of Medicine, Associate Director of Clinical Investigation of the NYU Cancer Institute, Director of the Breast Medical Oncology Program and co-Director of the Phase I Program at NYU. Dr. Esteva received his medical and doctoral degrees from the University of Zaragoza in Spain. He completed a residency in internal medicine at Cooper Hospital University Medical Center (Camden, NJ) and a fellowship in medical oncology at Georgetown University Medical Center (Washington, DC). He was a faculty member at MD Anderson Cancer Center (Houston, Texas) from 1997 to 2013. He received the K23 patient-oriented research career development award from the US National Cancer Institute and multiple research awards. He is a member of the American Society for Clinical Investigation. He is an author of more than 100 publications in the area of breast cancer research and treatment. Dr. Esteva is dedicated to the clinical development of novel therapies for breast cancer, with a passion for improving the survival and quality of life of cancer patients through innovative research and compassionate patient care. Philip Febbo No information received. xxii Thierry Gorlia Thierry Gorlia is a member of the team allocated by the EORTC Headquarters to provide scientific and logistic support to EORTC Brain Tumor Group and Translational Research (TR) Unit in the conduct of cancer clinical trials and TR projects. He reports to the Head of Biostatistics Unit and work in close cooperation with the other members of his teams (coordinating physician, data manager, etc.), and with other units of the Headquarter for particular projects. Thierry Gorlia has focused his research on the prognosis and diagnosis of Brain tumors in clinical trials. He has been involved in the development of disease specific quality of life modules (Gastric), health economics studies within clinical trials (brain, ovarian cancer). He has developed expertise in the statistical design of TR projects and their analyses. In particular, when microarrays are involved. Iordanis Gravanis I studied medicine at the University of Ioannina, in Greece between 1992 and 1998. I then worked for three years (1998–2001) as primary care physician in Greece, half of it as conscript in the Greek army. From January 2002 until April 2007, I studied towards a PhD in Molecular and Cellular Pharmacology at Stony Brook University, NY, working on mouse models of neurodegeneration. I joined the European Medicines Agency in August 2008 where I have been managing oncology drug authorisations and post-authorisation drug lifecycle as well as providing scientific secretarial support to working parties developing regulatory guidance for medicines. Richard Grose 1999 University College London, PhD in Cell Biology 1999–2001 Postdoctoral Fellow, Institute of Cell Biology, ETH Zurich, Switzerland 2001–2004 Postdoctoral Fellow, Cancer Research UK London Research Institute 2004–2010 Lecturer in Cell Biology, Barts Cancer Institute 2010–present Senior Lecturer in Cell Biology, Barts Cancer Institute My group is interested in understanding how Fibroblast Growth Factors (FGFs) and their receptors (FGFRs), which play critical roles during development, are hijacked by cancer cells to drive tumourigenesis. FGFR signalling can be a positive driving force for cell proliferation, survival and migration but is kept under tight control via feedback loops. In cancer, these controls can be bypassed by a variety of mechanisms and we are investigating how this happens. We are focusing currently on breast, pancreatic and endometrial cancer, using 2D and 3D cell based models to investigate how cellular behaviour changes when FGFR signalling is perturbed. We collaborate with clinical Faculty’s Biosketches colleagues to determine the clinical significance of our findings through analysis of patient samples. Jacqueline Hall Dr. Hall joined the EORTC (Belgium) Translational Research Unit in May 2009 after 8 years of research and training in multidisciplinary environments. She completed a bioinformatics PhD at Glasgow University where she worked alongside Epigenomics AG (Berlin) and Orion Genomics Inc (St Louis) investigating DNA methylation markers. After her PhD she joined McGill University (Montreal) for a post-doc in breast cancer at the McGill Center for Bioinformatics, Rosalind and Morris Goodman Cancer Centre. At the EORTC HQ she now focuses on infrastructure development and integration of translational research activities into EORTC clinical studies. This involves the implementation of human biological material collection and biobanking, developing and implementing quality assurance principles for biomarker assays in trials as well as managing processes for review and implementation of translational research in EORTC clinical trials. Dr. Hall also has a keen interest in bioinformatics and data sharing. She is also a member of the EORTC PathoBiology Group (PBG) and EORTC Pharmacology and Molecular Mechanisms Group (PAMM). Juhl Hartmut Founder and Chief Executive Officer of Indivumed GmbH and Inostics GmbH in Hamburg. Hartmut Juhl studied medicine from 1979 to 1986 and received his doctorate of medicine in 1989. From 1987 to 1992 he worked as a research associate at the Hamburg Eppendorf Surgical University Clinic.In 1995 he became a specialist in general surgery at the Christians Albrecht University in Kiel and obtained his habilitation in 1996. From 1999 to 2002 he led a gastrointestinal cancer research group as Associate Professor at the Lombardi Cancer Center of Georgetown University in Washington DC. In 2002, Hartmut Juhl became co-founder of Indivumed GmbH in Hamburg. Indivumed is focused on individualizing cancer therapy based on a highly standardized tumor tissue bank, a comprehensive clinical data collection and a broad range of specialized research services. In 2008, Indivumed GmbH and scientists from Johns Hopkins University, Baltimore/Maryland, founded the subsidiary Inostics GmbH which is specialized in detection of cancer-specific somatic DNA-changes. Hartmut Juhl holds an extraordinary professorship at the Hamburg University and is Adjunct Professor at the Lombardi Cancer Center of Georgetown University. Monika Hegi Monika Hegi is associate professor for Experimental and Translational Neuro-Oncology at the University Lausanne, Switzerland. After her PhD (Dr.sc. nat.) at the ETH Zurich she pursued post-doctoral training in molecu- Faculty’s Biosketches lar toxicology and carcinogenesis at the National Institute of Environmental Health Sciences (NIEHS), NIH, RTP, NC, USA, and directs the laboratory of Brain Tumor Biology and Genetics in the Department of Clinical Neurosciences since 1998.She works in close collaboration with international cooperative groups, in particular the EORTC Brain Tumor Group, for which she serves as coordinator for translational research. Analyzing molecular profiles of gliomas from patients enrolled in trials, has uncovered mechanisms of treatment resistance. Clinically most relevant was the demonstration of a predictive value for epigenetic silencing of the repair gene MGMT for benefit from the alkylating agent temozolomide in trials for glioblastoma. This has led to a paradigm change in neuro-oncology: The MGMT methylation status is now used as biomarker for stratification or patient selection in glioma trials and guides treatment decisions in particular for elderly glioblastoma patients. Manuel Hidalgo Manuel Hidalgo was born in Antequera, Malaga, in 1968. He received his MD from the Universidad de Navarra in 1992 and his PhD from the Universidad Autonoma de Madrid in 1997. Manuel specialized in Medical Oncology at the Hospital Universitario 12 de Octubre, Madrid, obtaining his license in 1996. He completed his training in drug development at the University of Texas Health Science Center, San Antonio (USA), where he briefly joined as Faculty. He then moved to Johns Hopkins University in 2001 as Co-Director of the Drug Development and GI Programs.In 2009, he joined the CNIO to lead the GI Cancer Clinical Research Unit. Manuel is a founding member of the pancreatic cancer research team.He has published 180 papers in peer-reviewed journals and his work has been funded by the NCI, AACR, and ASCO.In 2011, he was named Vice Director of Translational Research at CNIO with the mission to foster translational research. Susan Hilsenbeck Susan Hilsenbeck, PhD is a Professor of Medicine at the Baylor College of Medicine in Houston TX, USA and is Director of the Biostatistics and Informatics Shared Resource of the Duncan Cancer Center at Baylor. She is an internationally known biostatistician with particular interests in prognostic and predictive markers in cancer, especially breast cancer, and design and analysis of early phase, translational clinical trials. She is the statistician of record for a number of single and multi-center early phase breast cancer trials. She also has a long-standing commitment to education of clinical investigators. Dr. Hilsenbeck is a member of the Tutorial Committee for the Markers in Cancer Diagnostic Development Tutorial. xxiii Michail Ignatiadis Michail Ignatiadis, MD PhD is attending Physician (Medical Oncologist) at the Medical Oncology Department, Jules Bordet Institute (IJB), Brussels, Belgium since 2007. He obtained his MD from the Aristotle University of Thessaloniki Greece in 1996, his Medical Oncology specialization from the University of Crete, Greece in 2006 and his PhD on circulating tumor cells in early breast cancer from the University of Crete, Greece in 2008. Apart from his clinical appointment in IJB he is also actively involved in translational and clinical research. Dr. Ignatiadis is particularly interested in drug and biomarker development in breast cancer. Stefanie Jeffrey Stefanie Jeffrey, MD, is the John and Marva Warnock Professor and Chief of Surgical Oncology Research in the Department of Surgery at Stanford University School of Medicine. She received her undergraduate degree in Chemistry and Physics and master’s degree in Chemistry from Harvard University. She graduated from medical school at University of California San Francisco (UCSF), where she also completed her surgical residency. Professor Jeffrey was a key member of the Stanford/Norway team that pioneered the use of DNA microarrays to classify breast cancers by their molecular subtypes. She later led the Stanford team that invented the MagSweeper, a device that captures rare cells, such as circulating tumor cells, live and at extremely high purity for downstream single cell analyses. Using patient tissues obtained fresh from the operating room, her laboratory generates patient-derived orthotopic xenograft models of breast and colorectal cancer for use in pre-clinical studies of circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), and other circulating biomarkers during drug treatment. Richard Kaplan Pr. Kaplan is Senior Clinical Scientist and Chair at the MRC Clinical Trials Unit at University College London, Consultant in Oncology at UCL Hospital, and Associate Director of the National Cancer Research Network (NCRN) Coordinating Centre. He is a medical oncologist and trialist, formerly Chief of the Clinical Investigations Branch, National Cancer Institute (NCI),NIH, and Program Director for the NCI’s national program of Cooperative Groups and Consortia.He was responsible for scientific coordination of the NCs portfolio of funded or sponsored treatment trials in GI, GU and CNS malignancies. He has served on advisory committees and panels for NCI, NIH, FDA and EMA, as well as for other government agencies, clinical trials organisations and specialty societies in North America, Europe and Australia. In 2004 he came to the UK to become Associate Director for Industry for NCRN and Professor of Oncology in the Leeds Institute of Molecular Medicine and has promoted collaborative research, both xxiv commercial and investigator-led, between the NHS and the pharma and biotech industries. Since 2008 he has also led clinical research in colorectal, prostate, and renal cancer at MRC CTU. Shirin Khambata-Ford Dr. Shirin Khambata Ford is Global Head Correlative Sciences, Oncology Global Development, Novartis. She leads a group focused on clinical biomarker development and translational research for full development programs. Prior to this she was a Sr. Biomarker Experimental Medicine Leader, Roche where she led biomarker work for early development oncology programs. She was previously Director, Oncology Biomarkers, Bristol-Myers Squibb where she led a group responsible for the discovery, validation and implementation of clinical biomarkers for the late stage portfolio. Her key accomplishments include being the lead researcher on the team that discovered that EGFR ligand expression and KRas mutation status are predictive of benefit from cetuximab in metastatic CRC. This resulted in a seminal publication in Journal of Clinical Oncology. Her research has led to publications in journals such as New England Journal of Medicine, multiple presentations at oncology conferences – ASCO, AACR, ESMO and patent applications. As the Biomarker Lead for cetuximab, she was a key participant in regulatory and diagnostic activities in addition to providing scientific leadership and expertise. Edward Kim Edward S. Kim, MD is Chair of Solid Tumor Oncology and Investigational Therapeutics and the Donald S. Kim Distinguished Chair for Cancer Research at the Levine Cancer Institute, Carolinas HealthCare System in Charlotte, NC. He previously was Associate Professor and Chief of the Section of Head and Neck Cancer at MD Anderson Cancer Center. Dr. Kim studies novel targeted agents in the treatment and prevention settings and has expertise in lung, head and neck, as well as thymic cancers and chaired BATTLE, the personalized medicine program in lung cancer. Dr. Kim serves on the editorial boards of JCO, CCR, and Clinical Lung Cancer. Dr. Kim is also the recipient of several awards including the ASCO YIA and the AACR Scholar in Training Award. He also served as the PI of the MD Anderson SWOG U10 institutional grant and is the recipient of a V Foundation Grant. Dr. Kim is the author or coauthor of more than 100 published articles, book chapters, reviews in journals such as Lancet, Journal of Clinical Oncology, Cancer Discovery, Cancer, and Cancer Prevention Research. Shivaani Kummar Dr. Shivaani Kummar received her medical degree from Lady Hardinge Medical College, New Delhi, India and her Internal Medicine Residency training from Emory Univer- Faculty’s Biosketches sity, Atlanta, Georgia. Upon completion of her fellowship training in Medical Oncology and Hematology from the National Institutes of Health, Bethesda, MD, she joined Yale Cancer Center, Yale University, New Haven, Connecticut as an Assistant Professor in Medical Oncology. In 2004 she moved back to the National Cancer Institute to pursue her work in early drug development in oncology. At present, she is Head of Early Clinical Trials Developemnt, Office of the Director, Division of Cancer Treatment and Diagnosis, and head of the Developmental Therapeutics Section, Medical Oncology Branch, National Cancer Institute. Dr. Kummar has published in major peer-reviewed journals in the field of oncology, and serves of the scientific program committee for the American Society of Clinical Oncology (ASCO) annual meetings and is a member of ASCO and the American Association for Cancer Research (AACR). Denis Lacombe In 1988 Dr. Lacombe graduated as MD from the University of Marseilles (France). He was granted for a Master Post-doctoral Fellowship at the Roswell Park Cancer Institute, Buffalo, NY, USA for Fundamental and Clinical Pharmacokinetics from 1989 to 1991. From 1991 to 1993 he worked as Clinical Research Adviser in charge of the development of a new drug in oncology in the pharmaceutical industry. From September 1993 to January 2013 he has been working as Clinical Research Physician at the EORTC in Brussels for various Groups. Until 2007, as Assistant Director Medical Affairs, he coordinated the EORTC New Drug Development Team conducting new drug development studies. Currently amongst his activities, he is responsible for the Early Project Optimization Department dedicated to ensure EORTC studies being in line with the EORTC strategy. Since November 2010, he has been appointed as Director HQ of the EORTC with main missions being to ensure that the functioning of the EORTC HQ is in line with the decisions of the EORTC Board and to develop and coordinate the future EORTC Scientific Strategy in accordance with the evolving health care ecosystems. Christophe Le Tourneau Christophe Le Tourneau has been appointed as a senior Medical Oncologist at the Institut Curie in November 2009. He is heading the Phase I Program as well as the Head and Neck Clinic. He is also running a multicenter randomized personalized medicine trial (SHIVA). Christophe Le Tourneau was certified in Medical Oncology in 2005 and got his PhD in Clinical Epidemiology in 2007. He did a Clinical Research Fellowship at Princess Margaret Hospital in Toronto, Canada, in the Drug Development Program from November 2007 to November 2009 under the supervision of Pr Lillian Siu. His main interests are oncology phase I clinical trials with a special attention at the Faculty’s Biosketches methodology to conduct these trials, as well as Head and Neck oncology. Christophe Le Tourneau is the principal investigator of several phase I trials, as well as of clinical trials in Head and Neck oncology. Frank Lin Frank I. Lin, MD, is currently at the National Cancer Institute (NCI) in the U.S. National Institutes of Health. Dr. Lin is Board-certified in both Internal Medicine and in Nuclear Medicine, and completed his clinical imaging training at Stanford University and at the University of California UC Davis. Since joining the National Cancer Institute in 2010, Dr. Lin has served as the Medical Officer in the Clinical Trials Branch of Cancer Imaging Program, where he oversees all aspects of imaging trials from Phase 0 to Phase III that are sponsored by the NCI. Dr. Lin’s research interests involve the role of both functional and molecular imaging in the diagnosis of cancer and evaluation of cancer therapies. Tracy Lively Dr. Tracy Lively joined the NIH in 1996 as a program director in the Cancer Diagnosis Program of the National Cancer Institute. Prior to coming to the NIH she had been an assistant professor in the Division of Biomedical Sciences at the University of California, Riverside, and had completed post-doctoral fellowships in cancer biology and human genetics. As a program director, and later as Deputy Associate Director of the Cancer Diagnosis Program, Dr. Lively has been responsible for the scientific oversight of a portfolio of investigator-initiated research grants. She has also developed and implemented targeted research initiatives for exploratory research, for technology development and for patient-oriented research in cancer diagnostics. She reviews the correlative science aspects of protocols for NCI’s clinical trials program. She also organizes scientific meetings and working groups with investigators outside the NIH Patricia LoRusso Patricia LoRusso, D.O. is a Professor of Medicine in Wayne State University School of Medicine’s Department of Oncology, and Director of the Eisenberg Center for Translational Therapeutics at the Karmanos Cancer Institute, Detroit, Michigan. Her primary interest is translational therapeutics with a focus on phase I clinical research and novel trial designs. She currently directs one of only 14 National Cancer Institute (NCI) U01-funded phase I sites in North America. Dr. LoRusso serves as co-chair of the NCI Cancer Therapy Evaluation Program (CTEP) Investigational Drug Steering Committee. She has also served on the education and scientific committees of the American Society of Clinical Oncology (ASCO), the scientific committee of the American Association for Cancer Research (AACR), and xxv as a parent member of the NCI’s Quick Trials Clinical Subcommittee. She has served either ad hoc or as an appointed member on multiple study sections. She has also served four years as a faculty member of the Methods in Clinical Cancer Research Workshop in Flims, Switzerland and six years as a faculty member of the Methods in Clinical Cancer Research Workshop in Vail. CO. John Martens John WM Martens is Associate Professor heading the laboratory of Genomics and Proteomics of Breast Cancer in the Dept. of Medical Oncology, Erasmus MC, Rotterdam, the Netherlands. He received his PhD (1994) in Molecular Biology at Wageningen University (NL). After being a post-doc in Molecular Endocrinology at the Erasmus MC, Rotterdam (NL) (1994–1998) and at UCSF (1998–2001), he preceded his career in 2001 in translational breast cancer research at his current department. He currently is chair of EORTC-Pathobiology group, member of the Steering Committee of the Center for Personalised Cancer Treatment, member of BCAC and various national translational research advisory committees. The major aim of his research is to determine by using a multitude of state-of-the-art high-throughput genomics and proteomics technologies which biological factors are associated with disease progression and/or the development of resistance to therapies in breast cancer. Ultan McDermott Ultan McDermott is a clinician scientist with an interest in cancer genomes and how they impact on drug response in the clinic. He is a Group Leader in the Wellcome Trust Sanger Institute as well as a practicing Oncologist at Addenbrooke’s Hospital in Cambridge. He trained as a medical oncologist and obtained a PhD in cancer biology at Queen’s University, Belfast. He was accepted for a post-doctoral research position at Harvard Medical School and Massachusetts General Hospital in 2005, where he established a high-throughput cancer cell line drug screen to identify genomic alterations that could be used in the clinic to stratify patients for treatment. His research interests are in the area of predictive biomarkers to cancer therapeutics and in vitro models of drug resistance in human cancers. He joined the Sanger Institute in 2009 as a clinical research fellow and was appointed to the faculty as a Group Leader in 2010 with the award of a Cancer Research UK fellowship. He was made a Fellow of the Royal College of Physicians in 2011. Federico Monzon Dr. Monzon is board certified in anatomical/clinical pathology and molecular genetic pathology by the American Board of Pathology. Dr. Monzon received his MD degree from the Universidad Nacional Autónoma de México xxvi and did his Pathology residency training at Thomas Jefferson University Hospital in Philadelphia and Molecular Genetic Pathology subspecialty training at the University of Pittsburgh Medical Center. He has a broad background in molecular diagnostics and pathology informatics and significant experience in the translation of novel technologies into clinical molecular tests. At the University of Pittsburgh, he led genomics studies that validated a microarray-based clinical assay for the diagnosis of tumors of unknown origin. These studies were used to support one of the first FDA cleared gene expression tests (Pathwork Tissue of Origin). Until recently, Dr. Monzon was Director of Molecular Pathology at the Cancer Genetics Laboratory of Baylor College of Medicine. Prior to Baylor, he was Director of Molecular Pathology at The Methodist Hospital in Houston, founder and head of the advisory board for iKaryos Diagnostics, and Director of the Clinical Genomics Facility at the University of Pittsburgh Cancer Institute. He recently joined Invitae, a company specialized in genetic testing. Helen Moore Dr. Helen Moore directs the National Cancer Institute’s Biospecimen Research Network program (BRN) within the Biorepositories and Biospecimen Research Branch (BBRB). The BRN encompasses extramural research programs, a Web-based biospecimen literature database, and community outreach activities including the annual BRN Symposium, “Advancing Cancer Research Through Biospecimen Science”. Dr. Moore is a Molecular Biologist with a broad background in research and development. She joined the NCI in 2006 from Celera Genomics, where she led and managed cross-functional teams to develop bioinformatics products focused on Comparative Genomics and data visualization; developed new drug targets for complex diseases using multiple approaches including genetic analysis of disease association study data, biological pathways analysis, literature mining, and genomic analysis; and contributed to the assembly and annotation of the human genome. Dr. Moore leads an NCI collaboration with the EU SPIDIA program, is a member of the ISBER Science Policy Committee, and is the Biospecimen Science section editor for the journal Biopreservation and Biobanking. Uwe Oelmueller Dr. Uwe Oelmueller is the coordinator of the European Collaborative Grant Project SPIDIA within the European Commission FP7 program. The SPIDIA consortium is working on the standardization and improvements of preanalytical tools and procedures for in vitro diagnostics. It is build by 7 public research organizations, 8 research companies and an official European standards organization. Dr. Oelmueller is a vice president RandD at QIAGEN and a management committee co-chair at the QIAGEN / BD joint venture company PreAnalytiX. He is head- Faculty’s Biosketches ing the technology center "Diagnostic Sample Preparation" within QIAGEN’s global MDx Development. The center involves technology and product development projects for clinical sample collection, preservation, storage, transport and archiving, and the isolation and analysis of human and pathogen nucleic acids. Prior to QIAGEN, Dr. Oelmueller headed a research group at the Clinical Microbiology Center, University of Goettingen, Germany, working on the AIDS disease stage relevant HIV RNA expression pattern. Dr. Oelmueller’s scientific work is presented in more than 70 patents, patent applications, scientific papers and abstracts. Kelly Oliner Kelly S. Oliner, PhD is currently a Scientific Director in the Medical Sciences – In Vitro Diagnostic Group at Amgen. Dr. Oliner has been involved in prospective testing of the KRAS predictive biomarker, next generation sequencing analysis of a phase 3 study, a circulating tumor DNA analysis aimed at describing mechanisms of acquired resistance and the demonstration that MET IHC has the potential to stratify gastric cancer patients who respond to anti-HGF antibody treatment. Most recently she has completed studies on the predictive nature of the RAS (KRAS and NRAS) biomarker. Dr. Oliner received her BA from Smith College and her PhD from the Johns Hopkins School of Medicine. Her PhD thesis was focused on the molecular genetics of colorectal cancer in the laboratory of Bert Vogelstein. Two post-doctoral fellowships at Somatix Gene Therapy and Chiron Corporation followed. Currently, Dr. Oliner focuses on leading late stage oncology biomarker and in vitro diagnostic development teams. Xavier Paoletti Xavier Paoletti, PhD, is senior statistician at the Institut Curie and co-head for the research team of clinical biostatistics at INSERM U900. He is also the referent statistician for the early clinical study group (GEP) of the French federation of anti-cancer centers and for the Innovative Therapies agains Cancer in Children European network. He is in charge of the Statistical Training Applied to Clinical Research (STARC) from the Paris VI Pierre & Marie Curie University. Xavier Paoletti developed expertises in statistical methods for early phase clinical trials with both methodological and applied works. He has also been interested in the validation of surrogate markers in gastric cancer while conducting individual patient-based meta-analyses. Finally he is closely involved in biomarkers-driven trials, such as the SHIVA trial with Dr. C. Le Tourneau and is interested in the statistical issue of detecting treatment variation in patients’ subgroups based on intermediate endpoints. Faculty’s Biosketches xxvii Mario Pazzagli Full Professor of Clinical Biochemistry and Clinical Molecular Biology, Faculty of Medicine, at the Dept of Clinical and Experimental Biochemical Sciences, University of Florence, Italy. Former member of the Executive Board of the EFLM, Former Chair of the IFCC-C-MD commission, Advisor of the CLSI Committee on Molecular Methods; Chair of the Working Group on Personalized Laboratory Medicine of the EFLM, Co-ordinator of the FP6 EU projects “Multi-National External Quality Assay (EQA) Programmes in Clinical Molecular Diagnostics based on Performance and Interpretation of PCR assay methods including dissemination and training” (contract LSH no. 504842), Member of the FP7 EU Project SPIDIA: Standardisation and improvement of generic preanalytical tools and procedures for in vitro diagnostics 2008–2011 (Contract no. 222916). National representative of the CEN/TC 140/WG 3 “Quality management in the medical laboratory”. Editor of 7 books and author of about 150 publications in International peer reviewed journals. Michael Postow Dr. Michael Postow is a physician on the faculty at Memorial Sloan-Kettering Cancer Center in the MelanomaSarcoma Oncology Service. He completed medical school at New York University School of Medicine and internal medicine residency training at Brigham and Women’s Hospital/Harvard Medical School. He then returned to New York City to pursue a fellowship in Medical Oncology at Memorial Sloan-Kettering Cancer Center where he served as Chief Fellow. During fellowship, he conducted research with Dr. Jedd Wolchok studying interactions between radiotherapy and immunotherapy (the abscopal effect) and characterizing the absolute lymphocyte count as a pharmacodynamic biomarker for ipilimumab. He has organized a prospective study to further evaluate the abscopal effect and has been involved in a number of studies investigating promising immunotherapeutic approaches for patients with advanced melanoma. Aside from medicine, he enjoys playing music, snow skiing, and sailing in the New York Harbor. Eric Polley Eric Polley, PhD, is a Mathematical Statistician in the Biometric Research Branch of the Division of Cancer Treatment and Diagnosis at the NCI. Dr. Polley received his BAin Mathematics from Saint John’s University (MN), a MS in Biostatistics from Columbia University, and a PhD in Biostatistics from the University of California, Berkeley in 2010. His primary research area involves the development and evaluation of prediction models, statistical analysis of high-throughput genomic data and oncology clinical trial design. Jordi Rodon Jordi Rodon is attending physician at the Medical Oncology Department of the Vall d’Hebron University Hospital and Clinical Coordinator of the Research Unit for Molecular Therapy of Cancer (UITM) – “la Caixa”. He obtained his specialization in Medical Oncology from the Institut Catala d’Oncologia, Barcelona. He has been a Research Fellow in the Advanced Drug Development Fellowship program at the Institute for Drug Development in San Antonio, Texas (USA), and Senior Research Fellow at the Investigational Cancer Therapeutics Department at the MD Anderson Cancer Center in Houston, Texas (USA). He joined the Medical Oncology Department of the Vall d’Hebron University Hospital in 2008 and has been Principal Investigator or Co-Investigator for more than 80 Phase I clinical trials. He is also involved in translational research derived from the clinical trials at UITM, as well as projects in Personalized Oncology as Principal Investigator of the BKM120 Stand-Up-to-Cancer (SU2C) “Dream Team”, and member of the Worldwide Innovative Networking in personalized cancer medicine (WIN) Consortium. Mei Polley Dr. Mei-Yin “May” Polley received her doctoral degree in Biostatistics from Columbia University. Her dissertation was honored The Joseph L. Fleiss Memorial Prize in Biostatistics for an Outstanding Dissertation. Upon graduation, she joined the biotechnology company Amgen as a Senior Biostatistician. In 2007, she joined the faculty of the Department of Neurosurgery at the University of California San Francisco as an Assistant Professor where she provided statistical expertise to support a broad array of brain tumor research projects. In 2010, she moved to the Biometric Research Branch at the National Cancer Institute. At present, she provides statistical leadership to both the Cancer Diagnosis Program (CDP) and the Cancer Therapy Evaluation Program (CTEP) through participation in reviews of protocols and grants involving correlative science studies. She also serves as a statistical reviewer on several correlative science committees. Her current research interests include biomarker trial designs, early phase trials and assay reliability. Dr. Polley has published in many professional journals in the field of statistics and oncology. She is an Associate Editor for Neurosurgery. Roberto Salgado Roberto Salgado, MD, PhD is board certified in Anatomic Pathology since 2006, has obtained his medical training at the University Hospital of Antwerp (Belgium) and the University Hospital in Leiden (The Netherlands). A PhD thesis was obtained working with the Translational Cancer Research Group of the AZ Sint-Augustinus Hospital/Antwerp and at the Department of Pathology at the University Hospital of Antwerp; studying the interactions of haemostasis and angiogenesis in breast cancer. His training in Anatomic Pathology and Molecular Pathology took place at the Uni- xxviii versity Hospital Antwerp, the University Hospital Leuven and at the Jules Bordet Institute. Currently he works at the Breast International Group (www.breastinternational group.org) located in the Jules Bordet Institute and at the Department of Pathology/Translational Cancer Research Group/AZ Sint-Augustinus Hospital in Antwerp. He works in close collaboration with the EORTC, having as main tasks integrating tumorbanks and molecular testing laboratories in clinical trials. He is also an auditor on Molecular Pathology/Genetic laboratories for the Federal Belgian Government. Nicolas Servant Nicolas Servant has a background in biological analysis and computer science (M.Sc.) and a strong experience in cancer bioinformatics and high-throughput analysis. In the last few years, he was mainly involved in the analysis of breast local recurrence in young women. Since 2012, he leads the next-generation analysis team of the bioinformatics platform of the Institut Curie (Paris, France). His team is today involved in the development of bioinformatics workflows for next-generation sequencing applications and provides bioinformatics support to the research and medical teams of the Institut Curie. Steven Shak Steven Shak, MD, is Executive VP Research and Development of Genomic Health, Inc, which is focused on improving the quality of treatment decisions for cancer patients. He and his colleagues have worked together with leading oncology clinical research groups to develop and commercialize the Oncotype DX® breast cancer, colon cancer, and prostate cancer assays. Dr. Shak has previously served as Senior Director and Staff Clinical Scientist at Genentech, Inc. where he led the clinical team that gained approval of trastuzumab (Herceptin®) and of the companion diagnostic HercepTest for metastatic breast cancer. In addition, Dr. Shak cloned and expressed the therapeutic human enzyme human DNase I or dornase alfa (Pulmozyme®), a mucus-dissolving enzyme that is approved worldwide for the treatment of cystic fibrosis. Dr. Shak previously held faculty positions in Medicine and Pharmacology at New York University School of Medicine. Dr. Shak has an undergraduate degree from Amherst College, an MD degree from New York University School of Medicine, and post-graduate training in medicine and research at Bellevue Hospital in New York City and the University of California, San Francisco. David Sidransky Dr. Sidransky is a renowned oncologist and research scientist named and profiled by TIME magazine in 2001 as one of the top physicians and scientists in America, recognized for his work with early detection of cancer. Since 1994, Dr. Sidransky has been the Director of the Head Faculty’s Biosketches and Neck Cancer Research Division at Johns Hopkins University School of Medicine and Professor of Oncology, Otolaryngology, Cellular & Molecular Medicine, Urology, Genetics, and Pathology at John Hopkins University and Hospital. Dr. Sidransky is one of the most highly cited researchers in clinical and medical journals in the world, in the field of oncology during the past decade, with over 450 peer-reviewed publications. He has contributed more than 60 cancer reviews and chapters and has published more than 30 articles on UCC and 150 on SCC cancers including seminal papers on genetic and epigenetic changes in these tumor types. He has pioneered the detection of various molecular alterations in human cancer and bodily fluids. Dr. Sidransky is the recipient of a number of awards and honors, including the 1997 Sarstedt International Prize from the German Society of Clinical Chemistry, the 1998 Alton Ochsner Award Relating Smoking and Health by the American College of Chest Physicians, and the 2004 Richard and Hinda Rosenthal Award from the American Association of Cancer Research. Walter Stadler Walter Stadler, MD, is an expert in prostate, kidney, bladder, and testicular cancers. He concentrates on the use of chemotherapy, immunotherapy, anti-angiogenic therapy, and molecularly targeted therapy for patients with locally advanced or metastatic disease. His research focuses on the development of new treatments for these urological cancers. Dr. Stadler’s recent research includes development of molecular and imaging markers for predicting response to various anti-cancer therapies. Dr. Stadler has authored and co-authored more than 170 articles in medical journals such as Cancer Research and the Journal of Clinical Oncology, in addition to more than 100 book chapters, reviews, letters, and editorials. He is an active member of several committees and boards, including the NCI Investigational Drug Steering Committee, the NCI Board of Scientific Counselors, medical advisory boards of the Kidney Cancer Association and the Bladder Cancer Advocacy Network, and the editorial boards of Cancer and UpToDate in Oncology, an information source for oncologists. Fred Sweep Fred Sweep (1959) is full professor of Chemical Endocrinology and head of the department of Laboratory Medicine at the Radboud University Nijmegen Medical Centre. He is board certified in Clinical Chemistry and Endocrinology. He studied Medical Biology at the University of Utrecht, where he also obtained his PhD degree (Pharmacology). He had his training as a clinical chemist at the RUNMC. Fred Sweep has published over 365 papers (Hirsch-index 43). Sweep is active member of many different national and international societies devoted to cancer Faculty’s Biosketches biomarkers. Presently, Sweep is elected president of the SKML. Sweep’s department has developed EQA programs for steroid hormone receptors and other biomarkers since 1975. His current research interests are focused on development of new antibody based assays for biomarkers in oncology with emphasis on proteases and angiogenesis. The department of Laboratory Medicine harbours all up-to-date laboratory facilities for Clinical Chemistry, Endocrinology, Haematology, (transplantation) Immunology, Blood transfusion, Paediatrics and Neurology. Frazer Symmans Dr. Symmans is Professor of Pathology at The University of Texas MD Anderson Cancer Center where he practices surgical pathology and co-directs the Breast Cancer Pharmacogenomics Program. Dr. Symmans received his medical degree from the University of Auckland before completing his residency training at Columbia University College of Physicians and Surgeons and fellowship training at MD Anderson. Dr. Symmans joined the faculty of MD Anderson in 2000. Dr. Symmans’ research is focused on breast cancer, with specific emphasis on neoadjuvant treatment trials for evaluation of chemo sensitivity and development of diagnostic tests to select the most effective treatments for individuals with breast cancer. He has adapted genomic technologies to clinical needle biopsies of breast cancer in order to use gene expression profiling to identify important genes for response to chemotherapy and, independently, to endocrine therapy; to validate gene expression tests with clinical potential; and to establish their performance in the context of clinical testing. He has published over 70 peer-reviewed articles, and has been awarded several prestigious grants in breast cancer pharmacogenomics. Sabine Tejpar Sabine Tejpar, MD, PhD is Associate Professor in the Department of Oncology, University of Leuven, Belgium. She works as a part time clinician treating GI malignancies, part time researcher (Senior Clinical Investigator of the Fund for Scientific Research- Flanders (Belgium), with a focus on basic and translational research in colorectal cancer. Main projects involve molecular sub classification of colorectal cancer, prognostic markers in adjuvant colorectal cancer and predictive markers for efficacy of EGFR inhibition. She is a member of EORTC board and chair of the EORTC Translational Research Advisory Committee (TRAC). Geraldine Thomas Gerry Thomas is a serial biobanker. She established the Chernobyl Tissue Bank (CTB: www.chernobyltissuebank. com) in 1998 which provides a platform for a systems biology approach to the mechanisms involved in radiation induced thyroid cancer and supports tissue collection for xxix international epidemiology studies. She has also developed two research tissue banks in the UK that are embedded within the NHS. Gerry is the Scientific Director for the Wales Cancer Bank (www.walescancerbank.com), which opened in 2004, and since appointment to the Chair of Molecular Pathology at Imperial College in 2008, she has revamped the Imperial College Healthcare Tissue Bank (www.imperial.ac.uk/tissuebank) to provide a flexible tissue resource to accommodate a varied portfolio of clinical studies. She sits on a number of advisory boards for tissue banking initiatives in Europe, and combines tissue banking with research interests in the molecular pathology of breast and thyroid cancer, and the effects of radiation exposure on health. Magdalena Thurin Program Director at Cancer Diagnosis Program, NCI, where her major focus has been on development of projects in the area of biomarkers for clinical application in cancer. Her specific interest is on research programs in the cutting-edge areas of molecular markers in melanoma and the immune response for prognosis and prediction of response to treatment of cancer. Prior to joining NCI in 2001 she conducted research at The Wistar Institute of Anatomy and Biology in Philadelphia focusing on characterization of carbohydrate antigens, their function in cancer progression and cancer vaccines. Her research experience includes structural and immunochemical approaches to study expression of carbohydrate structures including blood group type antigens that correlate with tumor progression. After completing her PhD and MS degree in Biochemistry at the University of Warsaw, Poland, she received postdoctoral training at the Wistar Institute in Philadelphia and subsequently began an academic career at the same institution. She has published over 70 papers in peer-reviewed journals, is an active member of many different societies devoted to cancer biomarkers and is an editor for several journals. Fernando Ulloa Montoya Fernando Ulloa Montoya, PhD, Senior Scientist, RandD, Head of Molecular Biology in Cancer Immunotherapeutics, GlaxoSmithKline (GSK) Biologicals, Rixensart, Belgium. Dr. Ulloa Montoya received his Biochemical Engineering degree from the National Polytechnic Institute in Mexico in 1998, his MS degree in 2001 from Washington State University and completed his PhD training in Chemical Engineering at the University of Minnesota Twin Cities in 2006. During his post-doctoral training at the Katholieke Universiteit Leuven, Belgium (2006–2008), he also held the position of stem cell core facility leader; in which he used transcriptional profiling for characterizing and predicting the functionality of adult stem cells. He joined GlaxoSmithKline Biologicals in 2008 and has since led a group working on discovery and development of transcrip- xxx tome based biomarkers to predict cancer patient’s response to antigen specific cancer immunotherapy (ASCI). John Welch Jack Welch, MD, PhD, is Head of Gastrointestinal and Thoracic Cancers Therapeutics in the Clinical Investigations Branch of the Cancer Therapy Evaluation Program (CTEP), Division of Cancer Treatment and Diagnosis, at the NCI. Dr. Welch came to CTEP from the European Organisation for the Research and Treatment of Cancer (EORTC), where he was a senior clinical research physician. Dr. Welch received his BS in Biology from Georgetown University, his MD degree from the State University of New York at Buffalo, and his PhD in Pharmacology from Roswell Park Cancer Institute. He completed a pediatric residency at Georgetown University, and subsequent fellowship in Pediatric Hematology-Oncology at the Children’s Hospital of Philadelphia. Paul Williams The Molecular Characterization Laboratory was established to focus on development of state of the art genomic Faculty’s Biosketches technologies for clinical research. Laboratory goals are to assist in the development and application of well characterized and validated clinical assays to support cancer patient management. One of several ongoing pilot projects is the development and implementation of massively parallel sequencing assays for selection of patients for early stage clinical trials. He has been active in the use of molecular technologies for drug target discovery. During 13 years at Genentech, he developed novel assays to support clinical studies and discover new therapeutic targets. He was the author of the first quantitative “real-time” PCR papers and contributed to the development of this powerful technology. Prior to joining FNLCR, he was a senior research group leader at Roche Molecular Diagnostics, where he led the research effort and managed two large multi-national clinical assay studies: The Microarray Innovations in Leukemia Study and collaborated with the Leukemia and Lymphoma Molecular Profiling Project. He has published over 50 manuscripts and holds over 30 issued US Patents. Kurt Zatloukal No information received. 27,000,000 The number of new cancer cases expected globally by 2030. As the Janssen Pharmaceutical Companies, we are dedicated to addressing and solving the most important unmet medical needs of our time. We are committed to making cancer a preventable, chronic or curable disease by pursuing innovative ideas wherever they are – inside or outside our company. We work hand in hand with the best scientific minds in oncology today to make cancer prevention, diagnosis and treatment accessible and affordable to patients worldwide. As Janssen, we are passionately pursuing science for the benefit of all patients. For more information please visit http://oncology.janssenrnd.com. Janssen Research & Development, LLC SPECTAcolor An innovative path to new colorectal cancer treatments EORTC has connected over 30 leading clinical centers across Europe for the first pan-European Biomarker Screening Platform dedicated to patients with advanced colorectal cancer (CRC). SPECTAcolor ✓ Increases opportunities for CRC patients to access clinical trials with new molecularly defined approaches ✓ Pan-European network of top level recruiting clinical centers ✓ Exhaustive molecular characterization of CRC patients ✓ Certified labs for diagnostic assays ✓ Translational research-based clinical trials SPECTAcolor is funded by a corporate social responsibility initiative of Alliance Boots The future of cancer therapy www.eortc.org/contact SPECTAcolor is a trademark of the EORTC. All rights reserved. spectacolor.indd 1 17/09/2013 16:13:24 European Journal of Cancer (2013) 49, Supplement 4, S1–S38 Abstracts Markers in Cancer: A joint meeting by ASCO, EORTC and NCI, 7–9 November 2013, Brussels, Belgium Speakers’ Presentations Topic 1: New trial concept and challenges for the future SP001 Biostatistical challenges for the design of biology-driven trials X. Paoletti. Biostatistics, Curie, Paris, France An important objective of early phase-trials (phase I and phase II) is to understand the mechanisms of action of a new compound and to identify potential activity that would justify going into larger trials. Biology then plays a central role but using biological measurements to drive the design raise numerous issues. In this communication, we will focus on challenges of biology-driven early phase trials. In phase I clinical trials, three main types of dose finding trials that incorporate translational research can be identified. We will see what is feasible and what type of information can be expected in each situation. We will emphasize the importance of having reliable sensitive endpoints and assays and highlight the risk of erroneous conclusions at this very early stage of the clinical development. We will then focus on proof of concept phase II designs and show that randomization is a key component for evaluating the added value of biology measures and to distinguish between prognostic factors and factors predictive of the response to treatment. The SHIVA trial that compares standard practice to the administration of targeted agent based on molecular profile of the tumour, will serve as an illustration of the importance of an adequate design to obtain reliable conclusions. We will review several possible designs for such a trial and detail the strength and limits of each of them. In particular, we will emphasize the importance of consolidating our knowledge on the biomarkers in order to design powerful phase III trials. Multistep designs appear as promising yet delicate design. We will also plead for simple well-designed trials addressing well-defined objectives. Again, we will insist on the importance of sensitive endpoints to detect variations of activity according to biomarkers and to the necessary steps to go from an exploratory research to a validated biomarker. its tumour. The last decade witnessed the development of high-throughput technologies such as microarrays, and more recently next-generation sequencing, which paved the way to personalized medicine in the field of oncology. While the cost of these technologies decreases every day, we are facing an exponential increase in the amount of data generated. Our ability to quarry this genetic information in daily practice relies strongly upon the availability of a performant bioinformatics system that assist in the translation of knowledge from the bench to molecular targets and diagnosis. Clinical trial and routine diagnosis constitutes different approaches in clinical practice, yet they both require a bioinformatics environment, which ensures the integration and the traceability of all samples’ information as well as the processing and analyses of the genomic data, and which implements well-defined procedures for workflow management and decision making. A knowledge integration system was developed at the Institut Curie to facilitate the integration of the clinical and pathological data, and to track in real-time the processing of individual samples. Checkpoints from the biopsy, to the technological platforms and to the final analysis results are also installed. In addition, the bioinformatics environment must ensure secure storage of large datafiles as well as access to high performance computing enabling rapid analysis of data and secure results delivery within a few hours. For the molecularly-based analysis or profiling, the bioinformatics challenges include establishing validated computational pipelines for identifying reliably genomic alterations and mutations, with adequate quality control. Currently, several methods are available but no consensus on a standard computational tool dedicated to such questions was reached yet. User-friendly interface are also needed to allow physicians to control the analysis workflow and to facilitate the interpretation of individual patient data. Personalized medicine projects are now a reality. Thus, real-time medical informatics systems that integrate clinical and genomic data, as well as efficient sample traceability are today mandatory. Developing high-throughput technologies like sequencing into validated diagnostics that can be used for patient care is still a major challenge, which can be addressed by sharing expertise in cancer genomics, bioinformatics, pathology and clinical genetics. SP003 Pathology challenges for biology-driven trials: The Ki67 experience SP002 Bioinformatics analysis for real-time applications N. Servant 1 , P. Hupé 1 , M. Kamal 2 , C. LeTourneau 3 , E. Barillot 1 . 1 INSERM U900 – Plateforme de Bioinformatique, Institut Curie, Paris, France; 2 Département de Transfert, Institut Curie, Paris, France; 3 Département de Médecine oncologique, Institut Curie, Paris, France Genetically guided medicine requires the delivery of individually adapted medical care based on the genetic characteristics of each patient and of 0959-8049/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. M. Dowsett. Academic Department of Biochemistry, Royal Marsden Hospital, London, United Kingdom Ki67 is a highly favoured marker of proliferation in multiple malignant tissues because of its ease of measurement by immunohistochemistry (IHC) and general analytic ruggedness of the assay. It has been of particular interest in breast cancer where it has multiple potential applications. Of greatest clinical interest is its possible role in assessing prognosis and likelihood of benefit from specific treatments. So far as prognosis is concerned we have shown S2 that it is best assessed as part of a panel of other IHC markers including ER, PgR and HER2 to form an index termed IHC4 and this is most appropriately applied by integration with standard clinicopathological parameters. In the research context Ki67 has been validated as a tool for assessing the comparative efficacy of endocrine treatments in the presurgical context and as such is being applied as a primary end-point in many biology-driven trials particularly where proliferation is a direct target of the therapy. In addition, a 4,000 patient clinical trial called POETIC (PeriOperative Endocrine Therapy for Individualised Care) in ER+ breast cancer is in follow-up in part to determine whether measuring Ki67 during endocrine treatment is a better index of long-term outcome than the conventional approach to measuring it before treatment. Measurement of Ki67 during or at the end of neoadjuvant therapy (endocrine or chemotherapy) is being employed to plan modified treatment on the basis of the residual risk of recurrence. All of these applications require a standardized approach to analysis and scoring that is not currently present. To rectify this guidelines for harmonization of methodology have been published (Dowsett et al., JNCI, 2011; 103:1–9) and a series of interlaboratory studies has been designed. Until these report valid between-laboratory comparisons of data are limited in value. Although Ki67 remains a very helpful outcome measure within individual research studies, the direct application of cut-offs for clinical management can be highly misleading unless performed by laboratories who have their own reference data or have cross-validated against the methodology of a group that has such data. As we move forward it will be important to recognize that different levels of precision and accuracy are demanded by the many potential uses of Ki67. Topic 2: Biology-driven clinical trials SP004 Rationale of the design of the WINTHER trial J. Rodon. Vall d’Hebron Institute of Oncology, Barcelona, Spain Recent advances in diagnostics and targeted therapies during the last decade have changed how Oncology is viewed. Stratified Medicine has emerged from the accumulated evidence garnered from matching targeted therapies with tumor molecular aberrations. Concomitantly, current knowledge derived from large-scale, massively parallel sequencing technologies, and global research initiatives have illuminated the utility of understanding the molecular basis of cancer through genome analysis. In addition, some pilot studies are transforming the way we analyze tumor tissue molecular aberrations, the design of clinical trials and the measurement of treatment efficacy. Taken together, these pilot studies are paving the way for designing clinical trials that empirically test the concept of Personalized Cancer Medicine. The WinTHER is an innovative phase 2 clinical trial that will assess the potential of selecting targeting therapies according to the tumor biology of patients. The trial will explore a rational choice of therapeutics and their efficacy beyond current limitations. From each patient’s biopsy of the tumor (or metastasis) and normal tissue, a complete biological analysis of DNA, RNA and microRNA will be undertaken. The choice of therapy will be rationally guided either by matching actionable targets found in the tumor analysis (matching drug and molecular alterations, arm A) or the tumor gene expression and predicted sensitivity of the drug (matching differentially expressed genes between tumor and normal tissue with drugs, arm B). In this presentation, I will describe the trial design and methods and how we incorporated many of the lessons learned from prior experiences in Personalized Cancer Medicine. SP005 Rationale of the design of the I-SPY trial A. DeMichele 1 , D. Yee 2 , N. Hylton 3 , L. Vant’Veer 4 , W.F. Symmons 5 , J. Perlmutter 6 , J. Lyandres 7 , S. Davis 7 , M. Buxton 7 , D. Berry 8 , L. Esserman 9 . 1 Abramson Cancer Center, University of Pennsylvania, Philadelphia, USA; 2 Masonic Cancer Center, University of Minnesota, Minneapolis, USA; 3 Department of Radiology, University of California, San Francisco, USA; 4 Helen Diller Cancer Center, University of California, San Francisco, USA; 5 Department of Pathology, MD Anderson Cancer Center, Houston, USA; 6 507 N. 5th Ave, Gemini Group, Ann Arbor, USA; 7 Department of Surgery, University of California, San Francisco, USA; 8 Department of Biostatistics, MD Anderson Cancer Center, Houston, USA; 9 Helen Diller Cancer Center, University of California, San Francisco, USA New approaches to drug development are needed to lessen the time, cost and Speakers’ Presentations resources to identify and optimize active agents. Strategies to accelerate drug development include testing drugs earlier in the disease process such as the neoadjuvant setting, in which the surrogate short-term endpoint, pathological response (pCR), may identify active agents and shorten the time to approval of both efficacious drugs and biomarkers identifying patients most likely to respond. I-SPY 2 is a multicenter, phase 2 trial using adaptive randomization within biomarker subtypes to evaluate a series of novel agents/combinations when added to standard neoadjuvant weekly paclitaxel (T) ×12 (with trastuzumab if Her2+), followed by doxorubicin & cyclophosphamide (AC) q2–3 wk ×4, for women with high-risk stage II/III breast cancer. The primary endpoint is pathologic complete response (pCR) at surgery. Participation is limited to patients with sufficiently poor prognosis (risk of early recurrence) with standard therapy to justify exposure to investigational agents; thus those with tumors <2.5 cm or hormone receptor (HR)+/MammaPrint (MP) low biomarker profiles are not eligible. Agents selected for inclusion in I-SPY2 must have adequate safety information from phase Ib studies in combination with a taxane at a dose and schedule approximating the I-SPY2 regimen. I-SPY2 safety data are collected electronically through a web-based application (“TRANSCEND”) developed specifically for the trial in collaboration with the NCICB. The I-SPY2 DSMB meets monthly to review safety and outcome data, providing the opportunity to conduct frequent monitoring for unexpected safety signals, and constantly re-evaluate the risk/benefit ratio for a given drug or combination in the trial. The study is designed to identify and “graduate” regimens that have ≥85% Bayesian predictive probability of success (statistical significance) in a 300-patient biomarker-linked Phase 3 neoadjuvant confirmatory trial defined by HR & HER2 status & MP “signature”. Experimental regimens can “graduate” in at least 1 of 10 possible signatures, with a maximum number of 120 patients enrolled per regimen. As of August 2013, over 400 women have been enrolled to the trial and randomized to T/AC with or without one of seven investigational therapies from five different pharmaceutical companies, including veliparib/carboplatin, neratinib, AMG-386, ganitumab/metformin, pertuzumab, pertuzumab/ado-trastuzumab emtansine (T-DM1) or MK-2206. Whether any of these investigational therapies have been graduated or dropped for futility has not been announced. The I-SPY3 TRIAL, currently in development, will provide a platform for drugs graduating from I-SPY2 to be definitively tested in a randomized phase 3 trial with primary endpoints of both pCR and event-free survival. SP006 Rationale of the design of the SHIVA trial C. Le Tourneau 1 , M. Kamal 2 , E. Mitry 3 , A. Goncalves 4 , N. Isambert 5 , C. Gavoille 6 , O. Tredan 7 , J.P. Delord 8 , M. Campone 9 , X. Paoletti 10 . 1 Medical Oncology, Institut Curie, Paris, France; 2 Translational Research, Institut Curie, Paris, France; 3 Medical Oncology, Institut Curie, Saint-Cloud, France; 4 Medical Oncology, Institut Paoli Calmettes, Marseille, France; 5 Medical Oncology, Centre Georges-Francois Leclerc, Dijon, France; 6 Medical Oncology, Centre Alexis Vautrin, Nancy, France; 7 Medical Oncology, Centre Leon Berard, Lyon, France; 8 Medical Oncology, Centre Claudius Regaud, Toulouse, France; 9 Medical Oncology, Centre Rene Gauducheau, Nantes, France; 10 Biostatistics, Institut Curie, Paris, France Two recent studies suggest that a histology-independent approach consisting in selecting molecularly targeted agents based on the molecular profile of patients’ tumors, whatever the tumor location and histology are, improves patients’ outcome [Von Hoff et al., 2010; Tsimberidou et al., 2012]. However, the lack of randomization versus standard of care in these studies did not allow drawing robust conclusions. The SHIVA trial is a multicentric randomized proof-of-concept phase II trial comparing molecularly targeted therapy based on tumor molecular profiling versus conventional therapy in patients with any type of refractory cancer. The primary endpoint is progression-free survival (PFS). The molecular profile performed on a mandatory biopsy includes the assessment of (1) hot spots mutations using the AmpliSeq cancer panel on Ion Torrent/PGM (Life Technologies), (2) gene copy number alterations using Cytoscan HD/Affymetrix, and (3) expression of estrogen, progesterone and androgen receptors by immunohistochemistry on formalin-fixed sample. The algorithm used by a Molecular Biology Board (MBB) to guide treatment in the experimental arm is presented in Table 1. The efficacy analysis will be performed on 200 randomized patients. Randomization is stratified according to the molecular pathway selected for drug selection as well as on the patients’ prognosis using the Royal Marsden Hospital score for phase I cancer patients. A cross-over is proposed at disease progression. Feasibility results on the first Speakers’ Presentations S3 Abstract SP006 – Table 1 Molecular abnormalities KIT, ABL, RET PI3KCA, AKT1 AKT2,3, mTOR, RAPTOR, RICTOR PTEN STK11 BRAF PDGFRA/B, FLT-3 EGFR HER-2 SRC EPHA2, LCK, YES ER, PR AR Type of molecular abnormality Molecularly targeted agents Activating mutation or amplification Activating mutation or amplification Amplification Inactivating mutation and LOH Inactivating mutation and LOH Activating mutation or amplification Activating mutation or amplification Activating mutation or amplification Activating mutation or amplification Activating mutation or amplification Amplification Protein expression >10% Protein expression >10% Imatinib Everolimus Everolimus Everolimus Everolimus Vemurafenib Sorafenib Erlotinib Lapatinib + Trastuzumab Dasatinib Dasatinib Tamoxifen (or letrozole if contra-indication) Abiraterone ER = Estrogen receptor; PR = Progesterone receptor; AR = Androgen receptor; LOH = Loss of heterozygosity. 100 included patients were presented at the 2013 ESMO meeting. More than 300 patients have been included as of September 2013. We present here the rationale for the design of the SHIVA trial and discuss its advantages and drawbacks in the context of personalized cancer medicine. SP007 Molecular profiling based assignment of cancer therapy (MPACT) S. Kummar 1 , A. Chen 1 , J. Lih 2 , M. Williams 2 , L. Rubinstein 1 , B. Conley 1 , J.H. Doroshow 1 . 1 National Cancer Institute, Maryland, USA; 2 SAIC-NCI-Frederick National Laboratory for Cancer Research, Frederick, MD, USA Targeting unique aberrations within the tumor is the goal of personalized medicine. Assignment of therapy based on underlying genetic aberrations is guided by the hypothesis that patients with mutations in a given pathway in their tumor are more likely to derive clinical benefit if treated with agents that target that pathway than if treated with agents that do not. Multiple trials and trial designs, the so-called umbrella and basket trial designs, have been proposed based on this hypothesis. However, apart from a few examples of successfully targeting such mutations, more data from randomized trials is required to establish the role of molecular profiling based assignment of cancer therapy for patients with refractory solid tumors. The US National Cancer Institute (NCI) is initiating a randomized pilot trial that aims to establish whether patients with advanced cancers who have no treatment options with proven benefit, and with tumor mutations in one of 3 genetic pathways (DNA repair, PI3K, or RAS/RAF) are more likely to derive clinical benefit if treated with agents targeting that pathway than if treated with agents that do not. Each patient will be randomly assigned to receive the recommended Phase II dose of either a study drug identified to work on their tumor’s mutation in the given pathway, or an agent from the complementary set not identified to work on the mutations of interest (MOIs). Patients enrolled on study will have a tumor biopsy sequenced in a CLIA-certified lab for specific MOIs in the DNA repair pathways, PI3K pathway, and the RAS pathway. Therapy will be assigned based on defined criteria following identification of an actionable mutation of interest. The objectives of the trial are to compare the response rate (CR+PR) and progression-free survival (PFS) for treatment with agents chosen based on the presence of specific mutations in patient tumors with the response rate and PFS for treatment with agents randomly chosen from the complementary set of agents not identified to work on the MOIs. Patients will be randomized 2:1 in favor of the target-directed arm; the trial will discriminate between tumor response rates of 20% vs. 5% and median PFS of 3.6 vs. 2 months Topic 3: Progress and implications of novel technology for clinical trials SP008 Clinical reporting of NGS data D. Eberhard. Pathology Pharmacology and Lineberger Comprehenesive Cancer Center, University of North Carolina, Chapel Hill, USA Next-generation sequencing (NGS), also known as massively parallel sequencing, presents great opportunities for advancing a tumor genomics-based approach to clinical tumor diagnostics, especially companion diagnostics and clinical development of targeted therapeutics. The various technologies of massively parallel sequencing potentially allow assessment of a variety of genomic alterations across many genes or even an entire genome, at an economic cost that is now competitive with other established clinical diagnostic tests. Medical research and diagnostic laboratories are now adopting NGS platforms for various purposes ranging from preclinical research on tumor genomics, through clinical research often related to targeted therapy development, into clinical diagnostics and personalized medicine schemes. Each of these settings have particular and differing needs that must be considered in how to deliver NGS results that are relevant and useful. Various professional and regulatory organizations are working on developing recommendations for the clinical implementation and reporting of NGS results. This presentation will give a background review, describe our efforts at UNC to implement a clinical research program using NGS (UNCseqTM), and present new recommendations from an NCI working group group regarding the use of “Omics”-based predictors in NCI-sponsored clinical trials. SP009 Are we ready for genomes in the clinic? U. McDermott. Cancer Genome Project, The Wellcome Trust Sanger Institute, Cambs, United Kingdom Cancer is a disease of the genome. Mutations in genes that confer a selective advantage to the cancer cell are perpetuated through subsequent generations of daughter cells and result in clonal expansion of that cell. In addition to being responsible for tumourigenesis, it is also clear that in some cancers the presence of specific gene mutations has major effect on treatment response and survival in the clinic. The most celebrated recent example of this is the V600E mutation in the BRAF oncogene. Melanoma tumours harbouring this mutation are exquisitely sensitive to small molecule inhibitors of BRAF, and patients are now routinely screened for this mutation in melanoma clinics prior to treatment stratification. However, it is likely that mutations in other genes within the same tumour will also play a role in treatment response and multi-gene signatures will become increasingly important as predictive tools. Analysis of the 10,000 tumours exome sequenced world-wide is enabling us to define the cancer genes within each tumour type, and to restrict the design of a targeted gene screen (TGS) to these “drivers”. Thus, today a panel of 300–400 genes would encapsulate all genes recurrently mutated, amplified or deleted in solid tumours and could be delivered at a fraction of the cost of a whole genome. We have demonstrated that such a targeted gene screen can deliver high depth NGS data from archived FFPE tissue samples and using low amounts of DNA. Although there is a significant interest in the application of whole genome sequencing in the clinic, we would argue that there are technical, financial and logistical reasons why this is not the best use of current healthcare resources and that a TGS approach adopted systematically across all cancer patients would be truly transformative for clinical practice. SP010 Application of next generation sequence for retrospective biomarker analyses in tumors samples from Phase II/III clinical trials S. Khambata-Ford. Oncology Clinical Development, Novartis Pharmaceuticals Corp, East Hanover, USA Next Generation Sequencing (NGS) has been applied increasingly in cancer S4 genomics research over the last seven years. Most of the reported results are genetic landscapes generated on tumor samples collected outside clinical trials or from early phase oncology trials. Application of this technology in large global Phase 3 trials provides an excellent opportunity for advancing personalized healthcare by identifying potential predictive markers of therapeutic benefit. This presentation showcases an exploratory retrospective biomarker analysis in BOLERO-2, a Phase 3 study of everolimus (mTOR inhibitor, Afinitor® ) in Hormone Receptor (HR) positive, Her2-negative advanced breast cancer. In BOLERO-2, EVE plus exemestane (EXE) more than doubled progressionfree survival (PFS) versus EXE alone, with consistent benefit in the overall population and in all clinically defined subgroups. Potential predictive biomarkers for EVE benefit were assessed in a panel of 182 cancer-related genes using NGS. NGS data from formalin-fixed archival tumor specimens were assessed for exon sequence and gene copy number variations. Univariate and multivariate Cox models were used to evaluate correlations with Progression Free Survival (PFS). Patients with NGS data (>250× coverage; n=227; 157 in EVE+EXE; 70 in EXE alone) had comparable baseline characteristics and PFS outcomes versus the overall trial population (hazard ratio for PFS=0.40 and 0.45, respectively). Treatment benefit with EVE was mostly maintained in subgroups defined by a single altered gene such as PIK3CA, CCNDI, FGFR1 with a >10% rate of alteration. Patients with no or only 1 genetic alteration in key signaling (PI3K, FGFR) or cell cycle pathways had greater PFS benefit from EVE than patients whose tumors contained multiple alterations. This is the first global registration trial in which predictive biomarkers of therapeutic benefit were explored by correlating broad genetic variations with clinical efficacy. It demonstrated the feasibility of using NGS and the power of large scale correlative analysis in a randomized Phase 3 trial. The observations suggest the potential interplay of multiple oncogenic pathways in determining optimal therapeutic benefit from EVE in HR+, HER2− advanced breast cancer. These exploratory results also point to potential novel rational targeted therapy combinations that will be further investigated. Topic 4: Regulatory issues SP011 Translational scientists – Overview of process of taking an assay fropm the lab to the clinic J. Martens. Medical Oncology, Erasmuc MC Cancer Institute, Rotterdam, The Netherlands Genomics research is expanding our understanding of cancer biology in general as well as our insight in clinical relevant aspects of cancer progression such as metastatic potential and therapeutic vulnerability. As a results research on prognostic and predictive markers is a vivid field and the number of diagnostic markers is rapidly expanding leading a multitude of novel and potentially informative and clinically useful biomarkers. Rather than measuring these as single markers in an uniplex assay, in the future one or more multiplex assays will likely be used to guide clinical decisions. In this session we will hear the experts view on the regulatory aspects of the implementation of next generation diagnostics in the clinical arena. To set the stage I will discuss the standard route a biomarker assay has to travel before it can be used for clinical decisions. Aspects which mainly involve providing solid evidence defined as Level Of Evidence (LOE) from category 5 till 1 will be discussed as well as various aspects of proper measuring such biomarker as we as to ensure that the standards of those assays are maintained over a long period time. As examples I will use the prognostic markers uPA/PAI-1 developed by the EORTC Pathobiology group. This marker panel has travelled all the way from being discovered in a small retrospective cohort to independent multicenter validation and prospective testing and is now listed in the ASCO and St Gallen guidelines for the clinician to guide them for taking decisions in patients visiting the clinic in the current daily routine. Other examples being discussed include the current prognostic gene expression signatures which are following or have travelled a similar route towards clinical application. Also examples of predictive markers will be given. In the end, and to set the stage for the discussion, the impact of NGS is discussed likely leading to an explosion of informative markers moving towards clinical application and how this might impact the current practice of biomarker implementation. After this presentation, experts in the field which include a clinician, a biologist, a representative from a regulatory body and from a diagnostic company and a statistician will give their professional view on the most critical issues, in their opinion, relevant for the rapid advancement Speakers’ Presentations of the plethora of potentially useful markers currently being discovered from the bench to the bed-side. SP012 EMA perspective on approval issues I. Gravanis. European Medicines Agency (Bio)Markers may serve a variety of functions in oncology drug development and ultimately authorisation. The most prominent function is as a means to determine patient populations expected to benefit or not benefit from a new drug. This is commonly linked to the fact that novel drugs often belong to the class of “targeted agents”, i.e. they act upon specific biological pathways/functions implicated in cancer. However, the approval of numerous targeted therapies in the recent years has not been granted in biomarkerdefined patient populations. This discrepancy is being attributed to the relative non-specificity of targeted agents acting promiscuously on multiple pathways, so that no single marker can be predictive of drug activity while the development of a marker-signature involving multiple ones is challenging. At the same time, disease biology is either complex or evolving in the course of the disease, so that targeted agents are often not impressively effective or resistance to them develops quickly. Systems approaches promise to address this lack of understanding of the static or evolving disease, but concrete examples of this have not reached drug regulatory authorities. The latest EU regulatory guidance on anticancer drug evaluation (EMA/CHMP/205/95/Rev.4) addresses the use of biomarkers in guiding oncology drug development. It encourages the use of biomarkers in early phases of drug development to inform PK/PD relationships and correlations. It also assumes that, irrespective of pharmacological class, entrance into clinical development of new molecule today is guided by translational research. This means that in most cases there are hypotheses to be tested and candidate biomarkers available. However, in practice this is not the case so that the majority of drugs currently reaching not only confirmatory trials but even approval do not have biomarkers defining their target population, while historically such use of biomarkers has been necessitated by lack of adequate efficacy in an unselected population. This is not meant to say that activity and efficacy in the biomarker-negative population should be overlooked when this exists and holds promise of benefit in these patients. Use of biomarkers to define eligibility in confirmatory trials ultimately defines the target population upon approval. However, retrospective biomarker analyses may also lead to restrictions to the target population when appropriate and when the benefit/risk balance is not considered positive in the unselected population. Determination of biomarker status is crucial and, although approval of relevant diagnostics in the EU is separate from drug approval, a diagnostic assay complying with the relevant EU legal requirements, as appropriate, should be available at time of drug authorisation. Moreover, the adequacy of the available diagnostic assays to correctly identify patients eligible for treatment and the feasibility of its application in wide clinical practice is taken into consideration when drug approval decisions are made. Finally, EU guidance also opens the possibility towards biomarker-defined authorisations across histologies, provided appropriate conditions are met. SP013 FDA view on approval issues and new guidelines R. Becker No abstract presentation received. SP014 R&D’s view on obstacles to development of diagnostics for clinical development S. Shak. Genomic Health Inc., Redwood City, USA To develop and clinically validate a new HER2-specific monoclonal antibody for breast cancer treatment, a rigorous multi-step drug development program was required. Similarly, to develop and clinically validate a new quantitative RT-PCR gene expression assay to guide breast cancer chemotherapy treatment planning, it was necessary to bring the rigor of drug development to the process for cancer diagnostic development. In both instances, it was critical to optimize and standardize new technologies, antibody humanization for trastuzumab and precise, reproducible RNA quantitation from tumor blocks for Oncotype DX (detailed data on the performance characteristics of the Speakers’ Presentations Oncotype DX assay will be presented). In both instances, it was also equally important that academia, industry, advocacy, and regulatory authorities collaborate to ensure that the studies were well-designed and successfully implemented. It requires a team with all accountable, and the skills, processes, and resources to perform the early exploratory studies and the later controlled validation studies. Most importantly, tests must be “Fit for Purpose” with evidence relevant to that specific purpose, with consistent results across multiple well-designed studies and value beyond traditional measures. Multiple well-designed clinical studies using a standardized quantitative assay system can successfully provide the evidence required by physicians to use the genomics of individual tumors to individualize cancer treatment. SP015 Clinican’s view TBC No abstract presentation received. SP016 Statistical issues on biomarker ascertainment and power to detect interactions J. Bogaerts. EORTC Headquarters, Brussels, Belgium Biomarkers may be construed from different perspectives: diagnosis, prognosis, or various predictive purposes. Predictive biomarkers can identify classes of patients that are specially sensitive to a particular treatment, or the opposite, or that have no need for a (class of) treatment. They could also identify classes of patients for whom currently no good treatment options are available. At the statistical level, prognostic analyses have better properties than predictive questions. The latter, while conceptually not complex, do involve interaction questions: a comparative treatment effect needs to be proven to be different in one group as compared to the treatment effect in the other group. Imprecision in biomarker assessment (i.e. assay performance) can lead to frightening loss of effect size and power in such analyses. We will discuss mathematical repercussions of various types of predictive biomarker questions, and try to indicate settings where the full penalty of a statistical interaction may not need to be endured. From the statistical point of view, the key point in biomarker development is its purpose and intended usage. This will drive the design of the trials to be undertaken, and the analysis path to follow. Topic 5: Circulating / Imaging biomarkers SP017 Circulating tumor cells (CTCs) and circulating DNA to monitor tumor heterogeneity in clinical trials S. Jeffrey. Surgery, Stanford University, Stanford, USA Dr. Jeffrey will discuss circulating tumor cells and circulating tumor DNA in the context of intratumoral heterogeneity, single cell analysis, and applications related to cancer monitoring and drug selection. SP018 PET imaging to monitor tumor heterogeneity in clinical trials E. de Vries, M. van Kruchten. Medical Oncology, UMCG, Groningen, The Netherlands Increasingly cancer patients are treated based on tumor characteristics. Typically this is done through the use of tumor biopsies, which may be performed on archival tissue that does not represent the contemporary status of the tumor. Even with a recent tissue sample, there are still problems of heterogeneity among lesions and within lesions. Heterogeneity plays a role in limiting the efficacy of targeted therapy and is a barrier to precision medicine. Given this heterogeneity, and the fact that often not all lesions can be biopsied, other supporting techniques are required. Molecular radionuclide imaging with positron emission tomography (PET) can potentially fulfill this task. Most clinical molecular imaging data are gathered on the visualization of general processes, such as glucose metabolism with 18 F-fluorodeoxyglucose (FDG) and DNA synthesis with 18 F-fluoro-L- S5 thymidine PET-imaging. But PET imaging can also visualize other important tumor cell characteristics, like hormone receptors, growth factors and growth factor receptors. Visualization of the estrogen receptor (ER) with 18 F-fluoroestradiol (FES) proved to be feasible in breast cancer patients. Heterogeneity of ERexpression is shown by both FES-positive and -negative lesions present in 15–47% of patients with an ER-positive primary tumor. Furthermore, up to a 6-fold difference in quantitative FES-uptake can be observed among metastases within individual patients. Finally, dual imaging with FDG-PET and FES-PET showed that FES/FDG ratios varied greatly across patients. The androgen receptor (AR) was imaged with 18 F-fluoro-5α-dihydrotestosterone (FDHT) in prostate cancer patients. FDHT-PET showed that MDV3100, an AR antagonist, that blocks androgens from binding to the AR, substantially reduced FDHT binding (Scher H et al. Lancet 2010). HER2 can be visualized with 89Zr-trastuzumab-PET. In metastatic breast cancer patients quantifiable tracer uptake in liver, bone and brain lesions was seen. Within patients, higher 89Zr-trastuzumab uptake was observed in liver lesions, compared to bone and brain lesions. Furthermore, tracer uptake of liver, bone and brain lesions varied within and across patients. With 89Zr-bevacizumab-PET, visualizing VEGF-expression, we showed in renal cell carcinoma patients a wide range of 89Zr-bevacizumab uptake by tumor lesions, a reduction in uptake by bevacizumab treatment and a wide range in uptake changes upon sunitinib treatment. Also, for the anti-VEGF antibody HuMV833 a heterogeneous distribution and clearance was observed between and within tumors using 124 I-labeled HuMV833 (Jayson GC et al. J Natl Cancer Inst. 2002). Thus, new molecular imaging can identify the levels of a specific tumor target across the entire body over time, can provide new mechanistic and pharmacological insights and can contribute to distinguish patients most likely to benefit from a specific treatment. Topic 6: Predictive markers for immunotherapy SP019 Monitoring response to anti CTLA-4 therapy M. Postow. Medicine, Memorial Sloan-Kettering Cancer Center, New York, USA Immunotherapy directed against cytotoxic T lymphocyte antigen-4 (CTLA-4) has improved overall survival (OS) for patients with melanoma in two randomized phase III trials and is being evaluated for additional malignancies. Pre-treatment or on-treatment biomarkers, reflective of the mechanistic activity of CTLA-4 blockade, are urgently needed to identify patients most likely to benefit. The investigation of immunotherapeutic biomarkers is conceptually distinct from investigations of biomarkers pertaining to targeted agents which typically interact directly with intratumoral oncogenic pathways. Since CTLA-4 blocking antibodies target lymphocytes, we have explored the absolute lymphocyte count (ALC) as a pharmacodynamic biomarker for ipilimumab. Immunosuppressive myeloid derived suppressor cells (MDSC) inhibit lymphocytes, and additional data suggest MDSC quantity is also associated with CTLA-4 therapeutic outcomes. Other emerging, novel possible biomarkers for CLTA-4 therapy will be discussed as will plans for prospective evaluation. SP020 Predictive gene signature (GS) in MAGE-A3 antigen-specific cancer immunotherapy F. Ulloa Montoya. R&D, GSK Vaccines, Rixensart, Belgium Antigen-specific cancer immunotherapeutics aim at educating the immune system to treat cancer. The candidate cancer immunotherapeutic used here is a combination of the MAGE-A3 tumor-specific antigen with an immunostimulant designed to enhance immune response. Two Phase II trials yielded signs of activity for this cancer immunotherapeutic in metastatic melanoma and NSCLC adjuvant setting (NCT00086866–NCT00290355). Gene expression profiling by microarray analysis was performed on fresh frozen tumor samples taken before treatment and used to identify biomarkers potentially predictive of benefit from Antigen specific cancer immunotherapy treatment. Such predictive biomarkers were initially identified in Phase II melanoma patients and then tested in NSCLC Patients. S6 In the metastatic melanoma patients, a set of 84 genes was identified whose expression was potentially associated with improved clinical response. A classifier based on gene-expression data was cross-validated and showed a greater overall survival in the population of patients whose tumor presented the GS: median OS was 16.2 months (95% CI, 9.0 to 20.0 months) in the GS-negative population and 29.0 months (95% CI, 20.5 to 40.2 months) in the GS-positive population. The genes identified were mainly immune-related, involving interferon gamma pathways and specific chemokines, suggesting that their pre-treatment expression may influence the tumor’s immune microenvironment and the patient’s clinical benefit. The predictive value of the melanoma signature in NSCLC was corroborated in samples from the proof-of-concept Phase II study in adjuvant NSCLC patients. The disease-free interval, primary endpoint of the Phase II study, for the overall population and for the patients whose tumor presented the predictive “metastatic melanoma” GS showed that the relative improvement in the risk of recurrence upon treatment with the MAGE-A3 Cancer Immunotherapeutic was increased in the patients with the predictive GS as compared to the overall population: HR GS(+) patients = 0.42 (95% CI [0.17;1.03], HR GS(−) patients = 1.17 (95% CI [0.59; 2.31]), while the respective HRs were 0.85 (95% CI [0.50; 1.43]) in the 157 patients analyzed for GS and 0.78 (95% CI [0.49; 1.24]) in the entire study population of 182 patients). Furthermore, this GS classification did not show a difference in DFS between the GS+ and GS− patients in the placebo group (HR=1.23 (95% CI: [0.51; 2.98], suggesting that this signature is not prognostic in this patient population. In conclusion, an 84-gene GS appeared associated with improved clinical response to the MAGE-A3 immunotherapeutic in metastatic melanoma and in resected NSCLC (primary tumor sampling). Phase III trials to confirm these findings prospectively are ongoing. Reference: [1] Ulloa Montoya F, Louahed J, Dizier B, et al. Predictive gene signature in MAGE-A3 antigen-specific cancer immunotherapy. J Clin Oncol 2013; 31(19):2388–95. doi: 10.1200/JCO.2012.44.3762. Epub 2013 May 28. SP021 Bone marrow minimal residual disease (MRD) was the strongest predictor of survival from high risk metastatic neuroblastoma (NB) following anti-GD2 immunotherapy, when tested in multivariate models that include FcR polymorphism and missing ligand for inhibitory killer-immunoglobulin-like receptor (KIR) N. Cheung, I. Cheung, D. Kuk, I. Ostrovnaya, S. Modak, K. Kramer, B. Kushner. Department of Pediatrics, Epidemiology-Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, USA Background: Immunotherapy using anti-GD2 antibody is now a standard of care for children with high risk metastatic neuroblastoma. Identifying predictive markers of response and survival can provide a better understanding of the underlying immunobiology and help optimize efficacy among patient subgroups. Methods: Prognostic markers were analyzed in 343 patients treated at Memorial Sloan-Kettering under 4 protocols (NCI-V90-0023, NCT00002634, NCT00002560, NCT00072358). All patients had stage 4 disease, diagnosed at >18 m at diagnosis and/or with MYCN amplification. They were treated with anti-GD2 antibody 3F8, without or with granulocyte-macrophage colonystimulating factor (GM-CSF), given either intravenously (iv) or subcutaneously (sc). Group 1 (n=169) was first treated in first remission, group 2 (n=105) had primary refractory disease, and group 3 (n=69) in > second complete remissions (2nd CR). Immunotherapy cycles were given every 1–3 months over 2 years if human anti-mouse antibody (HAMA) response was negative. Progression-free survival (PFS) and overall survival (OS) were estimated by Kaplan-Meier method; prognostic variables were tested using log-rank test, and their independent significance was assessed by multivariate Cox regression. Prognostic variables included tumor MYCN amplification, gene polymorphism of Fc-receptors (FcgR2A and FcgR3A), missing inhibitory KIR ligands for natural killers (NK), route of GM-CSF, pre-3F8 (pre-MRD) and the 2-month post-3F8 bone marrow MRD (post-MRD) using a 4-marker panel (GD2 synthase, PHOX2B, CCND1 and ISL1) as measured by quantitative RT-PCR, HAMA and other relevant treatment variables. Results: For PFS, post-MRD and disease group were statistically significant prognostic factors. For OS, HAMA, disease group, post-MRD, FCGR2A polymorphism, and missing KIR ligand were significant. In addition, interactions between FCGR2A and disease group, as well as between post-MRD and disease group were significant; FCGR2A being strongest among primary refractory, and post-MRD strongest among 1st remission group. Speakers’ Presentations Table 1 Endpoint Prognostic Variable HR Lower Upper p-value Bound Bound PFS 4.23 0.68 1.24 0.41 3.06 0.31 0.61 0.17 0.52 OS postMRD (positive vs negative) 1st remission vs primary refractory 2nd remission vs primary refractory HAMA (positive vs negative) postMRD (positive vs negative) FCGR2A (“HR or RR” vs “HH”) Missing KIR ligand (favorable vs unfavorable) 1st remission vs primary refractory 2nd remission vs primary refractory 3.14 0.49 0.85 0.28 1.65 0.16 0.42 0.07 0.22 5.70 0.95 1.81 0.59 5.69 0.60 0.87 0.39 1.21 <0.001 0.023 0.270 <0.001 <0.001 0.001 0.006 <0.001 0.129 Conclusion: Marrow MRD determined after two cycles of immunotherapy was the strongest predictor of outcome, irrespective of disease status at the beginning of anti-GD2 immunotherapy. While the significance of missing KIR ligand supports the role of NK-mediated cytotoxicity, the impact of high affinity FCGR2A allele primarily among patients with primary refractory chemo-resistant disease suggests a role of 3F8-mediated myeloid-dependent anti-NB effect for metastasis in the bone marrow. Topic 7: Pathway driven approaches SP022 Challenges and opportunities for molecular screening platforms: The EORTC SPECTA program D. Lacombe 1 , S. Tejpar 2 , F. Cardoso 3 , R. Salgado 4 , J. Hall 1 , V. Golfinopoulos 1 . 1 EORTC, Brussels, Belgium; 2 University Hospitals Leuven, Leuven, Belgium; 3 Champalimaud Cancer Center, Lisbon, Portugal; 4 Bordet Institute, Brussels Belgium Newly developed targeted drugs are tested, through clinical trials, on restricted groups of patients exhibiting specific biomarkers. When the prevalence of those biomarkers is low, a high number of patients need to be screened. Recruitment of patients in a biomarker-led clinical trial requires significant effort due to the number of patients screened and the complexity and cost of such screening. Moreover, when various clinical trials are targeting the same population of patients, the availability of samples may be restricted and patients may lose the opportunity to enter a potential clinical trial. To avoid duplication of effort and guarantee efficient patient access to clinical trials, the European Organization for Research and Treatment of Cancer is building a program of collaborative molecular screening platforms, for patients with various advanced tumor entities. The first platforms are dedicated to colo-rectal, central nervous system tumors and melanoma. SPECTA is the first European initiative intended to accelerate clinical trial access by ensuring robust screening of markers that can potentially allow patients to be enrolled in open clinical trials. Patients with predefined tumor stages will be proposed to have their material centrally processed for pathological examination and screening of cancer gene alterations. Common cancer genes are investigated using respectively certified assays and next generation sequencing technologies. By matching clinical, pathological and molecular profiles of patients to eligibility criteria of available biomarker-led clinical trials, SPECTA achieves a real-time identification of eligible patients across Europe and patient-oriented parallel screening for multiple trials. In this presentation, we will describe SPECTA as a new business model which is based on currently active network of clinical centers, a central biobanking and pathology facility, qualified assays laboratories, software to collect, trace and secure the data and algorithms to match the patients in real time with trials. We will also detail the model of partnership that has been developed to engage all stakeholders in the project, including the patient organizations, the European Society of Pathology, the Sanger Institute and our funders as well as the pharmaceutical industry sector. SP023 Pathways triggered by FGF and other growth factors R. Grose. Centre for Tumour Biology, Barts Cancer Institute, London, United Kingdom Fibroblast Growth Factor Receptor (FGFR) plays critical roles during embryogenesis, homeostasis and repair but is kept under tight control in normal tissues. A number of mechanisms have been used by cancers to co-opt FGFR signalling; ligands or receptors can be amplified/over-expressed, fusion Speakers’ Presentations proteins can be generated, autocrine stimulation can occur, and novel nuclear signalling pathways can be adopted. FGFR signalling has been implicated in a number of different cancer types. Genome wide association studies have identified key SNPs in FGFR2 that associate significantly with risk of developing ER positive breast cancer and, similarly, FGFR1 over-expression has been associated with poor prognosis in ER positive breast cancer. Activating mutations have been reported in several cancer types and, recently, we have identified a role for nuclear FGFRs in regulating cancer cell behaviour. I will discuss the current status of FGFs and FGFRs as markers in cancer and the mechanisms by which they are being targeted therapeutically. SP024 Prediction of the efficacy of mTOR targeted therapies T. Alain. McGill Cancer Centre, Montréal, Québec, Canada mRNA translation is dysregulated in cancer. This is caused by altered expression of translation factors and the hyperactivated oncogenic pathways that promote protein synthesis (e.g. the mammalian target of rapamycin (mTOR)). Consequently, the loss of translational control results in increased expression of proteins with proliferative, survival, and angiogenic functions. Active-site mTOR inhibitors (asTORi) hold great promise for targeting dysregulated mTOR signaling in cancer. However, because of the multifaceted nature of mTORC1 signaling, identification of reliable biomarkers for the sensitivity of tumors to asTORi is imperative for their clinical implementation. We found that cancer cells acquire resistance to asTORi by downregulating eukaryotic translation initiation factor (eIF4E)-binding proteins (4E-BPs). Loss of 4E-BPs or overexpression of eIF4E renders neoplastic growth and translation of tumor-promoting mRNAs refractory to mTOR inhibition. Conversely, moderate depletion of eIF4E augments the anti-neoplastic effects of asTORi. The anti-proliferative effect of these inhibitors in vitro and in vivo is therefore significantly influenced by perturbations in eIF4E/4E-BP stoichiometry, whereby an increase in the eIF4E/4E-BP ratio dramatically limits the sensitivity of cancer cells to asTORi. We propose that the eIF4E/4E-BP ratio, rather than their individual protein levels or solely their phosphorylation status, should be considered as a paramount predictive marker for forecasting the clinical therapeutic response to mTOR inhibitors. SP025 Genotype-based combinations of RAS/RAF and PI3K pathway inhibitors U. Banerji. Division of Cancer Therapeutics & Division of Clinical Studies, Institute of Cancer Research & The Royal Marsden, Sutton, United Kingdom Background: There are currently multiple inhibitors targeting the RAS/RAF and PI3K pathways (BRAF, MEK PI3K, AKT and m-TOR inhibitors) in development. Each class of drugs have shown clinical activity in single agent studies and are active in relatively small, defined subsets of tumours which harbour specific mutations. It is hypothesized that combinatorial inhibition of both signalling networks could broaden range of tumours where these drugs could be used in. Our aims included; a) To define tumours which were most likely to respond to the combination compared to each single agent alone b) To define the degree of target inhibition needed to achieve maximal growth inhibition (as it may not be possible to clinically deliver full doses of both drugs in a combination due to overlapping toxicity). Materials and Methods: We used two tool compounds (MEK inhibitor PD0325901 and AKT inhibitor AKT1/2 inhibitor) in an attempt to answer these questions. We exposed a panel of cell lines (4 BRAF mutant, 5 PIK3CA mutant 3 KRAS mutant and 5 cell lines with no mutation in BRAF, PIK3CA or KRAS) to the MEK and AKT inhibitor separately for 24 hours. We then calculated the concentration of the MEK inhibitor required to reduce the phosphorylation of ERK by 25, 50, 75 and 100% of and the concentration of the AKT inhibitor required to reduce phosphorylation of S6 by 25, 50, 75 and 100%. We then exposed the panel of cell lines to various concentrations of the combinations of the MEK and AKT inhibitors known to inhibit pre-defined degrees of signalling output for 96 hours and studied growth inhibition using sulforhodamine assays. Results: In 4/4 BRAF mutant cells, there was significantly more growth inhibition caused by maximal inhibition of MEK as compared to maximal inhibition of AKT and in 5/5 PIK3CA mutant cells, there was significantly more growth inhibition caused by maximal inhibition of AKT as compared to maximal inhibition of MEK. Interestingly, in 4/5 cell lines with no BRAF, S7 PIK3CA or KRAS mutations, cells were significantly more susceptible to AKT inhibition compare to MEK inhibition and 1/5 of these cell lines were equally susceptible to AKT and MEK inhibition. Further 1/3 cell lines with KRAS mutations were more sensitive to MEK inhibition, 1/3 more susceptible to AKT Inhibition and 1/3 were equally sensitive to maximal MEK or AKT Inhibition. Further experiments are being conducted to expand the cell line panel to 20 and study the inhibition of different combinations of MEK and AKT inhibitors known to inhibit MEK and AKT by different degrees. This will help to define the cells most likely to respond to the combination and further define the degree of target inhibition needed while designing these combinations. Conclusion: Combinations of MEK and AKT inhibitors are likely to be more effective than single agents only in defined subsets of cancers and this information is crucial to the design of clinical trials evaluating the efficacy of these combinations. Topic 8: Genomics driven approaches SP026 Genomic wide biomarker discovery in personalized patient derived xenografts M. Hidalgo. Centro Integtral Oncologico Clara Campal, Madrid, Spain Pancreatic cancer (PDAC) remains one of the most deadly cancers. Over the last few years, the genomic landscape of pancreatic cancer as well as precursor pancreatic cancer lesions have been deciphered in great depth. These studies show that PDAC develops as the consequence of accumulation of mutations in key oncogenes and tumour suppressor genes. The disease, once established, is characterized by high complexity, heterogeneity and genomic instability. Despite this facts, some patients harbour actionable mutations which targeting has resulted in significant clinical benefit. Indeed, one of the most active areas of research in PDAC is the development of strategies and approaches to personalize the treatment of patients. This is a complex field that can be tackle from many complementary angles. Our group has been interested in using patient derive xenogaft (PDX) models, aka Avatar mouse models, to guide cancer treatment. A piece of freshly collected tumour is implanted in immunodeficient mouse models, expanded, treated with different anticancer agents alone and in combination to select the most effective drug/regimen to treat the patient cancer. Our data show that the approach is highly predicted but, because of complexity and cost issues, not widely applicable to clinical practice at the present stage. To solve some of these limitations we are working on different aspects. One area is technological development to increase the take rate of tumours and to speed time to engraftment and expansion time. Currently, these figures are approximately 60–80% and 5–7 months. Studies are in progress to optimize this aspect. Another key question is the selection of agents, both alone and in combination, to be tested in the model. In this regard, it is important to integrate biomarker assessment in the tumour to pre-select a series of treatment candidates that can then be tested in the PDX models. To this end, we have now integrated next generation sequencing and assessment of copy number variation in patient’s tumour. These studies provide us with an unbiased overview of the tumour genomic landscape. From this data, using different bioinformatics and biological methods we extract the most relevant drug targets that are then bench tested against the patient Avatar mouse model to select the most effective treatment. SP027 RAS mutations as markers of resistance for colorectal cancer patients treated with the anti-EGFR monoclonal antibody panitumumab K. Oliner. Medical Sciences – In Vitro Diagnostics, Amgen Inc., Thousand Oaks, USA Panitumumab (pmab) is an anti-EGFR antibody that is now indicated in the EU for the treatment of adult patients with WT RAS (KRAS and NRAS) metastatic colorectal cancer. Data will be presented from analyses of tumor RAS mutations from three randomized pmab studies. The hypothesis generating study was a multigene analysis of the phase 3 monotherapy study of Pmab vs. best supportive care. Final RAS result ascertainment of this study employing a combination of next-generation sequencing, Sanger sequencing and WAVE-based SURVEYOR® analysis was 75%. This study suggested that RAS mutations beyond the most commonly tested KRAS codons 12 & 13 might be predictive of lack of response to Pmab therapy (Peeters, 2013 and Patterson, 2013). This led to two prespecified analyses of the RAS S8 biomarker, a prospective analysis of the phase 2 PEAK study of Pmab plus FOLFOX vs. bevicuzumab plus FOLFOX and the prospective-retrospective analysis of the phase 3 PRIME study of Pmab plus FOLFOX vs. FOLFOX alone. Banked patient tumor specimens, that were previously found to be WT KRAS exon 2 by the Therascreen® assay, were subjected to parallel Sanger sequencing and WAVE-based SURVEYOR® analysis of KRAS exons 3 & 4 and NRAS exons 2, 3 & 4. RAS result ascertainment was 80% and 90% in PEAK and PRIME, respectively. In the larger PRIME study 17% of tumors that were previously designated WT KRAS were found to harbor mutations in this expanded RAS analysis. In the pmab + FOLFOX studies, the additional RAS mutations were associated with an inferior PFS and OS in the pmab arm, consistent with previous findings for the KRAS exon 2 mutations. No new safety signals were found. Clinically, these randomized studies clearly demonstrate that expanded RAS analysis leads to the selection of patients who have increased PFS and OS compared to selection of patients who are wild-type for KRAS exon 2 alone (Oliner, 2013 and Schwartzberg, 2013) when treated with pmab. Most strikingly, the median OS benefit in the PRIME study was increased to 5.8 months from previously reported 4.2 month benefit when testing KRAS exon 2 alone. By selecting patients whose tumors do not harbor these additional RAS mutations there is a clinically meaningfully increase in the benefit:risk profile of panitumumab. This presentation will focus on how exploratory hypothesis generation testing followed by a prospective retrospective analysis led to the rigorous demonstration of RAS as a primary resistance biomarker. SP028 Epigenetic markers in cancer M. Hegi. Clinical Neurosciences, University Hospital Lausanne, Lausanne, Switzerland Epigenetic alterations have been recognized as important mechanisms in neoplastic transformation, malignant progression of cancer, and response to therapy. Efforts using high through put sequencing in diverse tumor types has identified a growing number of mutations in epigenetic modifier genes (e.g. MLL family member genes or TET2), histone genes (H3F3A), or in genes indirectly affecting epigenetic modifiers such as mutations in IDH1 and 2. These mutant proteins can have more generalized effects on DNA methylation through distinct mechanisms, resulting e.g. in a CpG island methylator phenotype (CIMP, IDH1 or 2 mutations, TET2), or a CpG island hypomethylator phenotype (CHOP, H3F3A-mt G34) deregulating expression of large numbers of genes, but also of regulatory non-coding RNAs. Epigenetic silencing by aberrant methylation of CpGs in the gene promoter region affects diverse cancer relevant pathways leading to activation of oncogenic pathways such as the WNT pathway, mediated by silencing of WNT antagonists, or by inactivation of tumor suppressing pathways, including DNA repair (MLH1, MGMT). Most interestingly, some of these epigenetic alterations can be converted into the “Achilles heel” of the affected tumors upon treatment with certain classes of anti-cancer agents. In fact, we have shown previously for glioblastoma that promoter methylation of the MGMT gene that encodes a DNA repair enzyme, confers a benefit to treatment with the alkylating agent temozolomide. Hence, the methylation status of the MGMT promoter has become the first predictive biomarker in this tumor type and is used in the clinic. Aberrant DNA methylation patterns have tumor type specific characteristics that are also distinct from respective normal tissue. Hence DNA methylation in characteristic genes can also serve as biomarkers for early detection of neoplastic cells, such as methylation of GSTP1 in prostate cancer or SEPT9 in colon cancer. For some tumor types non-invasive tests using plasma-derived DNA are in development. Taken together, development of markers for epigenetic alterations is a growing field and will inform on many questions, ranging from early tumor detection, tumor classification, prognostication, detection of treatment targets, or as predictive factors. Hence, beside mutation analysis of epigenetic modifier genes, reliable and validated tests for detection of aberrantly methylated gene promoters will be required. Speakers’ Presentations An EORTC Workshop on Biospecimen Pre-Analytical Stability and Diagnostics SP029 Keynote address: Overcoming the challenges of biospecimen collection for R&D, clinical trials and biobanking G. Thomas. Surgery and Cancer, Imperial College London, London, United Kingdom Risk analysis is a useful tool in designing protocols to address the challenges of biospecimen collection. An assessment of risk is based on the answers to three questions – What can go wrong?, How likely is it to happen? and, if it does happen, what are the consequences? In order to answer these questions, a detailed understanding of all the steps involved in biospecimen acquisition is needed. This includes how the multidisciplinary team interacts during a clinical procedure which leads to the acquisition of the biospecimen of interest. The second step then involves identifying the weak links in the chain of custody for the biospecimen, and the final step requires a detailed understanding of biospecimen science. For samples obtained from operative material, the chain from removal of tissue from a patient to stabilization of the material either by fixation and processing to paraffin or freezing can be complicated, dependent on the interaction of staff in the operating theatre and pathology staff and their relative locations within a hospital. Protocols that are to be used in multi-centre studies should identify areas of difference in procedure between sites that are most likely to lead to significant changes in specimen quality and provide information on how these effects may be mitigated. A biospecimen is only useful for research if it is fit for purpose. In reality, the majority of researchers will not have identified which of the pre-analytical variables that can occur during a chain of custody affect the suitability of that specimen for their specific analysis. It therefore becomes the responsibility of the protocol author to identify key steps in the chain from acquisition to use that are likely to result in degradation of the analyte within the sample. Procedures can then be put in place to decrease the likelihood of these events occurring. Acquisition of samples in the clinical setting is not a simple procedure, but an approach using risk analysis to design protocols for procurement is a useful tool in improving sample quality, and hence reducing noise due to pre-analytical variables in the final analysis. SP030 A controlled assessment of pre-analytical variables and their impact on the proteome in the collection and storage of plasma and serum David Chelsky No abstract presentation received. SP031 Evidence-based guidelines for the pre-analytical phase of DNA, RNA and cell-free DNA testing in blood samples M. Pazzagli 1 , F. Malentacchi 1 , R. Wyrich 2 , C.C. Hartmann 2 , P. Verderio 3 , S. Pizzamiglio 3 , C.M. Ciniselli 3 , M. Kubista 4 , S. Weisbuch 5 , S. Gelmini 1 . 1 Dept. of Experimental and Clinical Biomedical Sciences, University of Florence, Florence, Italy; 2 Qiagen GmbH, Qiagen GmbH, Hilden, Germany; 3 Unit of Medical Statistics and Biometry, IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 4 TATAA Biocenter AB, TATAA Biocenter AB, Gothenburg, Sweden; 5 Immunid, Immunid, Grenoble, France Molecular diagnostics have allowed for great progress in medicine, but its full potential is hampered by the lack of guidelines for the collection, handling, stabilisation and storage of the biosamples. The development of evidence-based quality guidelines for blood samples require identification of the critical steps in the pre-analytical procedure that need to be controlled and possibly further developed. To reach this goal, within the FP7 EU Granted Project “Standardisation and Improvement of generic pre-analytical tools and procedures for in vitro diagnostics” (SPIDIA; Grant Agreement no. 222916, www.spidia.eu), pan-European external quality assurance schemes (EQAS), specifically designated for the monitoring of the performance of the pre-analytical phase of DNA, cell-free DNA and RNA testing in blood samples, have been executed. With the support of the European Federation for Clinical Chemistry and Speakers’ Presentations Laboratory Medicine (www. efcclm.org) more than 320 applications from 220 laboratories representing 30 European countries were obtained. The participants received the same sample/s (whole blood, plasma) and performed, in two separated runs, sample extractions using their own standard protocol and reagents. Participants then sent back the extracted DNA/RNA to the SPIDIA laboratories for analysis, providing also details about the reagents and protocols used for the extraction. At the SPIDIA laboratories the extracted samples were tested for quality/quantity/integrity and stability. Each participant received a specific report with a summary of its performance obtained by considering the results of all the other participants. From the SPIDIA-EQAs, the most critical steps in the pre analytical procedure were identified. The results of these studies will enable the development of several European Technical Specifications (CEN/TS) currently under evaluation by the European Committee for Standardization (CEN; www.CEN.eu). SP032 Effects of intrasurgical and postsurgical variables on stability of signaling molecules in cancer tissues J. Hartmut. Indivumed, Hamburg, Germany Surgical biospecimens are the main source for cancer research and tissue diagnosis. With rapidly increasing numbers of targeted therapies the understanding of tumor biology in patients becomes important for patient stratification and prediction of drug response. Therefore, it is extremely important to understand variables that affect expression of genes and proteins and their phosphorylation status in tissues during surgery and postsurgical processing. Indivumed together with three partner clinics in Hamburg, Germany (Israelitisches Krankenhaus, Agaplesion and University Clinic Hamburg-Eppendorf) performed a study, funded by the NCI, to analyze the impact of surgical procedures and postsurgical processing of tissue with respect to gene expression, the expression of cancer therapy relevant proteins and the phosphorylation of key proteins regulating cancer pathway activity in primary and metastatic colorectal cancer tissue. The ultimate goal has been to define genes and proteins which are stable and, thus, can be regarded to be uncritical for research and diagnostic purposes as well to identify potential markers that can serve as quality markers for tissue testing. Tissue samples from the tumor and adjacent normal tissue were collected from 80 patients with colorectal cancer at different time points before and after tumor resection. They were subjected to an analysis of defined proteins of the EGFR-family and the related Akt and MAPK signaling pathway. In addition, gene expression was analyzed in total RNA samples using whole genome chips. Our data show that there is a significant difference in the molecular composition of tissue specimens that were collected after tumor resection compared to those tissue specimens that were collected before tumor resection. This difference is larger than the difference between various post-resection times between 10 and 45 minutes. The observed effect is either due to warm- and cold ischemia, or is caused by a combination of the surgical manipulation of the tissue, length of warm ischemia, drug application and other individual variables. Postsurgery the processing time is the most important variable for preserving molecular composition. This study presents an important contribution to the understanding of molecular changes that are being introduced into tissue samples during the preanalytical phase, i.e. by the tissue collection procedure itself. SP033 Evaluation of novel alternatives to formalin fixation for companion diagnostics K. Zatloukal 1 , K.F. Becker 2 , D. Groelz 3 , S. Guendisch 2 , M. Kap 4 , M. Kruhoffer 5 , P.H.J. Riegman 4 , P. Turano 6 , C. Viertler 1 , R. Wyrich 3 . 1 Medical University of Graz, Institute of Pathology, Graz, Austria; 2 Technical University, Institute of Pathology, Munich, Germany; 3 Qiagen GmbH, Research and Development, Hilden, Germany; 4 Department of Pathology, Josephine Nefkens Institute, Rotterdam, The Netherlands; 5 Aros Applied Biotechnology A/S, Science Park Skejby, Aarhus, Denmark; 6 University of Florence, Magnetic Resonance Center and Department of Chemistry, Florence, Italy The efficacy of targeted therapies mostly relies on the selection of the appropriate patients to treat. This requires the analysis of the status of S9 the drug target and affected downstream biological mechanisms. Therefore, the application of companion diagnostics based on different types of molecular biomarkers related to DNA, RNA, proteins or metabolites on tissues becomes increasingly important. It is know that these biomolecules are markedly modified by the formalin fixation procedure which limits their analysis in tissue specimens collected for histopathological examination. A novel non-formalin based fixative (PAXgeneR, Qiagen) has been extensively tested within the large integrated project “Standardisation and improvement of generic pre-analytical tools and procedures for in-vitro diagnostics” (SPIDIA; www.SPIDIA.eu) which was funded by the 7th European framework programme. Data obtained showed that PAXgene preserved DNA and RNA almost as good as snap freezing in liquid nitrogen. The efficiency and accuracy of next generation DNA sequencing was comparable to snap frozen tissues. PAXgene revealed far less gene-to-gene variations as compared to formalin fixation and resulted in up to 100-fold increased sensitivity of qRT-PCR based assays. Furthermore, proteins including phospho-proteins could be recovered well from PAXgene-fixed paraffin-embedded tissue samples. Epitopes were well preserved. However immunohistochemistry protocols required optimization for some of the antibodies tested. Morphologic features and diagnostic criteria could be well reproduced as compared the classical formalin-fixed tissues. Safety evaluation showed similar pathogen inactivation capacity as formalin. These findings suggest that PAXgene is a promising new fixative that would allow state-of-the-art histopathological analysis and application of a broad range of molecular biomarkers from the same tissue sample without the need of different sample processing workflows (e.g., formalin fixation and snap freezing). This would markedly improve the feasibility of application of novel companion diagnostics in drug development and routine health care. The work was funded by the 7th European Union Framework Programme under grant agreement no. 222916. SP034 Preanalytical variation affecting detection of RNA in breast cancer tissues F. Symmans No abstract presentation received. SP035 NCI biospecimen evidence-based practices H. Moore, K.B. Engel. Biorepositories and Biospecimen Research Branch – Cancer Diagnosis Program Division of Cancer Treatment and Diagnosis, National Cancer Institute, USA The U.S. National Cancer Institute’s BRN (Biospecimen Research Network) program focuses on the systematic study of different biospecimen collection, processing, and storage parameters and their effects on downstream analyses. The BRN supports new research and maintains an online literature resource, the Biospecimen Research Database, which catalogs and describes the published literature in Biospecimen Science. Together, these resources help to build a much-needed knowledge base in Biospecimen Science. A remaining challenge is the translation of this knowledge to improved biospecimen practices, with the ultimate goal of improving the quality of scientific research and development. To address this challenge the NCI has developed the concept of Biospecimen Evidence-based Practices. This concept will be elaborated in the presentation. The Biospecimen Evidence-based Practices can be used to support and speed the development of evidence-based Standard Operating Procedures for biospecimen collections that are integral to basic research and clinical trials. S10 SP036 SPIDIA – Dissemination of results into CEN technical specifications for biospecimen handling U. Oelmueller 1 , K.F. Becker 2 , M. Heinrich 3 , L. Krieger 3 , M. Kubista 4 , M. Pazzagli 5 , P. Riegman 6 , P. Turano 7 , C. Viertler 8 , K. Zatloukal 8 . 1 MDx Development, QIAGEN GmbH, Hilden, Germany; 2 Department of Pathology, Technical University of Munich, München, Germany; 3 DIN Institute, DIN Deutsches Institut für Normung e. V., Berlin, Germany; 4 TATAA Biocenter AB, TATAA Biocenter AB, Goeteborg, Sweden; 5 Dept. of Experimental and Clinical Biomedical Sciences, University of Florence, Florence, Italy; 6 Department of Pathology, Erasmus Medcial Center, Rotterdam, Netherlands Antilles; 7 Magnetic Resonance Center (CERM), University of Florence, Florence, Italy; 8 Institute of Pathology, Medical University of Graz, Graz, Austria Molecular in vitro diagnostics and biomedical research have allowed great progress in medicine. Further progress is expected by new technologies analyzing cellular biomolecule profiles such as nucleic acids, proteins, and metabolites. New biomarkers based on these biomolecule classes including disease specific biosignatures will be key value drivers for future personalized medicine and improved health care. Studies have demonstrated that profiles of these molecules as well as their qualities can change significantly during Speakers’ Presentations sample collection, processing, transport, storage, archiving and biomolecule isolation thus making diagnostics or research unreliable or even impossible. High quality clinical samples with preserved bioanalyte profiles are therefore critical to diagnostics, research, and biobanking. Developing new pre-analytical technologies and tools as well as standardizing pre-analytical workflows is therefore of high importance. The large-scale integrating research project SPIDIA within the European Union FP7 program is one of the major contributors to these efforts (grant no. 222916, www.spidia.eu). The project is supported by 7 private research and diagnostic companies, 8 public research organizations, and the European Committee for Standardization CEN. The consortium is working on the standardisation and improvement of generic pre-analytical tools and procedures for in vitro diagnostics. SPIDIA’s research has focused on analyzing and improving pre-analytical workflows for blood, tissues, and body fluids. The project results serve as important evidence for developing new standards. In October 2012 the European Committee CEN/TC 140 “In vitro diagnostic medical devices” accepted the SPIDIA consortium’s proposal for a first Technical Specification Document “Molecular in-vitro diagnostic procedures – specifications for pre-analytical workflows for selected sample types”. Up to 16 Technical Specifications for different individual sample types and bioanalytes such as RNA, DNA, proteins, metabolites are currently being developed within a dedicated CEN/TC 140 working group. S11 Six Best Poster Abstracts – Oral Presentations MC13-0026 KRAS mutated plasma DNA as predictor of outcome from irinotecan monotherapy in metastatic colorectal cancer K. Spindler 1 , A.L. Appelt 1 , N. Pallisgaard 2 , R.F. Andersen 2 , A. Jakobsen 1 . 1 Oncology, Vejle Hospital, Vejle, Denmark; 2 Biochemistry, Vejle Hospital, Vejle, Denmark Background: A major proportion of colorectal tumors harbour KRAS mutations, a negative predictive marker for outcome of EGFR targeted therapies combined with irinotecan. The role of mutation status for outcome of chemotherapy alone has only been sparsely investigated and with negative results. Purpose/Objective: We investigated the clinical implications of KRAS mutations in patients treated with irinotecan monotherapy when detected in archival tumor tissue and plasma cell free DNA. Materials and Methods: Patients were included in a prospective nonrandomised phase II and biomarker study (Protocol ID S-20090114). Inclusion criteria were; histopathologically verified mCRC, measurable disease according to RECIST, indication for secondline irinotecan monotherapy according to local guidelines, informed consent to therapy and biobank collection and age ≥18. Patients received irinotecan 350 mg/m2 q3w. Response was evaluated according to RECIST v 1.1. Plasma was obtained from a pre-treatment EDTA blood-sample, and KRAS mutations detected in tumor and plasma using an in-house qPCR. Results: The median number of cycles was 4 (range 1–15), response rate (RR) 13%, and disease control rate (DCR) 57%. Median PFS was 4.6 mo (95% CI 3.7–5.8) and OS 9.5 mo (95% CI 8.4–11.8). Ninety-two patients had matching tumor and blood samples and concordance between tumor and plasma KRAS status (pKRAS) was 82% (69/84). Mutation status in archival tumor tissue did not correlate to efficacy, but none of the patients with mutations detectable in plasma responded to therapy (RR=0). The RR in pKRAS wt patients was 19%, (p=0.014). DCR in pKRAS wt patients was 66%, and 37% in the patients with pKRAS mutations (p=0.01). Tumor mutation status was associated with OS but not PFS, whereas pKRAS had a strong influence on both parameters. Median OS was 13.0 mo (95% CI 9.5–15.1) in pKRAS wt patients, and 7.8 mo (4.6–8.4), in patients with pKRAS mutations, HR 2.26 (95% CI 1.31–3.90), p<0.0001. PFS was 4.6 mo (95% CI 3.3–6.4) and 2.7 mo (95% CI 2.1–4.5), respectively, HR 1.69 (95% CI 1.03–2.77), p=0.01. Cox regression analysis confirmed an independent prognostic value of pKRAS status, but not KRAS tumor status. Conclusions: This study indicates that tumor KRAS has minor clinical impact in patients treated with irinotecan monotherapy compared to plasma KRAS which seems to hold important predictive and prognostic information. These data call for reconsideration of the role of KRAS in mCRC and for validation in a randomised trial. MC13-0049 Two-stage adaptive cutoff design for building and validating a prognostic biomarker signature M. Polley, E. Polley, E. Huang, B. Freidlin, R. Simon. Biometric Research Branch, National Cancer Institute, Bethesda MD, USA Background: The scientific community has committed expansive resources during the last decade to identify useful biomarkers for clinical use. Purpose/Objective: Cancer biomarkers are frequently evaluated using archived specimens. Routine collection of high quality specimens is an expensive and time-consuming process. Therefore, care should be taken to preserve these precious and scarce resources. Here we propose a novel statistical design to allow efficient evaluation of a panel of biomarkers while making economic use of available resources. Materials and Methods: Motivated by the use of futility monitoring for a treatment effect in clinical trials, here we propose a Two-Stage Adaptive Cutoff (TACO) design that affords the possibility to stop performing biomarker assays if an early evaluation of the model performance indicates that the hypothesized biomarker effect will not be confirmed. Our study design includes four components: (a) signature building with a statistically valid test of the model performance based on data in stage 1; (b) an early futility stopping if the model performance in stage 1 is unsatisfactory; (c) a completely locked-down biomarker signature in stage 1 with a properly cross-validated cutoff to classify patient status; and (d) an independent validation of the locked-down biomarker signature (including the cutoff) in stage 2. Results: We give an example based on a publicly available dataset to illustrate the use of the procedure. Simulation studies are presented to evaluate the operating characteristics of the design. We demonstrate that with the proposed design, substantial savings in specimens is possible under the null hypothesis when the model performance is undesirable. The practical aspects of the proposed design are discussed. Conclusions: In this work, we propose a statistical design useful for developing and validating a biomarkers signature which includes a stopping rule for futility in the event of poor model performance. Novel statistical designs like this are needed to allow conservation of precious tissue specimens for future research. MC13-0060 Analytical validation of the MPACT assay, a targeted next generation sequencing clinical assay for cancer patient treatment selection C. Lih 1 , D. Sims 1 , E. Polley 2 , Y. Zhao 2 , M. Mehaffey 1 , T. Forbes 1 , R. Harrington 1 , W. Walsh 1 , P. McGregor 1 , R. Simon 2 , B. Conley 3 , S. Kummar 4 , J. Doroshow 4 , P.M. Williams 1 . 1 Molecular Characterization Laboratory, Frederick National Laboratory for Cancer Research, Frederick, USA; 2 Biometric Research Branch, National Cancer Institute, Shady Grove, USA; 3 Cancer Diagnosis Program, National Cancer Institute, Shady Grove, USA; 4 Division of Cancer Treatment and Diagnosis, National Cancer Institute, Bethesda, USA Background: Robust and analytically validated assays are essential for development of targeted cancer therapies. MPACT (Molecular Profiling based Assignment of Cancer Therapeutics) is a proposed clinical trial that randomizes patients who failed standard treatment to therapy targeted or not to their own tumor’s molecular abnormalities. Purpose/Objective: Using next generation sequencing, we developed a multi-analyte somatic variant assay to support the treatment selection for MPACT trial. Here we report the results from analytical validation study. Materials and Methods: The MPACT assay interrogates a total of 59,150 bp that represents minimally, 392 variant loci with reported actionable value to guide treatment decisions. In this analytical validation study, we assessed the sensitivity, specificity, and reproducibility for 5 classes of sequence variants, single nucleotide variant (SNV), insertion and deletion (Indel, 3bp or less), large indel (greater than 3bp), SNV at homopolymeric region (HP, greater than 2 identical bases in a row), and Indel at HP. Results: Using DNA samples derived from a hapmap normal cell-line (CEPH, NA12878) with 27 spiked positive control plasmids and from cultured cancer cell-lines or FFPE xenografts derived from cells harboring known variants, we found that the sensitivity is 99.3% in SNVs, 91.1% in SNVs at HP, 100% in Indels, 66.7% in Indels at HP, and 88.9% at large Indels. Reproducibility analysis in a subset of DNA samples revealed that the interand intra-operator concordances are 95.1% and 98.3% respectively, with greater than 0.99 R square values in allele frequency for detected variants. Sanger sequencing of 22 loci in 5 xenograft samples demonstrated 100% accuracy with MPACT assay results. By sequencing three hapmap normal cell-lines (CEPH, Yoruban, Chinese female) multiple times, we showed that the MPACT assay achieved 100% specificity in reportable range for all 5 classes of variants. In addition, we blindly tested 10 unknown tumor specimens that were genotyped previously by a CLIA assay in another lab and found that the MPACT assay reported accurately all 9 previously identified variants at near identical frequency. Finally, we compared the data from MPACT assay and exome capture sequencing using Hiseq for 5 pairs of clinical tumor specimen and matched blood samples. The high concordance confirmed that the MPACT assay results are validated by an independent NGS platform. Conclusions: This validation study demonstrated that the MPACT assay was well-suited for the intended clinical use. S12 MC13-0071 Immune response against non-targeted tumor antigens after treatment with sipuleucel-T and its association with improved clinical outcome D. Guhathakurta, L.Q. Fan, T. Vu, N.A. Sheikh, J.B. Trager. Research, Dendreon Corporation, Seattle, USA Background: The development of immunotherapies targeting cancer provides a unique challenge in biomarker development [1]. Clinical benefits of immunotherapy are often not observed with the typical measures used to assess tumor progression (such as the RECIST or WHO criteria). Therefore, the identification of novel post-treatment biomarkers associated with clinical benefit is important for cancer immunotherapy development as well as patient care management. One consequence of an effective immunotherapy may be epitope spread or antigen spread [2]: tumor cell death during the initial response to an immunotherapy may lead to the release of tumor-associated antigens and the priming of self-reactive T and/or B lymphocytes specific to these antigens [2]. Antigen spread may subsequently promote more efficient tumor killing and may occur with a higher frequency in clinical responders [2], therefore providing avenues for the identification of novel, mechanism-based, biomarkers of clinical outcome. Purpose/Objective: Sipuleucel-T, an FDA approved immunotherapy for the treatment of symptomatic or minimally symptomatic, metastatic castrateresistant prostate cancer [3], is designed to elicit immune responses to the prostate-specific protein, Prostatic Acid Phosphatase (PAP). We have sought to determine if antigen spread occurs in response to sipuleucel-T. Materials and Methods: Using high-content protein microarrays, followed by independent technical validations, we have evaluated antigen spread in patients after sipuleucel-T treatment in IMPACT, a controlled phase 3 clinical study. Results: Subjects treated with sipuleucel-T consistently mounted an elevated IgG responses against a range of cancer antigens, whereas subjects in the control arm did not. Responses were observed against targets aberrantly expressed in prostate tumors as well as targets in pathways involved in prostate cancer progression. IgG responses to specific cancer or prostate antigens were technically validated using Luminex xMAP, and were also observed in an independent clinical trial. Importantly, these responses were associated with improved clinical outcome (overall survival). Conclusions: This study provides further insight into the mechanism of action of sipuleucel-T and potential biomarkers to assess clinical outcome after treatment. The methods and results presented here may benefit the development of biomarkers of clinical outcome for other cancer immunotherapies. References: [1] Topalian, SL, et al. Cancer immunotherapy comes of age. J Clin Oncol 2011;29:4828–36. [2] Ribas A, et al. Determinant spreading and tumor responses after peptide-based cancer immunotherapy. Trends Immunol 2003;24:58–61. [3] Kantoff PW, et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010;363:411–22. MC13-0075 Rapid assessment of TORC1 suppression as a functional biomarker predicting responsiveness to RAF and MEK inhibitors in BRAF-mutant melanoma patients R.B. Corcoran 1 , S.M. Rothenberg 1 , A. Piris 1 , R.M. Nazarian 1 , S. Maheswaran 1 , J. Settleman 2 , J.A. Wargo 1 , K.T. Flaherty 1 , D.A. Haber 1 , J.A. Engelman 1 . 1 Cancer Center, Massachusetts General Hospital, Boston, USA; 2 Discovery Oncology, Genentech, South San Francisco, USA Background: Selective RAF and MEK inhibitors have transformed the treatment of BRAF -mutant melanoma. Still, a substantial percentage of patients fail to respond to therapy, and most responses are partial and short-lived. Presently, there are no clinically useful biomarkers to guide the treatment of BRAF -mutant melanoma. Purpose/Objective: To identify effective biomarkers to predict which BRAFmutant melanoma patients are more or less likely to benefit from RAF or MEK inhibitors. Materials and Methods: BRAF-mutant melanoma cell lines were characterized before and after treatment with RAF or MEK inhibitors in vitro and in xenografts. Paired biopsies obtained from BRAF-mutant melanoma patients before and ∼2 wks after initiation of RAF inhibitor therapy were assessed by IHC for a change in phosphorylation of ribosomal protein S6 (P-S6). Serial fine-needle aspiration (FNA) biopsies were obtained prospectively from Best Poster Abstracts – Oral Presentations patients before and after initiation of therapy, and P-S6 was assessed by quantitative immunofluorescence microscopy. Results: Suppression of TORC1 activity in response to RAF or MEK inhibitors, as measured by decreased P-S6 levels, effectively predicted sensitivity in BRAF -mutant melanoma cell lines in vitro and in mouse tumor xenografts. In sensitive melanomas, TORC1 and P-S6 were suppressed in response to RAF or MEK inhibitors, but in resistant melanomas, TORC1 activity was maintained, in some cases despite robust suppression of MAPK signaling by these inhibitors. In paired biopsies obtained from patients with BRAF -mutant melanoma before treatment and after initiation of RAF inhibitor therapy, P-S6 suppression was associated with significantly improved PFS [HR=0.19, 95% CI 0.01–0.84, p=0.03]. In serial FNA biopsies obtained prospectively from patients before and during the first 2 weeks of RAF inhibitor therapy, a change in P-S6 in patients’ tumor cells could be readily monitored in real-time by multiplexed, quantitative immunofluorescence microscopy. A decrease in P-S6 correlated with tumor response, providing a rapid and minimally-invasive potential means to monitor the efficacy of treatment, before changes in tumor volume are apparent by traditional radiographic imaging. Conclusions: These results establish suppression of P-S6 after initiation of RAF inhibitor therapy as a robust potential functional biomarker to guide the treatment of BRAF -mutant melanoma, and present a powerful methodology for monitoring changes in potentially any signaling pathway in response to targeted therapies in patients. MC13-0079 Prospective mutational characterization of Japanese patients with non-small cell lung cancer using surgically resected tumor specimens by next-generation sequencing Y. Koh 1 , H. Kenmotsu 2 , M. Serizawa 1 , M. Isaka 3 , K. Mori 4 , T. Takahashi 2 , M. Endo 5 , T. Nakajima 6 , Y. Ohde 3 , N. Yamamoto 2 . 1 Drug Discovery and Development, 2 Thoracic Oncology, 3 Thoracic Surgery, 4 Clinical Trial Coordination Office, 5 Diagnostic Radiology, 6 Diagnostic Pathology, Shizuoka Cancer Center, Sunto-gun, Japan Background: Detection of tumor genetic alterations is critically needed for lung cancer clinic as well as for the development of molecular targeted therapeutics. Purpose/Objective: Here we report the results of a broad spectrum of genetic alterations identified in Japanese non-small cell lung cancer (NSCLC) patients by ultra-deep targeted sequencing. Materials and Methods: Highly multiplexed amplicon sequencing was performed using genomic DNA extracted from snap-frozen tumor specimens. TruSeq amplicon cancer panel was used for the detection of somatic mutations in 48 cancer related genes followed by ultra-deep sequencing (Illumina) at an average coverage of approximately 3,400×. ALK, ROS1 and RET translocations and EGFR, MET, PIK3CA, FGFR1 and FGFR2 amplifications were also detected by multiplex RT-PCR and quantitative PCR, respectively. Results: The demographics of 279 consecutive patients enrolled in this prospective study at Shizuoka Cancer Center between July 2011 and March 2013: median age 69 years (range: 38–92); male 66%; never smoker 25.8%; histology: adenocarcinoma 70.6%, squamous cell carcinoma (SQ) 25.1%, others 4.3%; tumor stage: I 58.1%, II 22.6%, III 15.4%, IV 3.9%. TP53 mutation was most frequently detected (35.9%) in all patients, particularly in SQ (59.2%). Mutations in genes such as CTNNB1 (4.0%), SMAD4 (1.5%), GNAS (1.0%), STK11 (1.0%), HRAS (0.5%) and PTPN11 (0.5%) were also detected in addition to major mutations in genes such as EGFR (44.8%), KRAS (18.4%) and PIK3CA (4.5%) in adenocarcinoma. PIK3CA (19.7%), HRAS (2.8%), APC (1.4%), FGFR2 (1.4%), FGFR3 (1.4%) and SMAD4 (1.4%) mutations were identified in SQ and notably, 40.9% of SQ patients harbored concurrent gene mutations, suggesting the genetic complexity of this histological subset. FGFR1 amplification was found in 7.0% of SQ, suggesting lower frequency in East Asian population than in Caucasian population. As for PIK3CA mutation, the majority (92.8% of all PIK3CA mutant SQ cases) was detected in exon 9 (residues E542 and E545) in SQ but in contrast, mutations in exon 1 and 20 (residues G106 and H1047, respectively) were more frequently detected (33.3% combined of all PIK3CA mutant adenocarcinoma cases) in adenocarcinoma. Conclusions: We managed to detect a wide range of genetic alterations and to identify additional actionable mutations along with popular driver mutations in NSCLC by next-generation sequencing technology. These data should be incorporated into lung cancer clinic to implement personalized cancer medicine. S13 Poster Presentations MC13-0001 Validation of a new tissue acqusition tool for molecular biology in oncology J. Janssens 1 , M. Verjans 2 , A. Cornelis 3 , M. Deleu 4 , J.P. Bogers 5 . 1 Oncology, University Hasselt, Hasselt, Belgium; 2 Gynecology, Regional Hospital Tienen, Tienen, Belgium; 3 Pathology, Regional Hospital Tienen, Tienen, Belgium; 4 Oncology, Regional Hospital Tienen, Tienen, Belgium; 5 Pathology, University Antwerp, Antwerp, Belgium Background: Histological and molecular examinations are a prerequisite to understand cancer, determine optimal individualized care, and identify targets for potential novel therapies. Despite this key role of tissue based research, the act of biopsy still remains troublesome. Purpose/Objective: New direct and frontal biopsy technologies have emerged with the aim to alleviate the bottleneck of inappropriate tissue acquisitions from minimal invasive interventions. This study was undertaken to validate the Spirotome type of instruments to harvest high quality tumor tissue in sufficient quantity. Materials and Methods: The new Spirotome instruments for minimal invasive biopsy procedures were evaluated in a population of 750 patients at risk or suffering from various types of cancer in the period 2004 to 2012. All patients were given informed consent according to institutional requirements. The biopsy specimen, harvested under direct imaging in various parts of the human body, was evaluated to provide enough high quality tissue for omic research in comparison to diagnostic surgery. Only when the pathological and molecular data were considered complete and comparable to surgery or in line with follow-up and with sufficient patient comfort, the procedure was defined as successful. Results: In all 750 patients tissues could be harvested. In less than 2% the tissue was considered non contributive to a complete diagnosis. Sample sizes were between 100 and 300 mg. Patient comfort was excellent as less than 5% of the patients experienced significant pain or hematoma that needed medication or intervention. Pneumothorax for lung applications was less than 10%. In only two cases hospitalization was needed or prolonged (pneumothorax). 53 cases had diagnosis when surgery was not possible. The cost of the materials was less than 200 Euro per procedure. Conclusions: New biopsy procedures are increasingly more patient friendly with appropriate comfort and safety. The new macrobiopsies are less expensive, making molecular biology at reach for every oncological patient, company and health care provider. Direct and frontal macrobiopsies open new avenues for future bio-banking, pharmacogenomics, omic research and personalized medicine. It is anticipated that drug discovery and clinical implementation of targeted therapies will be highly facilitated and that clinical research time for multicenter trials will be significantly shortened MC13-0004 Prognostic significance of MGMT and IDH1 in patients with secondary glioblastoma L. Bie, X. Zhang, M. Li. Neurosurgery, The First Hospital of Jilin University, Changchun, China Background: Several studies have observed a change in MGMT silencing and IDH1 mutation in secondary glioblastomas (sGBMs). However, reports about the prognostic value of promoter methylation of the MGMT gene and IDH1 mutations in secondary glioblastomas (sGBMs) are few in number. Purpose/Objective: In our research, pts with sGBM and IDH1 mutation had a significantly improved outcome. Moreover, MGMT methylation is also a powerful prognostic marker in sGBM pts. Materials and Methods: The study involved primary and secondary tumor tissue samples from 89 GBMs pts (P/S: 42/47) diagnosed and treated within the First Hospital of Jilin University from Jan 2006 to Nov 2011. After surgical treatment, all GBM pts were subjected to radiotherapy with concomitant administration of TMZ. Pts with GBM were screened for promoter status of MGMT gene, by the methylation-specific polymerase chain reaction (MSP), and, for IDH1 mutations by direct sequencing. Results: A total of 42 of 89 pts (47.2%) had primary GBMs (pGBMs) (Group 1), while 47 pts (52.8%) had secondary GBMs (Group 2). In Group 2, 23 pts (21/47, 44.7%) developed a sGBM through progression from a low grade glioma (LGG) WHO grade II and a secondary anaplastic gliomas (sAG) (Group 2a), whereas 26 pts (26/47, 55.3%) showed a direct malignant transformation from a LGG to a sGBM (Group 2b). MGMT methylation was observed in 38 (42.7%) GBMs, with a higher frequency in sGBMs than pGBMs (24/14, 51.1 vs. 33.3%; p=0.008). MGMT methylation status was associated with increased median survival time in pGBMs and sGBMs pts. IDH1 mutations are present in the majority of sGBMs but rare in pGBMs (31/5, 66.0 vs. 11.9%; p<0.001). Group 2a had a higher frequency with IDH1 mutation than group 2b (18/12, 78.3 vs. 46.2%; p<0.01). The median survival time after malignant progression of all sGBMs pts with an IDH1 mutation was longer than in pts with wild-type IDH1 (3.2 vs. 1.1 ys; p<0.01). IDH1 mutation had an improved outcome in group2a, which is compared with group 2b (3.9 ys vs. 2.1 ys; p<0.05). Conclusions: In our population, pts with sGBM and IDH1 mutation had a significantly improved outcome. In addition, MGMT methylation is also a powerful prognostic marker in sGBM pts. MC13-0005 An integrated molecular genetic profiling of low-grade gliomas identifies clinically relevant mRNA genes L. Bie, Y. Li. Neurosurgery, The First Hospital of Jilin University, Changchun, China Background: Most low-grade gliomas patients the prognosis is good, but the small number of patients can progress to high-grade gliomas. Less biomarker for prognosis of the patient clinical urgent need for new effective prognostic markers. Purpose/Objective: WHO Grade II glioma (G2G) is a pre-malignant brain tumor and continuous growth, migration along the white matter tracts and unavoidable malignant transformation. Thus we have identified an mRNA expression signature that can predict G2G patient survival. Materials and Methods: To identify an mRNA expression signature that can predict G2G patient survival, we analyzed the mRNA expression data of GBM patients (n=91) derived from The Cancer Genome Atlas (TCGA) and GEO microarray datasets by WebArrayDB software. We divided the patients randomly into training (n=46) sets and testing sets (n=45) with equal number in each group. Results: We identified 8 significant mRNAs using Cox regression analysis on the training set and formulated a risk score based on the expression signature of these mRNAs that segregated the patients into high and low risk groups with significantly different survival times (hazard ratio [HR] = 3.2; 95% CI: 2.1–6.7; p<0.001). This signature was independently validated in the testing set (HR=1.2; 95% CI: 0.5–2.7; p<0.01). G2G patients with high risk scores had overall poor survival compared to the patients with low risk scores. Cox multivariate analysis with patient age as a covariate on the entire patient set identified risk score based on the 8 mRNA expression signature to be an independent predictor of patient survival (HR=1.3; 95% CI: 0.47–3.12; p=0.02). Conclusions: Thus we have identified an mRNA expression signature that can predict G2G patient survival.These findings may have implications in the understanding of gliomagenesis, development of targeted therapy and selection of high risk G2G patients for adjuvant therapy. MC13-0006 Cell-free circulating DNA can be used as a noninvasive approach for detection of genetic/epigenetic alterations in brain tumors L. Bie, Y. Li, X. Hong. Neurosurgery, The First Hospital of Jilin University, Changchun, China Background: Intracranial meningioma tumors, the majority of patients with good prognosis, but some patients relapse, progression to malignancy.The blood markers tumors to assess patient prognosis in meningiomas has not been carried out. S14 Poster Presentations Purpose/Objective: We evaluated whether cell-free circulating DNA can be used as a noninvasive approach for detection of genetic/epigenetic alterations in brain tumors during the course of the disease. Materials and Methods: Paired tumor-serum samples from 37 primary patients with glioblastoma and 21 primary benign meningiomas were analyzed. 19 non-cancer individuals serum were used as control. The median interval between surgery and serum sampling was 1.5 month. The methylation status of O6 -methyl guanine methyltransferase (MGMT) was studied by methylationspecific PCR. The mRNA expression of CCNB1 was studied by qRT-PCR. Promoter hypermethylation in MGMT was detected at high frequencies in paraffin-embedded (FFPE) tumor sections and serum in glioblastoma. Results: Statistically significant tumor-serum concordance was found for MGMT methylation in patients (r =0.87, p<0.01). None of the control serum showed aberrant methylation. Hypermethylation in serum DNA was all accompanied with methylation in the corresponding tumor tissues with 100% specificity. Highly elevated MGMT methylation levels in serum were the sole independent factors predicting inferior overall survival in this cohort. Over-expression of CCNB1 was also detected in FFPE and serum in meningioma patients. None of the control serum showed over-express. CCNB1 expression was significantly higher in patients with tumor recurrence (p<0.01). Conclusions: CCNB1 serum mRNA was a potential prognostic factor predicting meningioma recurrence. glioblastoma could provide important information for the progression and host responses of glioblastoma. Purpose/Objective: We studied a panel of 120 cytokines and growth factors and investigated their prognostic values for glioblastoma. Materials and Methods: A protein antibody array was first performed to study the prognostic significance of 120 cytokines in the plasma samples of 45 glioblastoma patients prior to craniotomy or biopsy procedure. An independent set of plasma samples from 100 patients with glioblastoma with complete clinicopathologic data and follow-ups were used for validation. Results: Ten cytokines were identified by Significance Analysis of Microarray (SAM), in which four were associated with poor prognosis (IL-15, MCP-1, GDNF, IL-1R4/ST2), and six were associated with good prognosis (IGFBP-6, MIP-1δ, ICAM-3, IL-7, MIP-3β, and sgp130) of the glioblastoma patients. Moreover, a 4-cytokine panel composed of IL-7, IL1R4/ST2, sgp130 and MCP-1 showed significant correlation with overall survival of the glioblastoma patients (HR: 2.068; 95% CI: 1.357–3.153; p=0.001). In the validation set, the cytokine panel was significantly correlated with overall survival (HR 1.753; 95% CI 1.502–2.255, p<0.001). Conclusions: This panel of four cytokines: IL-7, IL1R4/ST2, sgp130, and MCP-1 can serve as a prognostic marker for patients with glioblastoma. MC13-0007 Metabolomic profile of multiple myeloma patients P.C. Boutros 1 , E. Lalonde 1 , A.S. Ishkanian 2 , J. Sykes 2 , N. Moon 1 , G. Zafarana 2 , J. Thoms 2 , C.L. Have 2 , C. Malloff 3 , V.R. Ramnarine 2 , A. Meng 2 , D.F.Y. Mak 1 , J. Squire 4 , I. Jurisica 2 , M. Pintilie 2 , A. Dal Pra 2 , T. van der Kwast 2 , W.L. Lam 3 , M. Milosevic 2 , R.G. Bristow 2 . 2 Ontario Cancer Institute, University Health Network, Toronto, Canada; 3 Integrative Oncology, British Columbia Cancer Research Centre, Vancouver, Canada; 4 Pathology and Oncology, Queen’s University, Kingston, Canada L. Puchades-Carrasco 1 , R. Lecumberri 2 , J. Martínez-López 3 , J.J. Lahuerta 3 , M.V. Mateos 4 , F. Prósper 5 , J.F. San Miguel 4 , A. Pineda-Lucena 1 . 1 Structural Biochemistry Laboratory, Centro de Investigación Príncipe Felipe, Valencia, Spain; 2 Hematology Service, Clínica Universidad de Navarra, Pamplona, Spain; 3 Hematology Service, Hospital 12 de Octubre, Madrid, Spain; 4 Hematology Service, Hospital Universitario de Salamanca, Salamanca, Spain; 5 Hematology and Cell Therapy Area, Clínica Universidad de Navarra, Pamplona, Spain Background: Multiple myeloma (MM) remains an incurable disease. New approaches to develop better tools for improving patient prognostication as well as monitoring treatment efficacy are very much needed. Purpose/Objective: In this study, the capability of metabolomics by 1 H-NMR to characterize the metabolic profile of MM patients was evaluated. Materials and Methods: Serum metabolic profiles from MM patients at diagnosis (n=27) and after achieving complete remission of the disease (n=23) were acquired using 1 H-NMR. A matched control set of 31 serum samples from healthy individuals was also included. Results: Multivariable statistical modeling of the data showed that MM patients exhibit a specific serum metabolic profile (R2 =0.686; Q2 =0.462) characterized by lower concentrations of different lipids and some amino acids. A similar analysis performed in MM patients after achieving complete remission indicated that some of these changes are partially reverted upon responding to treatment (R2 =0.710; Q2 =0.358), thus reflecting that the metabolic profile of a MM patient gets closer to that of a healthy individual after the disappearance of major disease manifestations. Overall, the analysis of the variations in the metabolites found to play an important role in the discrimination between the groups included in this study revealed three different behaviours that could be related to disease progression, treatment response and achievement of complete remission. Conclusions: Our results highlight the potential of metabolomics by 1 H-NMR for identifying MM biomarkers that could be used to objectively discriminate individuals with and without MM, and monitor response to treatment. Furthermore, the results associated with the response to treatment are particularly relevant because, if validated in prospective studies, they could represent a starting point in the individualization of the therapeutic regimes for different patient groups. MC13-0008 A panel of four cytokines predict the prognosis of patients with glioblastoma Y. Lin 1 , Y. Wang 1 , T.A.O. Jiang 2 , Z. Guozhen 2 , J.I.N.G. Zhang 2 . 1st Hospital of China Medical University, Shenyang, China; 2 Neurosurgery, Beijing Neurosurgical Institute, Beijing, China 1 Neurosurgery, Background: A comprehensive evaluation of cytokine levels in patients with MC13-0009 A prognostic CNA signature sub-stratifies intermediate-risk prostate cancer Background: Men with prostate cancer (CaP) are stratified into low, intermediate and high risk groups based on clinical factors such as pre-treatment prostate-specific antigen (PSA) levels, tumour grade and tumour stage. Intermediate-risk patients vary widely in clinical outcome, with a 20–40% recurrence rate, as measured by a rise in post-treatment PSA concentration (biochemical recurrence). Purpose/Objective: In practice, there is no way to accurately identify the intermediate-risk patients that derive benefit from therapy. To address this issue, we developed prognostic signatures to further stratify these patients into sub-groups with distinct risk-profiles by applying machine learning to gene copy number profiles from intermediate risk patients. Materials and Methods: rray comparative genomic hybridization (aCGH) was applied to frozen biopsies from 126 intermediate-risk CaP patients prior to image-guided radiotherapy. Copy number aberrations (CNAs) were extracted and used to develop signatures which were then evaluated in an independent cohort of 129 low to intermediate risk patients treated by radical prostatectomy. Unsupervised and supervised machine-learning techniques were applied to identify disease subtypes and to develop a prognostic signature. Results: With unsupervised hierarchical clustering, we identified four distinct patient clusters within the radiotherapy cohort. Patients from the surgery cohort were matched to these clusters and the resulting clusters have statistically different biochemical recurrence rates. We also used a supervised learning approach with RandomForests to develop a CNA-signature. This signature is effective at identifying patients at risk of biochemical recurrence, while accounting for clinical covariates (HR=8.77, p=1.08×10−5 ). Finally, we found that including a measure of genomic instability in both models maintained the prognostic effects, suggesting that these measures contain independent prognostic information. Conclusions: We have recapitulated known genomic heterogeneity and shown that there are prognostic genomic subtypes within the intermediate-risk group. We also developed a clinically-relevant CNA-signature which stratifies intermediate-risk patients into two refined risk groups, independent of clinical features and genomic instability. The higher risk sub-group could be triaged to more aggressive systemic therapies given their rapid failure after local therapy alone. The lower risk sub-group may be over-treated, suggesting novel hypotheses for future clinical trials. In conclusion, we have identified a genomic biomarker which is promising in improving clinical management of intermediate-risk CaP patients. Poster Presentations MC13-0010 HLA and cytokinic markers in cervical cancer S. Zidi 1 , A.M.E.L. Mezlini 2 , H. Verdi 3 , Y. Yilmaz-Yalcin 3 , A. Yazici 4 , F. Atac 3 , B.Y. Loueslati 1 . 1 Biology, Faculty of Sciences of Tunis, ElManar, Tunisia; 2 Oncology, Salah Azeiz Oncology Institute, Tunis, Tunisia; 3 Medical Biology, Faculty of Medicine, Ankara, Turkey; 4 Biostatistics, Faculty of Medicine, Ankara, Turkey Background: It is increasingly evident that the variability in host immunogenetic background, especially in human major histocompatibility genes and pro-inflammatory cytokines, may modulate the susceptibility to Human Papilloma Virus infection and cervical cancer. Purpose/Objective: We have conducted a case-control study in Tunisian women to examine the effect of genetic variation in HLA class II DRB1 and DQB1, Interleukin (IL) 10, Interferon (INF)-γ, IL1-α, IL-1β and tumour necrosis factor (TNF)-α gene polymorphisms in cervical cancer susceptibility. Materials and Methods: Blood samples were collected from histopathologically confirmed patients of cervical cancer and unrelated healthy female controls of similar ethnicity. HLA genotyping was performed by PCR sequence specific primers technique. The IL 10-1082, and INFγ +874 polymorphisms were typed by Amplified Refractory Mutation Sequence PCR and the IL1-α +4845, IL-1β −511 polymorphisms were analyzed by PCR Restriction Fragment Length Polymorphism. Results: The data revealed significant positive and negative associations, suggesting either predisposing or protective effects of these genes in the disease outcome. DRB1*15, alone or linked to DQB1*06, was associated with a 2.7 and 3.5 fold increase in risk for cervical cancer, respectively. DRB1*13DQB1*03 showed a similar 3.5 risk effect. In contrast, only one haplotype – DRB1*13-DQB1*06 – provides evidence for a weak protection (about 0.3 fold reduction) of cervical cancer. At the polymorphic nucleotide 1082 of the IL-10 promoter and the IL1-α +4845 polymorphism, no differences were found between patients and control subjects. Women carrying IFN-γ+874 T/T genotype were at lesser risk of cervical cancer (about 0.5 fold reduction). However, IL-1β −511 the C/C genotype was associated with 2.3 increased risk for cervical cancer. Conclusions: Our results demonstrate a possible increased risk for cervical cancer associated with the DRB1*15 allele, DRB1*13-DQB1*03 haplotype and homozygote genotype IL-1β −511C/C and a possible protective effect with the DRB1*13-DQB1*06 haplotype and homozygote genotype IFN-γ +874 T/T. MC13-0012 Vitamin D downregulates biomarkers in the prostaglandin pathway E. Sauter, W.Q. Qin, C.S. Smith. Surgery, University of Texas HSC, Tyler, USA Background: Epidemiologic evidence suggests that women with higher circulating levels of vitamin (vit)D have a lower breast cancer risk, and preclinical studies suggest that vitD mediates growth inhibition through prostaglandin (PG) inhibition. Celecoxib, in clinical use to treat arthritis, inhibits PGs through a mechanism different than vitD. Purpose/Objective: Our objective was to determine if there was synergy between vitD and celecoxib in the cancer promoting PG pathway reduction in healthy women. Materials and Methods: 36 healthy women of normal cancer risk were randomized to receive daily for one month or one menstrual cycle one of four treatments: placebo, 400 international units (IU) vitD3, 2000 IU vitD3, or 2000 IU vitD3 + 400 mg celecoxib. Serum, plasma, nipple aspriate fluid (NAF), +/− mammary duct (MD) samples were collected. Serum and NAF were analyzed for PGE2 , serum for vitD3, plasma for celecoxib, and MD RNA for cyclooxygenase (COX)-2. Results: vitD3 levels increased in women receiving 2000 IU daily vitD (with or without celecoxib), but not in the other groups (p<0.01). Celecoxib increased (p<0.01) after treatment only in participants receiving the agent. PGE2 decreased in the breast after treatment with 2000 IU vitD3, but not in the other groups (p=0.01), and 2000 IU vitD3 worked better at lowering PGE2 than 2000 IU vitD3 + celecoxib (p=0.018). COX-2 decreased in the breasts of women taking 2000 IU vitD3, but not in the other groups. Conclusions: vitD3 alters the PG pathway in the breasts of healthy women in a dose dependent fasion. Adding celecoxib does not provide synergy. S15 MC13-0014 Cost-effectiveness of FDG-PET/CT for cytologically indeterminate thyroid nodules D. Vriens 1 , E.M.M. Adang 2 , R.T. Netea-Maier 3 , J.W.A. Smit 3 , J.H.W. de Wilt 4 , W.J.G. Oyen 1 , L.F. de Geus-Oei 1 . 1 Nuclear Medicine, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 2 Health Evidence, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 3 Medical Endocrinology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 4 Oncological Surgery, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands Background: Screening for malignancy in thyroid nodules (TNs) is performed using fine-needle aspiration cytology (FNAC). In approximately 25% of patients FNAC is inconclusive, mostly due to follicular neoplasia or cellular atypia, and diagnostic (hemi)thyroidectomy is proposed. This strategy results in futile surgery in 75% of these patients as histology reveals a benign TN. Our recent meta-analysis reported a high negative predictive value of 96% for FDG-PET/CT in this category of patients. As over a third of patients do have a negative FDG-PET/CT, we hypothesized that its routine use in FNAC-indeterminate TNs leads to better and more cost-effective patient care. In a decisionanalytic approach we set out to calculate the efficacy and economical value of FDG-PET/CT in order to support implementation. Purpose/Objective: In a decision analytic approach we set out to calculate the efficacy and economical value of FDG-PET/CT in order to support implementation. Materials and Methods: We developed an 8-state Markov decision model. Based on literature, reimbursement schedules of diagnosis/treatment combinations and expert panel opinion, we attributed distributions to the transition probabilities, costs and utility scores. Analysis of the model was performed by probabilistic sensitivity analysis for hypothetical adult patients with FNAC-indeterminate TNs over a duration of 5 years. Means and confidence intervals of discounted costs and QALYs were determined. Efficiency of FDG-PET/CT was presented by the incremental cost effectiveness ratio (ICER) in thousand euros per QALY gained (k€/QALY). One-way sensitivity analysis was performed over a plausible range for all variables. Results: Modifying current practice with routine use of FDG-PET/CT resulted in 47.9% fewer surgeries for benign nodules. Compared to surgery in all patients, the fraction of untreated cancers was 1.3%, similar as reported in literature. Over 5 years, mean discounted cost estimates were €8,934 (95% CI: €8,902–€8,965) for current practice and €8,102 (95% CI: €8,063–€8,140) with the routine use of FDG-PET/CT. Current practice and FDG-PET/CT produced no significantly difference in QALYs (4.56 and 4.60 QALY, respectively). FDG-PET/CT therefore saved €832 per 0.04 QALY gained (ICER: −25 k€/QALY): i.e. the dominant alternative. Only the utility attributed to observation after a negative FDG-PET/CT could lead to a negative effect on QALYs. At the minimum value tested (0.90), a FDG-PET/CT-driven approach would lead to 0.1 less QALYs over 5 years (ICER: +9 k€/QALY) compared to conventional work-up (utility of post-surgery surveillance: 0.99). Conclusions: Markov decision modeling showed the potential costeffectiveness of FDG-PET/CT in TN patients with inconclusive FNAC. A prospective randomized study is necessary to confirm these observations. MC13-0015 p14ARF methylation is a common event in the pathogenesis and progression of mixoid and pleomorphic liposarcoma J. Sopta 1 , R. Davidovic 2 , R. Kovacevic 1 , N. Lujic 3 , D. Ristic 4 . 1 Faculty of Medicine, Institute of Pathology, Belgrade, Serbia; 2 Faculty of Medicine, Institute Vinca, Belgrade, Serbia; 3 Faculty of Medicine, Institute Banjica, Belgrade, Serbia; 4 Faculty of Medicine, Institute for Cancer, Belgrade, Serbia Background: Liposarcoma represents the most abundant group of soft tissue sarcomas. The group can be divided into three different classes: differentiated/undifferentiated (WDLPS/DDLPS), mixoid/round cell (MLPS/RCLPS) and pleomorphic liposarcoma (PLS). It has become apparent that p53–p14 and Rb–p16 pathways play important roles in the pathogenesis of various sarcoma types. Molecular studies of the genes involved in these two pathways showed wide variations between the liposarcoma subtypes or even within the same subtype. Purpose/Objective: We sought to examine mutational status of p53 and methylation status of p16INK4a /p14ARF genes in primary and recurrent liposarcoma tumors. In order to follow mutational and epigenetic status of the selected genes eighteen primary and recurrent tumor tissue samples were selected. S16 Materials and Methods: In order to follow mutational and epigenetic status of the selected genes eighteen formalin fixed paraffin embedded (FFPE) primary and recurrent tumor tissue samples were selected. Eleven samples had pure mixoid histology, one was mixoid with <5% of round cell component and the others were PLS. Results: Immunohistochemical analysis revealed that p53 protein was overexpressed in 3/12 MLPS (25%) and 6/6 PLS (100%). Mutational analysis showed that two out of eleven MLPS (2/11, 18.2%) and two out of six PLS (2/6, 33.3%) contained mutated p53 gene. However, the frequencies of the p14ARF gene methylation were 83.3% (10/12) and 50% (3/6) in mixoid and pleomorphic group, respectively. Overall, 15 out of 18 (83.3%) samples had either p53 mutation or methylated p14ARF . Conclusions: The results from the current study suggest paramount importance of the p14ARF gene methylation in the pathogenesis and progression of mixoid and to a lesser extent pleomorphic liposarcoma. MC13-0017 Upregulation of TRF1 and TRF2 (telomere repeat binding factors) protein contributes to telomere shortening in renal cell carcinoma D. Pal 1 , U. Sharma 1 , R. Khajuria 1 , S.K. Singh 2 , N. Kakkar 3 , R. Prasad 1 . PGIMER, Chandigarh, India; 2 Urology, PGIMER, Chandigarh, India; 3 Histopathology, PGIMER, Chandigarh, India 1 Biochemistry, Background: TRF1 and TRF2 are telomere repeat binding proteins that are exclusively found at the telomeres. Both TRF proteins have a Myb-like helix-turn-helix domain in their carboxyterminus and a central conserved domain that includes sequences responsible for the formation of homodimers. The two proteins do not heterodimerize, and they differ substantially at the N terminus which is acidic in TRF1 but basic in TRF2. The protection of human telomeres crucially depends on these factors and it is reasonable to assume that the requirement for TTAGGG repeats at chromosome ends reflects the need for TRF1 and TRF2 binding. Telomere dysfunction is believed to be the significant factor in carcinogenesis. Purpose/Objective: To elucidate the carcinogenesis mechanism, the expression of TRF1, TRF2 and change in telomere length were investigated in renal cell carcinoma (RCC). Materials and Methods: Total 80 cases of RCC treated by surgery under advanced Urology services of Nehru Hospital, at Postgraduate Institute of Medical Education and Research, Chandigarh were included in the present study. For comparison, normal renal cortex samples were taken in each case. Transcriptional expression of TRF1 and TRF2 were estimated by real time PCR. Whereas Protein levels were detected by using immunohistochemical and immunofluorscence method. The mean telomere length was determined by southern blotting followed by hybridization. Results: The expression of TRF1 and TRF2 were significantly higher in the tumor tissue in comparison with normal renal parenchyma. The mean telomere length in RCC tissue was significantly shorter than that in normal renal parenchyma. The mean telomere length in all tissue samples were inversely correlated with the level of TRF1 and TRF2 expression. Conclusions: Our result suggests that the upregulation of TRF 1 and TRF2 may work to reduce the telomere length in RCC and could contribute to the carcinogenesis of renal cell carcinoma. MC13-0018 Circulating miRNA-148a as a predictive biomarker for early relapse of UICC stage II and III colorectal cancer patients following curative resection J.Y. Wang, H.L. Tsai. Cancer Center and Surgery, Kaohsiung Medical University and Hospital, Kaohsiung, Taiwan Background: The recurrence of colorectal cancer (CRC) is frequent within the first year of curative resection surgery and may be unavoidable. MicroRNA-148a is proven to be an oncomiRNA or a tumor suppressor miRNA in various cancers. Purpose/Objective: We recently identified microRNA-148 (miRNA-148) as a predictor of early recurrence in CRC. In the present study, we further investigated the function and serum level of miRNA-148 in relation to early recurrence of CRC. Materials and Methods: First we further confirmed overexpression of miRNA-148 in non-early relapse subjects. Gain-of-function in vitro studies were used to evaluate the effect of miRNA-148 on cell proliferation, migration, invasion, and cell cycle progression. The colon cancer cell line Caco2 and Poster Presentations a stable clone overexpressing miRNA-148 were xenografted to evaluate the in vivo effect of miRNA-148 in null mice. Finally, circulating miRNA-148 was investigated as a potential biomarker for identifying early relapse. Results: The results demonstrate that miRNA-148a overexpression is related to early relapse of CRC through suppression of cell proliferation, migration, and accumulation in the G2 phase but also to a better disease-free survival and overall survival. The serum miRNA-148 increased significantly in early relapsed patients compared to non-early elapsed patients (P =0.015). Conclusions: miRNA-148 shows anti-tumorigenesis activity, and preoperative circulating miRNA-148 levels can be used to predict postoperative early relapse of CRC. MC13-0019 Tumor microenvironment of metastasis: An imaging based marker of risk for distant metastasis of breast cancer J. Jones 1 , X. Xue 2 , H. Lin 3 , M. Oktay 1 , B. Robinson 4 , F. Gertler 5 , A. Glass 6 , J. Sparano 7 , J. Condeelis 8 , T. Rohan 2 . 1 Pathology, Albert Einstein College of Medicine, Bronx, USA; 2 Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, USA; 3 Health Evidence and Policy, Mount Sinai School of Medicine, New York, USA; 4 Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, USA; 5 Koch Institute, Massachusetts Institute of Technology, Boston, USA; 6 Health Research, Kaiser Permanente Northwest, Portland, USA; 7 Oncology, Montefiore Medical Center, Bronx, USA; 8 Anatomy and Structural Biology, Albert Einstein College of Medicine, Bronx, USA Background: Hematogenous metastasis is the principal cause of death in breast cancer. Intravital imaging in mouse models has identified intravasation sites called TMEM (Tumor MicroEnvironment of Metastasis). Previously, we confirmed TMEMs in human breast cancer samples and observed that increased TMEMs are associated with increased metastatic risk. Purpose/Objective: The purpose of this study was to assess TMEM as a marker of metastatic risk in a large cohort study of breast cancer patients, and to compare TMEM with IHC4, a score that provides prognostic information comparable to that provided by Oncotype DX® . Materials and Methods: TMEM is a microanatomic structure in which there is the direct contact of an endothelial cell, a perivascular macrophage, and an invasive tumor cell. TMEMs are assessed in whole tissue sections by a sequential triple immunostain for the 3 cell types: macrophages (CD68), endothelial cells (CD31), and invasive tumor cells (pan-Mena). IHC4 (ER, PR, Her2, and Ki67) was assessed using standard methods. Stained slides were read by pathologists blinded to outcome. The population for this study was a nested case-control study of 259 case-control pairs diagnosed between 1980 and 2000, chosen from a total cohort of 3760 invasive ductal carcinomas, and followed through 2010. Cases were women who ultimately developed distant metastasis; controls, women who did not. Odds ratios (OR) (95% CI) for the association of metastasis with TMEM were estimated overall and for 3 subgroups (ER+/Her2−, Her2+, and triple negative). The ORs for TMEM were compared with those for IHC4. Where TMEM was significantly associated with risk we performed a receiver operating characteristic analysis (ROC) and computed the area under the curve (AUC). Results: TMEM was associated with an almost 3 fold increased risk of metastasis in ER+/Her2− tumors (multivariate OR (95% CI)high vs. low tertile = 2.70 (1.39–5.26), Ptrend = 0.004), whereas IHC4 had a borderline positive association with risk (OR10 unit increase = 1.06 (1.00–1.13)). The association for TMEM persisted after adjustment for IHC4. There were no significant associations with the other two subgroups. Conclusions: TMEM score is a novel, mechanism-based assay that is positively associated with risk of distant metastasis in women with ER+/Her2− breast cancer and provides prognostic information that is complementary to IHC4 and other clinicopathologic risk factors. Strategies for automating TMEM scoring, and additional validation studies, including assessment of clinical utility, are underway. Poster Presentations MC13-0020 Targeted next-generation sequencing (NGS) of circulating tumor cells (CTCs) and matched primary tumors S.V. Liu 1 , P.W. Dempsey 2 , W. Strauss 2 , Y. Xu 1 , T. Xu 1 , T.J. Triche 3 , D.I. Quinn 1 , A. Goldkorn 1 . 1 Internal Medicine/Genitourinary Oncology, University of Southern California, Los Angeles, USA; 2 Bioengineering R&D, Cynvenio Biosystems, Westlake Village, USA; 3 CHLA Genomics Core Facility, University of Southern California, Los Angeles, USA Background: Personalizing cancer care relies on accurate detection of actionable genomic aberrations in tumor cells. Conventionally, this strategy relies on analysis of primary tumor samples, which are often temporally and biologically distinct from recurrent, metastatic or treatment-resistant disease. As an alternative, CTCs offer real-time cancer tissue for analysis that may more accurately represent the current state of a patient’s disease. Purpose/Objective: In this pilot, we used CTCs as source material for targeted NGS across a range of malignancies. Materials and Methods: Under IRB approval, blood samples from patients with advanced cancer were labeled with EpCAM ferrofluid and placed into the LiquidBiopsy® platform (Cynvenio Biosystems, Inc.), an immunoaffinity-based microfluidic device tailored to query genomic events. CTCs were identified by CK, CD45 and DAPI expression. A matched WBC pellet served as a control representing germline sequence. Amplicon libraries were generated using Life Technologies AmpliSeq 2.0 and sequenced with an Ion Torrent sequencer. When available, matched formalin fixed paraffin embedded (FFPE) primary tumor tissue from the same patient was analyzed in parallel. Somatic single nucleotide variants (SNV) present in CTCs or FFPE samples but not in WBC were identified. Results: CTCs were detected in 18 of 19 patients with advanced prostate (8), breast (6), renal cell (2), bladder (1), lung (1) and rectal (1) cancer (CTC median 54, range 15–421). Germline SNPs were consistently detected across WBC, CTC and FFPE samples. Significant SNVs (occurring in >1% of DNA in a sample) were found in 6 of 18 patient CTC samples (range 1–6 SNVs/sample, frequency 1.1–11.9% with 620X–14,422X sequence coverage depth). Numerous SNVs were identified in all 9 matched primary tumor FFPE specimens but did not correlate with the SNVs identified in the CTCs. Conclusions: This pilot demonstrates the feasibility of using CTCs as a real-time disease relevant substrate for NGS to identify personalized genomic targets. A high number of CTCs were detectable across malignancies, and CTC germline variants correlated with matched WBC controls. Cancer relevant SNVs were detected in a third of patients even using the relatively narrow primary tumor derived AmpliSeq platform. The FFPE specimens generated a high number of SNVs but did not correlate with CTC profiles, likely reflecting biological disparity between early localized tumors and advanced metastatic disease, as well as genomic artifacts introduced into primary tumor specimens by FFPE preservation. These data demonstrate the feasibility and potential biological and technical advantages of CTCs over traditional FFPE samples for genomic analysis in the pursuit of personalized cancer medicine. MC13-0021 Impact of post-protocol anti-epidermal growth factor receptor therapy on survival in wild-type KRAS/NRAS metastatic colorectal cancer: Data from the PRIME study M. Peeters 1 , J.Y. Douillard 2 , S. Siena 3 , T. Price 4 , J. Tabernero 5 , R. Sidhu 6 , S. Braun 7 , A. Rong 8 . 1 Department of Oncology, University Hospital Antwerp (UZA), Antwerp, Belgium; 2 Department of Medical Oncology, Centre René Gauducheau, Nantes, France; 3 Department of Haematology and Oncology, Ospedale Niguarda Cà Granda, Milan, Italy; 4 Haematology/Medical Oncology Unit, Queen Elizabeth Hospital and University of Adelaide, Woodville, Australia; 5 Department of Medical Oncology, Vall d’Hebron University Hospital, Barcelona, Spain; 6 Department of Clinical Development, Amgen Inc., Thousand Oaks, USA; 7 Department of Medical Development – Oncology, Amgen GmbH, Zug, Switzerland; 8 Department of Biostatistics, Amgen Inc., Thousand Oaks, USA Background: Updated intent-to-treat (ITT) overall survival (OS) results with longer follow-up and higher OS event attainment from the phase III, 1st-line PRIME study demonstrated improved OS in patients with wild-type (WT) KRAS ( exon 2) metastatic colorectal cancer (mCRC) receiving panitumumab (pmab) + FOLFOX4 vs FOLFOX4 alone (23.8 vs 19.4 months, respectively, hazard ratio [HR] 0.83; p=0.03 [Douillard et al. J Clin Oncol 2013; 31{Suppl; abstract 3620}). New data from this trial suggest that mCRC patients with S17 mutations in KRAS or NRAS beyond KRAS exon 2 may not respond to pmab treatment (Oliner et al, J Clin Oncol 2013;31(Suppl): abstract 3511). Purpose/Objective: To assess the OS impact of crossover to post-protocol anti-epidermal growth factor receptor (EGFR) therapy in the control arm of PRIME for patients with wild-type (WT) KRAS/NRAS exons 2–4 (RAS ) mCRC. Materials and Methods: In PRIME, patients with no prior chemotherapy for mCRC were randomised 1:1 to pmab 6.0 mg/kg Q2W + FOLFOX4 or FOLFOX4. ITT OS analyses and inverse probability-of-censoring weighted (IPCW) analyses were performed. IPCW censors patients at the time of subsequent anti-EGFR therapy use and estimates HRs that are valid in the presence of informative censoring, enabling imputation of OS for crossover patients using OS data from those who did not crossover. Data from a post-hoc OS update conducted when ≥80% of patients had an OS event were analysed for the WT RAS population overall and in the subgroup of patients with an Eastern Cooperative Oncology Group (ECOG) score of 0/1 (stratification factor). Results: See table. OS update (WT RAS ) Pmab + FOLFOX4 FOLFOX4 Number of patients 259 253 ITT analysis of OS; median, months (95% CI) Overall 25.8 (21.7–29.7) 20.2 (17.6–23.6) 0.77 (0.64–0.94); 0.009 HR (95% CI); p-valuea ECOG 0/1 25.4 (21.4–28.7) 20.5 (18.1–23.2) HR (95% CI); p-valuea 0.79 (0.66–0.94); 0.0089 IPCWb–d analysis of impact of post-protocol anti-EGFR therapy, HR (95% CI); p-value Overall 0.69 (0.50–0.95); 0.024 ECOG 0/1 0.60 (0.43–0.84); 0.031 a Descriptive p-value; b Rimawi & Hilsenbeck, J Clin Oncol 2012; c Colleoni et al, J Clin Oncol 2011; d Robins & Finkelstein, Biometrics 2000. CI = confidence interval. Conclusions: To improve outcomes in mCRC, it is important to continue to refine the optimum patient population by assessing the OS impact of new biomarkers. IPCW may also be particularly suited for detecting OS benefits beyond those detected with an ITT approach that ignores selective crossover/drop-in bias. MC13-0022 Resection rates and survival in patients with wild-type KRAS/NRAS metastatic colorectal cancer and liver metastases: Data from the PRIME study M. Peeters 1 , J. Tabernero 2 , J.Y. Douillard 3 , S. Siena 4 , C. Davison 5 , S. Braun 6 , R. Sidhu 7 , K. Öhrling 8 . 1 Department of Oncology, University Hospital Antwerp (UZA), Antwerp, Belgium; 2 Department of Medical Oncology, Vall d’Hebron University Hospital, Barcelona, Spain; 3 Department of Medical Oncology, Centre René Gauducheau, Nantes, France; 4 Department of Haematology and Oncology, Ospedale Niguarda Cà Granda, Milan, Italy; 5 Department of Biostatistics, Amgen Ltd, Cambridge, United Kingdom; 6 Department of Medical Development – Oncology, Amgen GmbH, Zug, Switzerland; 7 Department of Clinical Development, Amgen Inc., Thousand Oaks, USA; 8 Department of Medical Development – Oncology, Amgen AB, Solna, Sweden Background: Hepatic resection has become a standard treatment option for patients with metastatic colorectal cancer (mCRC) and colorectal cancer liver metastases (CLM). Data are required on the proportion of mCRC patients identified using refined molecular staging (e.g. wild-type [WT] KRAS/NRAS [RAS ] status) and treated with aggressive 1st-line therapy (e.g. panitumumab [pmab] + FOLFOX4) who become eligible for potentially curative hepatic resection. Data on the rate of successful conversion from unresectable to resectable CLM (e.g. by achieving early tumour shrinkage by week 8) could help guide multidisciplinary treatment decisions for patients with mCRC and CLM. Purpose/Objective: To evaluate resection rates and overall survival (OS) outcomes for all patients with WT RAS mCRC and CLM who were treated in the PRIME study. Materials and Methods: In PRIME, patients with no prior chemotherapy for mCRC and unresectable disease at baseline were randomised 1:1 to pmab 6.0 mg/kg Q2W + FOLFOX4 or FOLFOX4. Using data from an exploratory analysis conducted when ≥80% of patients had an OS event, the proportion of patients achieving ≥30% tumour shrinkage/response by week 8 and S18 Poster Presentations rates of metastasectomy/complete resection were analysed in CLM patients with WT RAS (KRAS/NRAS exons 2–4 assessed) mCRC. Median OS and 3-year OS rates for WT RAS CLM patients receiving each treatment were also evaluated. The data were summarised descriptively and tested for significance using Cox’s proportional hazards models. Results: Baseline characteristics were generally similar between treatment groups. For efficacy results, see table. WT RAS CLM patients (n=89) Pmab+FOLFOX4 (n=48) ≥30% tumour shrinkage at week 8, n (%) Metastasectomy, n (%) Complete resection, n (%) Median OS, months Hazard ratio (95% confidence intervals) 3-year OS rate, % P-valuec FOLFOX4 (n=41) 34 (79)*a 18 (51)*b 15 (31) 9 (22) 14 (29) 7 (17) 40.7 33.4 0.71 (0.43–1.16) 55 44 0.015 0.349 0.216 0.174 *Patients assessed at baseline and Wk 8; a N=43; b N=35; c Descriptive p-value. Conclusions: In this post-hoc analysis from PRIME including 1st-line mCRC patients with initially unresectable WT RAS CLM, there was a greater reduction in tumour burden for pmab+FOLFOX4 vs FOLFOX4 at week 8 as well as numerically higher rates of metastasectomy, complete resection and 3-year OS. Pmab+FOLFOX4 treatment resulted in conversion of almost one-third of initially unresectable CLM patients, enabling metastasectomy in 31% and complete resection in 29%. MC13-0023 Revisiting TFF1 and TFF3 as biomarkers in breast cancers: A 246 cases study S. Delpous 1 , N. Reix 2 , A. Welsh 3 , C. Wendling 1 , F. Alpy 1 , J.M. Lessinger 2 , M.P. Chenard 3 , M.C. Rio 1 , C. Tomasetto 1 , C. Mathelin 4 . 1 Functional Genomics and Cancer, IGBMC, Illkirch, France; 2 Biochemistry, CHU, Strasbourg, France; 3 Pathology, CHU, Strasbourg, France; 4 Senology, CHU, Strasbourg, France Background: A better knowledge of mammary tumorigenesis allows the development of personalized patient care. For example, adjuvant chemotherapy is not justified for all patients. While it is beneficial for those at high risk of relapse, it might be detrimental for other patients. This therapeutic decision is based on several criteria and can be difficult in some cases. Therefore, it is necessary to develop new tools predicting risk of recurrence in early stage breast cancer. TFF1 and TFF3 proteins represent potential biomarkers. Indeed, TFF1 and TFF3 are two related proteins induced by estradiol, and expressed at high levels in some breast cancers. Studied in isolation, their value as prognostic markers remains controversial. Purpose/Objective: Our objective is to evaluate the prognostic potential value of combined TFF1 and TFF3 dosages. Materials and Methods: We analyzed by Western Blot (WB) both TFF1 and TFF3 expression in a prospective cohort of 246 human invasive breast carcinomas. In parallel, TFF1 and TFF3 expression was measured by ImmunoHistoChemistry (IHC) in 109 of them. An immunoscore for WB and a histoscore for IHC were established. Associations between TFF1 and TFF3 expressions with various clinico-pathological features were determined using GraphPad software. Results: Immunoscores and histoscores were significantly correlated both for TFF1 and for TFF3. Both methods appear complementary since IHC is highly sensitive and WB is highly specific. Unexpectedly, there is not a complete overlap between TFF1 and TFF3 expression in breast cancers. Tumors double positive for TFF1 and TFF3 were associated with lympho-vascular invasion and lymph node involvement compared to TFF1 or TFF3 simple positive tumors. Conclusions: In contrast with previous published reports, TFF1 and TFF3 can be expressed independently from each other. Of interest, this study indicates that TFF1 and TFF3 may distinguish subtypes within the luminal group. Recent functional data showed TFF1 and TFF3 are involved in cell migration and invasion. In addition, a microarray study showed TFF1 and TFF3 were in a five-gene signature predicting bone metastasis. Along the same line, our results show TFF1 and TFF3 expression are associated with some poor prognostic indicators. Taken together, these data suggest the TFF1 and TFF3 double positive tumors should be considered for adjuvant chemotherapy. They definitively show TFF1 and TFF3 expression in breast cancer merits further investigation on larger cohorts. MC13-0024 Impact of post-progression anti-vascular endothelial growth factor-containing therapy on survival in patients with metastatic colorectal cancer: Data from the PRIME study M. Peeters 1 , J.Y. Douillard 2 , S. Siena 3 , T. Price 4 , J. Tabernero 5 , R. Sidhu 6 , S. Braun 7 , C. Davison 8 . 1 Department of Oncology, University Hospital Antwerp (UZA), Antwerp, Belgium; 2 Department of Medical Oncology, Centre René Gauducheau, Nantes, France; 3 Department of Haematology and Oncology, Ospedale Niguarda Cà Granda, Milan, Italy; 4 Haematology/Medical Oncology Unit, Queen Elizabeth Hospital and University of Adelaide, Woodville, Australia; 5 Department of Medical Oncology, Vall d’Hebron University Hospital, Barcelona, Spain; 6 Department of Clinical Development, Amgen Inc., Thousand Oaks, USA; 7 Department of Medical Development – Oncology, Amgen GmbH, Zug, Switzerland; 8 Department of Biostatistics, Amgen Ltd, Cambridge, United Kingdom Background: Refining the identification of eligible patients (pts) for treatment with agents targeting EGFR and VEGF pathways and the optimum sequencing of biologic agents combined with chemotherapy remain relevant for the treatment of metastatic colorectal cancer (mCRC). Continuing anti-VEGF therapy from 1st-line through 2nd-line therapy resulted in median overall survival (OS) of 23.9 months in unselected mCRC pts (Bennouna et al Lancet Oncol 2012). Purpose/Objective: To explore OS outcomes for pts in the PRIME study who had been identified as having wild-type KRAS/NRAS (WT RAS ) mCRC and received 2nd-line therapy that did/did not include an anti-VEGF agent. Materials and Methods: In PRIME, 1183 pts with no prior chemotherapy for mCRC were randomised to panitumumab (pmab) 6.0 mg/kg Q2W + FOLFOX4 or FOLFOX4. Using data from an exploratory analysis conducted when ≥80% of pts had an OS event, median OS from start of 1st-line therapy was calculated for pts with WT RAS (KRAS/NRAS exons 2–4 assessed) mCRC who did/did not receive post-progression (PD) anti-VEGF therapy. Data were summarised descriptively and tested for significance using Cox’s proportional hazards models. Results: Overall, 505 pts had WT RAS mCRC and 346 also received post-PD therapy. Of these, 100 (29%) received post-PD anti-VEGF therapy and 246 (71%) did not; median OS was longer for those receiving subsequent anti-VEGF therapy (38.1 vs 23.6 months, respectively; HR 0.63 [95% CI 0.48–0.81]; p<0.0004). Similar results were found irrespective of 1st-line treatment received, but median OS was longer when anti-VEGF agents were used after 1st-line pmab+FOLFOX4 treatment (table). Pmab+FOLFOX4 Post-PD anti-VEGF therapy Yes No FOLFOX4 Yes No Pts, n 55 114 45 132 Median OS, months 40.0 26.0 36.2 20.6 Yes vs no anti-VEGF therapy HR (95% CI) 0.64 (0.44–0.94) 0.62 (0.44–0.89) Descriptive p-value 0.0211 0.0101 Pmab+FOLFOX4 vs FOLFOX4 – with post-PD anti VEGF HR (95% CI) 0.64 (0.41–1.00) Descriptive p-value 0.0494 Pmab+FOLFOX4 vs FOLFOX4 – without post-PD anti VEGF HR (95% CI) 0.69 (0.53–0.92) Descriptive p-value 0.0096 Conclusions: Treatment of WT RAS mCRC pts with pmab+FOLFOX4 followed by an anti-VEGF agent produced a median OS of 40 months in these exploratory analyses. We acknowledge bias in our results since, pts were not randomised to 2nd-line treatment and, in accordance with Bennouna et al, deaths and dropouts prior to 1st-line PD are not considered. However, prospective confirmation of this treatment strategy may be of value to clinical practice. MC13-0025 TP53 gene polymorphism and its Pro/Pro variant is potentially contributing to cancer susceptibility in epithelial ovarian carcinoma patients from North Indian population R. Mir, Z. Mariyam, D. Sagar, A. Imtiyaz, R. Prakash, G. Gauri, N. Khurani, Y. Prasant, J. Jamsheed, F. Shazia, S. Alpana. Cancer Genetics Lab, Maulana Azad Medical College and Associated Hospitals, New Delhi, India Background: Ovarian cancer is the leading cause of death from gynecological malignancies. The early stages of this disease are asymptomatic Poster Presentations and more than 75% of the cases are diagnosed with regional or distant metastases. P53 is a tumor suppressor gene and is involved in the etiology of ovarian cancer. Studies investigating the associations between the p53 codon 72 polymorphism and ovarian cancer risk showed conflicting results. A polymorphism at codon 72 of the human tumour-suppressor gene, p53, results in translation to either arginine or proline. Purpose/Objective: To investigate the association of p53 codon 72 polymorphism with susceptibility to epithelial ovarian cancer in North Indian women and to correlate them with clinicopathological characteristics of disease. Materials and Methods: The study was conducted on 100 epithelial ovarian cancer patients and 100 healthy controls. Genotyping of p53 codon 72 polymorphism was examined by PCR with allele-specific primers. Results: The proportions of individuals homozygous for the arginine allele, homozygous for the proline allele, and heterozygous for the two alleles were 33%, 17%, and 50% among women screened for ovarian cancer; 62%, 6%, and 32% among the control group. A significant correlation was found between the Arg/Pro (p<0.0004) and Pro/Pro (p<0.0006) genotypes with respect to the Arg/Arg genotype. Pro/Pro genotype emerged as the risk factor with an OR of 5.3 and a RR of 2.5. Conclusions: Our study suggests that Pro/Pro genotype of 72 codon polymorphism could be an independent susceptibility marker in northern Indian women with ovarian carcinomas. MC13-0027 Deregulation of cell polarity protein Lgl2 correlates with gastric cancer progression and loss of E-cadherin K. Nam 1 , M.A. Kim 2 , G. Choe 1 , W.H. Kim 2 , H.S. Lee 1 . 1 Pathology, Seoul National University Bundang Hospital, Seongnam-si, Korea; 2 Pathology, Seoul National University Hospital, Seoul, Korea Background: Lethal giant larvae 2 (Lgl2) is a tumor suppressor that regulate epithelial cell polarity and proliferation at the basolateral domain. The role of Lgl2 on cancer progression has been suggested with relation to epithelial-mesenchymal transition (EMT) in epithelial cancers. Purpose/Objective: The present study aimed to evaluate the clinical and prognostic significance of Lgl2 expression in gastric cancer and its correlation with EMT marker expression. Additionally, we have evaluated Lgl2 expression in non-neoplastic lesions including normal, active gastritis and intestinal metaplasia to whether Lgl2 may be involve in gastric carcinogenesis. Materials and Methods: To investigate the clinical implication of Lgl2 in gastric cancer, 409 cases of surgically resected gastric cancers and 154 cases of endoscopically removed neoplastic or preneoplastic lesions were examined. Immunohistochemistry of Lgl2, E-cadherin, β-catenin, MMP2, S100A4, Snail-1, vimentin and ZEB-1 was performed. The expression of Lgl2 mRNA was examined using RNA in situ hybridization method. Results: Loss of membranous Lgl2 staining was correlated with diffuse type, lymph node metastasis and advanced pTNM stage (p<0.05). Furthermore, loss of membranous Lgl2 expression in adenocarcinoma was associated with reduced E-cadherin expression (p<0.001). Among EMT markers, S100A4, MMP2 and Lgl2 expression was associated with poor survival (p<0.05). In addition, combined analysis of Lgl2, S100A4 and MMP2 allows more precise estimation of prognosis in gastric cancer. During gastric carcinogenesis, membranous expression of Lgl2 was progressively lost in 4% of normal mucosa, 75% of intestinal metaplasia, 58% of gastric dysplasia, 69% of intestinal type gastric cancer and 96% of diffuse type gastric cancer. The expression of Lgl2 protein was concordant with mRNA expression status (p<0.001). Conclusions: Our data suggest that Lgl2 might function as a tumor suppressor in gastric cancer and its mislocalization could be important in carcinogenesis and cancer progression. S19 role in cancer invasion and metastases. It degrades the extracellular matrix, allowing cancer cells to invade the surrounding tissue. The uPA receptor, uPAR, is a three domain protein anchored to the surface of cells. Soluble cleaved and intact forms of uPAR are continuously released into circulation. Previous studies have shown that the forms of soluble uPAR (suPAR) have prognostic value in a number of different malignancies, but none has investigated the prognostic value of suPAR in patients with cholangiocarcinoma. Purpose/Objective: To investigate the prognostic value of suPAR’s different forms in patients with cholangiocarcinoma recieving chemotherapy. Materials and Methods: A cohort of consecutive patients with inoperable cholangiocarcinoma was established retrospectively from two separate institutions using exact similar treatment guidelines. Serum samples from the patients were collected from a common biobank. The suPAR forms were measured by the use of three time-resolved fluorescence immunoassays (TR-FIAs) in serum samples taken before start of chemotherapy. A multivariable Cox regression analysis of overall survival (OS) was done including the suPAR forms, CA19-9, age, gender, locally advanced/metastatic disease, performance status and institution. Results: One hundred and sixty-eight patients treated with chemotherapy and with a baseline serum sample were identified. Seventy percent had WHO performance status 0 +1. The median age was 64.5 (range 24.1–78.7). Patients received a median of six cycles of chemotherapy (range 0.5–20). The levels of intact+cleaved suPAR forms (suPAR(I–III)+(II–III)) at baseline (entered as a continuous variable log transformed base 2) was an independent predictor of survival (HR 1.93 (95% CI; 1.41–2.64), p<0.0001) in a Cox multivariable regression model. In a multivariable model not including suPAR(I–III)+(II–III), uPAR(I) was an independent predictor of OS (HR=1.25, 95% CI: 1.06–1.47, p=0.007) and similarly for suPAR(I–III) (HR=1.32, 95% CI: 1.01–1.73, p=0.040). The inclusion of CA19-9 (n=114) at baseline, showed CA19-9 to also be a predictor of OS (HR=1.10 (95% CI: 1.01–1.19) p=0.023) independent of suPAR(I–III)+(II–III). Conclusions: Intact + cleaved serum suPAR (suPAR(I–III)+(II–III)) is an independent prognostic marker for OS in patients with inoperable cholangiocarcinoma. The other uPAR forms were also significant predictors of OS but were not independent of suPAR(I–III)+(II–III). Our results warrant validation in an independent cholangiocarcinoma patient cohort. MC13-0029 Diffusion Weighted Imaging (DWI) as a biomarker for evaluation of neoadjuvant treatement in localy advanced breast cancer S. Drisis, M. Ignatiadis, K. Stathopoulos, S.L. Chao, M. Lemort. Radiology, Institut Jules Bordet, Brussels, Belgium Background: The use of preoperative chemotherapy is nowadays the standard treatment not only for locally advanced but also for operable breast cancer with tumour size ≥2 cm. Since the field of new drug development is under continuous improvement, surrogate markers that would predict or early assess the response to treatment became essential. Purpose/Objective: To examine if diffusion weighed imaging (DWI) could MC13-0028 Prognostic significance of serum levels of intact and cleaved forms of urokinase plasminogen activator receptor (suPAR) in patients with inoperable cholangiocarcinoma M. Grunnet 1 , I.J. Christensen 2 , U. Lassen 1 , L.H. Jensen 3 , M. Lydolph 4 , I.K. Lund 2 , G. Hoyer-Hansen 2 , N. Brünner 5 , M. Sorensen 1 . 1 Oncology, Rigshospitalet, Copenhagen, Denmark; 2 Finsen Laboratory, Rigshospitalet, Copenhagen, Denmark; 3 Oncology, Lillebaelt Hospital, Vejle, Denmark; 4 Biochemistry and immunology, SSI, Copenhagen, Denmark; 5 Veterinary Pathobiology, University of Copenhagen, Copenhagen, Denmark Background: The urokinase plasminogen activator (uPA) system plays a vital Figure 1. Subtraction image (A) and diffusion weighted image (DWI) (B) at EX1 for a responder. At EX2 no enhancement is detected in subtraction image (C) and the ADC value is increased at DWI (D). S20 serve as a biomarker for evaluation of neoadjuvant chemotherapy (NAC) in localy advanced breast cancer. Materials and Methods: A retrospective study was performed, including 29 patients who received NAC and underwent two MRI examinations, one before (EX1) and one during NAC (EX2). The apparent diffusion coefficient (ADC) was measured at both examinations. Postoperative pathological examination was used as the golden standard for the evaluation of tumour response with a binary model: responders and non-responders according to the presence of residual vital tumour. Statistical analysis was performed using Wilcoxon rank sum test. Results: From the 29 patients 21 were non-responders (72.5%) and 8 showed complete pathological response (27.5%). There was no significant difference between responders and non-responders for EX1 (p=0.96) but a significant difference (p=0.004) was found for EX2 between the two response groups. After treatment, responders (mean ADC = 0.865) showed higher values than non-responders (mean ADC = 0.414) indicating higher cellularity in the non-responders group. Moreover, the relative difference of ADC values between the two time points was statistically different for the two response groups showing more important increase of the ADC value for the responder group. Conclusions: DWI-MRI could have a potential monitoring role in neoadjuvant treatment for locally advanced breast cancer. Responders showed higher ADC values than non-responders at EX2 and more important increase of the ADC value between EX1 and EX2. MC13-0030 Regulation of hepatocyte growth factor-mediated cell proliferation and apoptosis through high-motility group A1 in stomach cancer cells H. Lee 1 , K.H. Lee 2 . 1 Division of Hematology and Oncology Department of Internal Medicine, Dongnam Institute of Radiological and Medical Sciences, Busan, Korea; 2 Division of Hematology and Oncology Department of Internal Medicine, Yeungnam University College of Medicine, Daegu, Korea Background: High-motility group A (HMGA) genes are frequently reactivated in many types of human cancers, and over-expression of HMGA proteins is linked to malignant transformation and progression in human cancers. However, the molecular mechanisms by which HMGA1 participates in gastric carcinogenesis needs to be further analyzed. Purpose/Objective: In this study, the role of HMGA1 in invasion and metastasis was investigated and the mechanisms of action were elucidated. Materials and Methods: HMGA1 upregulation by hepatocyte growth factor (HGF) was further confirmed in NUGC3 and MKN28 gastric cancer cells by RT-PCR and Western blotting. To determine the role of HMGA1 in HGFmediated cell proliferation, cell invasion and apoptosis, stable HMGA1-shRNA cells and the control cells were prepared and treated with or without HGF. Results: HMGA1 knockdown decreased a decrease in HGF-mediated cell proliferation and invasion, but increased apoptotic cell death. HMGA1-shRNA cells showed higher level of p53 protein and lower level of Bcl-2 protein than the control cells. HMGA1 knockdown abolished HGF-mediated HMGA1 binding to the Bcl-2 promoter and thus the activation of Bcl-2 promoter by HGF. HMGA1 knockdown also inhibited HGF-mediated Akt phosphorylation. Conclusions: The results suggest that HGF-mediated upregulation of HMGA1 might be involved in cell growth, cell invasion, and apoptosis in gastric cancer through the regulation of p53, Bcl-2 and Akt activation and therefore, HMGA1 might be the potential therapeutic target for gastric cancer treatment. MC13-0031 Tumour miR-210 expression is elevated in malignant pleural mesothelioma patients with shorter survival undergoing extrapleural pneumonectomy M.B. Kirschner 1 , Y.Y. Cheng 1 , S.C. Kao 1 , B.C. McCaughan 2 , N. van Zandwijk 1 , G. Reid 1 . 1 Asbestos Diseases Research Institute, Asbestos Diseases Research Institute, Sydney, Australia; 2 Cardiothoracic Surgical Unit, Royal Prince Alfred Hospital, Sydney, Australia Background: Malignant pleural mesothelioma (MPM) is an aggressive cancer with a median survival of around one year. A selected group of patients with a potentially resectable tumour mass may be considered for extrapleural pneumonectomy (EPP), however the results of this form of treatment are variable. Purpose/Objective: In the present study we used microarray profiling to Poster Presentations identify microRNAs which might have the potential to serve as a prognostic biomarker in patients eligible for EPP. Materials and Methods: The study used 60 formalin-fixed paraffin embedded (FFPE) tumour tissues from MPM patients who underwent EPP and had sufficient tumour for RNA extraction, a series which had been previously used to assess the prognostic value of the NLR. MicroRNA microarray profiling was performed on the 8 patients with longest (median: 53.7 months) and the 8 patients with shortest (median: 6.4 months) survival. Candidate microRNAs were independently validated using TaqMan assay-based microRNA-specific RT-qPCR. Levels of validated candidates were then assessed by RT-qPCR in 44 additional tumour samples. The median relative expression level of each candidate was used as cut-off to determine high and low expression for examination using the Kaplan-Meier method. Individually significant (p<0.05) variables were entered into a multivariate model together with the established risk factors age, gender, histological subtype, NLR. Results: Microarray profiling identified 16 microRNAs with differential expression between long-term and short-term survivors RT-qPCR validation found levels of miRs-30e, -93, -106b, -210, and -222 to be significantly different between long-term and short-term survivors. Expression of miR-30e and miR-210 showed a significant association with survival. MiR-30e: median OS of 24.2 months for low expression vs 13.3 months for high expression (p=0.03); miR-210: median OS of 24.2 months for low expression vs 13.7 months for high expression (p=0.008). Multivariate analysis with age, gender, histological subtype, NLR and microRNA expression included as variables revealed that miR-210 was the only factor remaining significant (p=0.006; hazard ratio: 0.41; 95% CI: 0.2–0.85). Conclusions: This study has identified expression of miR-210 as a potential new prognostic factor for patients undergoing EPP. Further validation is needed, but this marker has the potential to assist in better selection of MPM patients eligible for radical surgery. MC13-0032 Validation of 1 H-NMR-spectroscopy based metabolomics as a tool to detect lung cancer via a simple blood sample E. Louis 1 , L. Mesotten 2 , M. Thomeer 3 , K. Vanhove 4 , K. Vandeurzen 5 , K. Darquennes 6 , G. Reekmans 7 , P. Adriaensens 7 . 1 Faculty of Medicine and Life Sciences, Hasselt University, Diepenbeek, Belgium; 2 Nuclear Medicine, Ziekenhuis Oost-Limburg, Genk, Belgium; 3 Lung Diseases, Ziekenhuis Oost-Limburg, Genk, Belgium; 4 Faculty of Medicine and Life Scienes, Hasselt University, Diepenbeek, Belgium; 5 Lung Diseases, Mariaziekenhuis Noord-Limburg, Overpelt, Belgium; 6 Lung Diseases, Ziekenhuis Maas en Kempen, Maaseik, Belgium; 7 Institute of Material Research, Hasselt University, Diepenbeek, Belgium Background: Lung cancer is the leading cause of cancer death worldwide. Until today no effective methods permit the early detection of lung cancer. Therefore, detection methods with an improved specificity and sensitivity are urgently needed. Over the past decade, accumulating evidence has shown that the metabolism of cancer cells differs from that of normal cells. Metabolites are the end products of cellular metabolism and disturbances in biochemical pathways which occur during the development of cancer consequently provoke changes in the metabolic phenotype. Recently, our research group has built a statistical classifier (i.e. a metabolic phenotype) using multivariate orthogonal partial least squares-discriminant analysis (OPLS-DA). After removing the outliers (original dataset: 78 lung cancer patients and 78 control subjects), this classifier allows to discriminate between lung cancer patients and control subjects with a sensitivity of 83% (62 out of 75) and a specificity of 96% (71 out of 74). Purpose/Objective: The present study aims to validate these promising results in an independent study population of 80 patients with anatomopathologically confirmed lung cancer (before any treatment) and 80 control subjects. Materials and Methods: Fasting venous blood samples are collected and analyzed by 1 H-NMR spectroscopy. Subsequently, the constructed classifier is used to predict this independent group of lung cancer patients and control subjects. Results: By using the constructed classifier, 64 out of 80 (80%) lung cancer patients and 55 out of 80 (69%) control subjects are correctly classified. Conclusions: The constructed classifier allows to classify the majority of the lung cancer patients and control subjects correctly. Once we have collected sufficient samples to validate this method, we want to investigate at random whether it allows to detect lung cancer in a population with a low prevalence, actually whether it can be used as a valid screening tool. Poster Presentations MC13-0033 Prognostic value of CD109+ circulating endothelial cells in recurrent glioblastomas treated with bevacizumab and irinotecan M. Eoli 1 , A. Calleri 2 , M.G. Bruzzone 3 , E. Anghileri 4 , S. Pellegatta 4 , P. Mancuso 5 , A. Di Stefano 6 , F. Bertolini 5 , L. Cuppini 4 , G. Finocchiaro 4 . 1 Neuro-Oncology, Fondazione IRCCS Istituto Neurologico C. Besta, Milan, Italy; 2 Laboratory of Hematology-Oncology, European Institute of Oncology, Milan, Italy; 3 Neuro-Radiology, Fondazione IRCCS Istituto Neurologico C. Besta, Milan, Italy; 4 Neuro-Oncology, Fondazione IRCCS Istituto Neurologico C. Besta, Milan, Italy; 5 Laboratory of Hematology-Oncology, European Institute of Oncology, Milan, Italy; 6 Neurology, Ondazione IRCCS Istituto Neurologico Nazionale C. Mondino, Pavia, Italy Background: Data suggest that circulating endothelial and progenitor cells (CECs and CEPs,respectively) may have predictive potential in cancer patients treated with bevacizumab, the antibody recognizing vascular endothelial growth factor (VEGF). Purpose/Objective: Here we report on CECs and CEPs investigated in 68 patients affected by recurrent glioblastoma (rGBM) treated with bevacizumab and irinotecan and two Independent Datasets of rGBM patients respectively treated with bevacizumab alone (n=32, independent dataset A: IDA) and classical antiblastic chemotherapy (n=14, independent dataset B: IDB). Materials and Methods: rGBM patients with KPS >50 were treated until progression, as defined by MRI with RANO criteria. CECs expressing CD109, a marker of tumor endothelial cells, as well as other CEC and CEP subtypes, were investigated by six-color flow cytometry. Results: A baseline count of CD109+ CEC higher than 41.1/ml (1st quartile) was associated with increased progression free survival (PFS; 20 versus 9 weeks, P=0.008) and overall survival (OS; 32 versus 23 weeks, P=0.03). Longer PFS (25 versus 8 weeks, P=0.02) and OS (27 versus 17 weeks, P=0.03) were also confirmed in IDA with CD109+ CECs higher than 41.1/ml but not in IDB. Patients treated with bevacizumab with or without irinotecan that were free from MRI progression after two months of treatment had significant decrease of CD109+ CECs: median PFS was 19 weeks; median OS 29 weeks. The presence of two non-contiguous lesions (distant disease) at baseline was an independent predictor of shorter PFS and OS (P<0.001). Conclusions: Data encourage further studies on the predictive potential of CD109+ CECs in GBM patients treated with bevacizumab. MC13-0034 Prognostic factors of biochemical recurrence after radical prostatectomy for localized and locally-advanced prostate cancer V. Chernyaev 1 , V.B. Matveev 1 , M.I. Volkova 1 , Z.N. Nikiforova 2 , V.E. Shevchenko 2 . 1 Department of Urology, N.N. Blokhin Cancer Research Center, Moscow, Russia; 2 Oncoproteomics Laboratory Research Institute of Cancerogenesis, N.N. Blokhin Cancer Research Center, Moscow, Russia Background: to reveal prognostic factors of PSA-failure following radical prostatectomy in patients with localized and locally-advanced prostate cancer. Purpose/Objective: to reveal prognostic factors of PSA-failure following radical prostatectomy in patients with localized and locally-advanced prostate cancer. Materials and Methods: medical data of 386 consecutive patients with localized and locally-advanced prostate cancer who underwent radical prostatectomy from 1997 to 2011 were analyzed. Median age was 61.0 years. Median PSA before surgery – 10.3 ng/ml. Plasma levels of VEGF, VEGFR2, VEGFR3, TGF-β1, CD105, IL-6 were measured using Enzyme Linked-Immuno-Sorbent Assay (ELISA) before radical prostatectomy in 77 patients. Postoperatively the tumours were categorized as pT2 in 288 (59.1%), pT3 – in 144 (37.3%), pT4 – in 14 (3.6); pN+ – in 34 (8.8%) cases. Gleason score <7 was present in 254 (65.8%), ≥7 – in 132 (34.2%) specimens. Perineural invasion was identified in 188 (48.7%), angiolymphatic invasion – in 126 (32.6) cases. Results: Biochemical recurrence occurred in 64 (16.6%) out of 386 patients at a median follow-up of 30.5 (12–164) months. Independent predictors of biochemical recurrence were PSA (HR 0.161 (95% CI: 0.058–0.449); p=0.001), Gleason sum in surgical specimens (HR 0.496 (95% CI: 0.268–0.917); p=0.025), pN (HR 0.415 (95% CI: 0.181–0.955); p=0.039). The patients were divided into 3 prognostic groups: good (0 factor), intermediate (1 factor), poor (2 factors) and very poor (3 factors) (AUC – 0.720 (95% CI: 0.656–0.784)). High preoperative levels VEGF (≥67 pg/ml) (p=0.005), VEGFR2 (≥3149 pg/ml) (p=0.036), VEGFR3 (≥2268 pg/ml) (p=0.001), TGF-β1 (≥14473 pg/ml) S21 (p=0.052) were identified as unfavorable prognostic factors for survival without PSA-failure. Conclusions: Independent prognostic factors of biochemical recurrence after prostatectomy were PSA, Gleason sum and pN. Joint effect of the factors allows to predict PSA-relapse with accuracy 0.720. Preoperative serum levels VEGF, VEGFR2, VEGFR3, TGF-β1 potentially are perspective markers for PSA-failure after surgical treatment prostate cancer, further trials are needed. MC13-0035 Generation and (genetic) characterization of pre-clinical glioma models for “targeted therapies” B. Tops 1 , A. Navis 1 , A. Van Raaij 1 , K. Verrijp 1 , H. Petersen-Baltussen 2 , M. Ter Laan 2 , P. Wesseling 1 , W. Leenders 1 . 1 Pathology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 2 Neuro-surgery, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands Background: Currently, few effective therapeutic options are available for patients with gliomas, i.e. the most frequent primary tumors of the brain. Due to the diffuse growth pattern of the tumor cells into the normal brain tissue, gliomas are impossible to remove surgically. Treatment with radiotherapy and/or chemotherapy is also never curative. The average survival of patients with a glioblastoma is little more than 1 year. In recent years an increasing number of so-called “personalized cancer treatments” have been made available for several tumor types, “personalized” meaning that the treatment is tailored to the molecular changes in the patient’s tumor. Examples of successful treatments are the tyrosine kinase inhibitors crizotinib, targeting ALK in non-small cell lung carcinoma, and vemurafenib, targeting a mutated form of BRAF in melanoma. Purpose/Objective: We aim to identify novel and rational therapeutic strategies for individual glioma patients. The approach in this project is twofold; (1) genetic profiling of gliomas to identify possible therapeutic targets and (2) the generation of pre-clinical models (cell culture and intra-cerebral xenografts) from the same gliomas, which can then be used to validate therapies, rationally designed based on the outcome of 1). Especially intra-cerebral mouse models are invaluable, since the diffuse infiltrative growth of glioma cells into the normal brain parenchyma, often with intact blood brain barrier, are difficult to mimic using in vitro systems. These characteristics of glioma biogenesis are important as these may hamper therapeutic efficacy. Materials and Methods: To generate pre-clinical models, freshly resected tumor tissue is processed for in vitro cell culture and the remaining cell suspension is aliquoted and stored for the generation of intra-cerebral xenograft models at a later time point. Parallel to the creation of pre-clinical models, the tumor tissue is genetically profiled by massive parallel sequencing (IonPGM) using the comprehensive cancer panel from Life Technologies (containing 409 cancer-related genes). Results: We are currently in the process of generating and characterising the first glioma models. Conclusions: Here we will present the first (genetic) data regarding the tumors and models we created so far, and discuss potential therapeutic targets. MC13-0036 Dynamics of IGF-1R expression during endocrine breast cancer treatment S. Heskamp 1 , O.C. Boerman 1 , J.D.M. Molkenboer-Kuenen 1 , W.J.G. Oyen 1 , W.T.A. van der Graaf 2 , H.W.M. van Laarhoven 3 . 1 Department of Nuclear Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 2 Department of Medical Oncology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 3 Department of Medical Oncology, Amsterdam Medical Center, Amsterdam, The Netherlands Background: The insulin-like growth factor 1 receptor (IGF-1R) is a potential new target for breast cancer treatment. IGF-1R may be involved in the development of resistance against conventional systemic anti-cancer treatment. So far, the effect of neoadjuvant therapy on IGF-1R expression in breast cancer patients is unknown. Purpose/Objective: The aim of our study was to assess changes in IGF-1R expression in breast cancer after neoadjuvant treatment and to study whether these changes were associated with survival. Materials and Methods: Retrospectively, paraffin embedded tumor material was collected from pretreatment biopsies and surgical resections of 64 breast S22 cancer patients who were treated between 1989 and 2010 with neoadjuvant chemotherapy (CMF, anthracycline-based or taxane-based schedule) or endocrine therapy. ER, PR, HER2, and IGF-1R expression were determined immunohistochemically. IGF-1R expression was scored by two independent readers as negative (0), incomplete weak (1+), complete weak to moderate (2+), and strong (3+) membranous staining. IGF-1R expression before and after chemotherapy was compared. The correlation between changes in IGF-1R expression and survival was analyzed by Kaplan-Meier (log-rank test). Results: At time of diagnosis, 12 (19%), 18 (28%), 27 (42%), and 5 (11%) tumors were scored 0, 1+, 2+, and 3+ for IGF-1R expression, respectively. High IGF-1R expression (2+ or 3+) at diagnosis correlated significantly with ER positivity, low tumor stage (stage I/II) and longer overall survival (p<0.05). After neoadjuvant treatment, IGF-1R expression remained the same in 42 (66%) patients, was upregulated in 11 (17%) patients and downregulated in 11 (17%) patients. Upregulation in IGF-1R expression was significantly associated with high tumor stage (stage III/IV) at diagnosis. Most importantly, changes in IGF-1R expression were significantly associated with overall survival (p=0.007, Fig. 1). Mean overall survival time for upregulation, no change, and downregulation in IGF-1R expression was 2.7±0.4 years, 6.9±0.9 years and 14.7±2.0 years, respectively. Poster Presentations generated by a typical gradient-echo sequence. Each time-intensity curves were assigned to a sixteenth division of the phantom image, thus repeating this allocation for each time points, we obtained the dynamic series images stack (Fig. 1). Afterwards, variable experimental conditions that could affect the performance of the processing software are incorporated. Figure 1 Figure 1. Kaplan-Meier survival curve estimates of overall survival for patients with a tumor showing no change, a decrease or an increase in IGF-1R expression after neoadjuvant breast cancer treatment. Conclusions: Neoadjuvant treatment can induce changes in IGF-1R expression of breast tumors. Upregulation of IGF-1R expression was associated with a short overall survival, while a downregulation was associated with a long overall survival. Patients whose tumors showed increased IGF-1R expression may potentially benefit from IGF-1R targeted therapy. MC13-0037 A synthetic phantom for quality assessment of DCE-MRI quantification S.L. Chao. MRI, Institut Jules Bordet, Brussels, Belgium Background: In the recent last years, the development of novel antiangiogenic and antivascular cancer therapies has led to important considerations regarding the potentiality for DCE-MRI measurement to be used as an imaging biomarker of drug efficacy in clinical trials of angiogenesis inhibitors. Purpose/Objective: Quantitative results are not comparable across sites for lack of standardization in data acquisition protocols and analysis tools. The purpose of this work is to synthetize a DCE-MRI test object from known values of kinetic parameters so that it could be analyzed by multiple software tools to compare their performance, and may provide prerequisites for standardization procedure and sites/systems certification in multi-center trials. Materials and Methods: We reused the simulated data obtained by Buckley (MRM Vol. 47 March 2002) with 1 s resolution time, consisting in a vascular input function and 13 tissues response curves. These data were generated by MMID4 (http://nsr.bioeng.washington.edu) using different combination of parameters Ktrans , νe and vp . We extended this parameter space by 2 additional curves. The concentration data are first converted to MR signal intensities as The final phantom can include images for T1 mapping and provides customizable extend towards the image noise, the tissue native T1 , the overall onset time, the time delay between arterial input and tissue vasculature, and the measurement protocol (sampling time, repetition time, echo time, flip angle). The phantom was tested by using in-house analysis software regarding the robustness of its algorithms in fitting the commonly accepted general pharmacokinetic model. Parameter estimates obtained from 2 usual optimization approaches, in time and frequency domains, were compared with the value used in simulations. Results: Regarding the robustness, tests on noisy data showed no error during the software execution. Ktrans was systematically overestimated, this was concomitant with an under-estimation of vp : the % deviations from simulation values were much more prominent in time domain (2–76, mean 26%) than in frequency domain (0–13, mean 3%). Estimations for νe were close to the simulation values along the parameters space. Conclusions: The single test performed here has demonstrated that the optimization approach affects the comparison of results, yet this indication is not always referred in pubished data. For developers, the phantom provide ways to define acceptance criteria for software. Irrespective of the actual QA/QC requirements for imaging, running on common test data should be considered for any multi-centric study which requires a quantification process. MC13-0038 Detection of circulating tumor cells-related biomarkers for predicting early relapse of colorectal cancer by weighted enzymatic chip array method S. Lin 1 , H. Liu 2 . 1 Division of Medical Research, Fooyin University Hospital, Ping Tung County, Taiwan; 2 Division of Laboratory Medicine, Fooyin University Hospital, Ping Tung County, Taiwan Background: Colorectal cancer (CRC) is a significant public health problem. In Taiwan, it is estimated that over 10,000 new cases of CRC were diagnosed in 2006, and over 4,100 patients died of CRC that year. It is important to note that as many as 25–40% of patients who undergo curative resection subsequently develop metastatic disease, suggesting that undetected micrometastases exist, and may play a key role in relapse. Purpose/Objective: The aim of this study was to detect the circulating tumor cells (CTCs)related multiple biomarkers for predicting early relapse of CRC patients by a weighted enzymatic chip array (WEnCA) method. Materials and Methods: We selected 15 CTC associated candidate genes including five genes demonstrated to correlated with cancer prognosis. The expression level of 15 candidate genes of all 105 postoperative CRC patients were detected by WEnCA platform and the correlation between the experimental data and patients’ clinical pathological features was statistically analysis using spss software. Poster Presentations Results: Postoperative relapse was significantly correlated with overexpression of four genes including EVI2B (P=0.001, OR=4.622), ATP2A2 (P=0.006, OR=4.688), S100B (P=0.001, OR=11.521), TM4SF3 (P=0.001, OR=6.756), and OLFM4 (P=0.008, OR=3.545). Sensitivity, specificity, and accuracy of WENCA operation platform were 94.7%, 93.5%, and 97%, respectively. Conclusions: Detection of CTC-related multiple biomarkers by WEnCA operation platform can significantly improve the early prediction rate of postoperative CRC relapse. MC13-0039 Id1/IGF2/IGF1R/PI3K/AKT signaling cascade as functional markers and therapeutic targets in esophageal cancer B. Li 1 , G. Tsao 1 , K. Chan 2 , A. Cheung 1 . 1 Department of Anatomy, 2 Department of Pathology, The University of Hong Kong, Hong Kong, China Background: Mounting evidence suggests that growth factors secreted by cancer cells play important roles in cancer progression. Id1 is commonly overexpressed in solid tumors including esophageal squamous cell carcinoma (ESCC). Purpose/Objective: We aim to determine if the functions of Id1 in promoting cancer progression and chemoresistance are mediated by growth factors, investigate the autocrine/endocrine effects of the key Id1-induced growth factor, and evaluate the treatment efficiency of relevant anticancer therapeutic strategies. Materials and Methods: Antibody array-based screening was used to identify differentially secreted growth factors from Id1-overexpressing ESCC cells. In vitro and in vivo assays were performed to confirm the induction of IGF2 by Id1, and to study the autocrine and endocrine effects of IGF2 in promoting ESCC progression. Human ESCC tissue microarray was analyzed for overexpression of IGF2 and its correlation with that of Id1 and p-AKT. Fluorouracil (5-FU)-resistant ESCC sublines were established by treating cells with increasing doses of 5-FU, and gene expression profiles were compared with the parental cells by cDNA microarray. The anti-tumor and anti-metastatic efficacies of IGF2 antibody and PI3K inhibitor wortmannin were evaluated using tumor xenograft and experimental metastasis models. Results: Id1 overexpression induced IGF2 secretion which promoted cancer cell proliferation, survival and invasion by activating AKT in an autocrine manner. IGF2 secreted by Id1-overexpressing ESCC xenograft could instigate the growth of distant esophageal tumors, as well as promote metastasis of circulating cancer cells in an endocrine manner. Tissue microarray data showed overexpression of IGF2 in 21/35 (60%) human ESCC tissues which was associated with up-regulation of Id1 and p-AKT. Moreover, Id1 was identified by cDNA microarray analysis as being up-regulated in 5-FUresistant ESCC sublines; gain- and loss-of-function experiments confirmed its importance in increasing chemoresistance. We found that IGF2 antibody and wortmannin treatment significantly suppressed tumor growth, metastasis, and chemoresistance in mouse models. Conclusions: Our study demonstrated that the Id1/IGF2/PI3K/AKT signaling cascade plays an important role in esophageal cancer progression and chemoresistance. Blockade of IGF2/PI3K/AKT signaling has therapeutic potential in the management of esophageal cancer. This study is supported by General Research Fund (HKU 762610M) and HKU Seed Funding Programme for Basic Research (201111159198). MC13-0040 A novel HER3-V855A driver mutation homologous to EGFR-L858R in lung cancer I. Umelo 1 , A. Noeparast 1 , G. Chen 1 , M. Renard 2 , C. Geers 3 , J. Vansteenkiste 4 , E. Teugels 1 , J. De Greve 1 . 1 Medical and Molecular Oncology, Vrije Universiteit Brussel, Brussels, Belgium; 2 Pediatric Hemato-Oncology, UZ Leuven, Leuven, Belgium; 3 Pathology, UZ Brussel, Brussels, Belgium; 4 Pneumology, UZ Leuven, Leuven, Belgium Background: Somatic mutations found within the tyrosine kinase (TK) domain of the human epidermal growth factor (HER or ErbB) family of transmembrane receptors have been implicated in the development and progression of non-small cell lung cancer. These mutations have been identified in the EGFR (epidermal growth factor receptor [HER1; ErbB1), HER2 (ErbB2) and HER4 (ErbB4) genes, but no NSCLC associated functional mutations have been described to date in the kinase inactive HER3 (ErbB3) gene. Here, we report the case of an adolescent patient with advanced NSCLC where DNA sequence analysis of all 4 HER family genes in his tumor biopsy S23 specimen identified a novel V855A somatic mutation located in exon 21 of the HER3 TK domain. Remarkably, the mutation maps at a position homologous to the frequently described EGFR tyrosine kinase inhibitor (TKI)-sensitive L858R activating mutation located in exon 21 of the EGFR TK domain. Purpose/Objective: The aim of this study was to characterize the biological effects of the novel HER3-V855A mutation in an in vitro mammalian cellular system and to further investigate its potential as a possible therapeutic target in NSCLC. Materials and Methods: To examine functional differences we characterized HER3-V855A alone or combined with its HER2, dimerization partner in a null Ba/F3 model. Stable transfectants were subjected to MTS analysis in order to assess cellular proliferation with HER-specific ligands as well as the growth inhibitory effects of HER-specific inhibitors. Similarly, western blot analysis evaluated the effect of growth factor stimulation and drug inhibition on the HER signalling pathway. In addition, Annexin V/7-AAD analysis assessed the effect of HER-specific inhibitors on apoptosis. Results: In vitro functional analysis in a null Ba/F3 background reveals that HER3-V855A when combined with its HER2 dimerization partner leads to enhanced neuregulin 1β-induced receptor activation and transforms interleukin-3 (IL-3) dependent Ba/F3 cells to neuregulin 1β-dependent growth. Afatinib, a small molecule ErbB family inhibitor, has anti-proliferative and pro-apoptotic effects on the HER3-V855A: wild-type HER2 Ba/F3 derivative that is logarithmically higher than the effect obtained in the wild-type HER3: wild-type HER2 combination. Conclusions: Together, our findings demonstrate that kinase impaired HER3 can be activated and that the HER3-V855A mutation confers a gain-offunction phenotype that is associated with sensitivity to afatinib. This finding could be relevant for the treatment of NSCLC or other cancers carrying such mutations. MC13-0041 Is the prognostic role of serum CXCR4 in metastatic or recurrent colorectal cancer? S. Kim, Y. Choi, Y. Kim. Medicine, Korea University, Seoul, Korea Background: The CXCR4 is involved in several aspects of tumor progression including angiogenesis, metastasis, and survival. However, whether serum CXCR4 level in metastatic or recurrent colorectal cancer (CRC) has a prognostic role has been not evaluated. Purpose/Objective: The purpose of this study is to investigate the role of serum CXCR4 level for the prognosis of metastatic or recurrent CRC Materials and Methods: We analyzed serum samples from 55 patients with advanced CRC diagnosed between March 2008 and July 2011. Blood was collected before beginning systemic chemotherapy and serum CXCR4 levels were quantified by commercially available ELISA kit. Results: The median age of the 55 patients was 62 years (range: 39–82) and all patients received systemic chemotherapy of 2 or more line. The median serum CXCR4 level was 0.8425 pg/ml. Patients with 2 or more of metastatic sites, liver metastasis, or over more normal level of CA 19-9 (37 <) showed significantly higher level of serum CXCR4 than patients without. The median overall survival of all patients was 19.53 months. There was significant difference for OS between patients with CXCR4 ≤ 0.766 and 0.766 < (p=0.046). Univariate analysis showed that liver metastasis, no debulking operation and higher level of CXCR4 (0.766 <) had significantly poor prognostic value regarding OS (p<0.05). Conclusions: Serum CXCR4 level was positively correlated with disease burden (2 or more of metastatic sites, liver metastasis, or over more normal level of CA 19-9). And there was significant difference for OS according to the level of CXCR4. These findings suggested that CXCR4 might be useful as surrogate marker of clinical outcome in metastatic or recurrent CRC. MC13-0044 Impact of pre-analytical sample handling on metabolomic studies R. Vettukattil, S. Lamichhane, T.H. Haukaas, S.A. Moestue, T.F. Bathen. Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway Background: Metabolomics offers a powerful tool to study the variations in metabolites in various clinical conditions. Magnetic resonance spectroscopy (MRS) is an analytical tool used in metabolomic studies, which allows rapid and non-targeted exploration of metabolites in biological samples. As metabolic profiling is gaining popularity in clinical research, it is important to S24 Poster Presentations Abstract MC13-0044 – Figure 1 assess the susceptibility of these techniques to systematic variations resulting from sample handling and analysis. Purpose/Objective: In this study we systematically investigated the effect of post-surgical delay prior to sample freezing on metabolomics studies of patient derived breast cancer xenograft tissue samples using high resolution magic angle spinning (HR MAS) MRS. Materials and Methods: Tissue samples were obtained from mouse xenograft models representing luminal-like (n=3) and basal-like (n=3) molecular subtypes. After dissection, tumor tissue was snap frozen in liquid nitrogen at 0, 15, 30, 60, 90, and 120 minutes after surgery. In total 8 samples were collected from each tumor, which includes a replicate at 15 minutes and a sample which was analyzed immediately after surgery, without any freezing. All samples were analyzed by HR MAS MRS on a Bruker Avance III 600 MHz/54 mm US at 4°C. Multivariate analysis of the spectral data was performed in the form of unsupervised principal component (PC) analysis using PLS-Toolbox 5.8 for MATLAB© . All procedures and experiments involving animals were approved by the European Animal Research Authority. Results: Multivariate analysis of the samples showed that individual metabolite variation had a higher impact on the metabolic profile than the post-surgical delay. The overall metabolic profile was fairly robust to variation in sample processing (Fig. 1). Basal and luminal xenografts were separated along first PC (accounting for 80.5% of overall variability) despite the wide variation in sample freezing times. Conclusions: MR spectroscopic metabolite profiles of tumor tissue samples are more reproducible and robust to variation in post-surgical delay prior to freezing than variation introduced by their biological differences. However, the degree of variation in metabolites differs based on the type and molecular heterogeneity of tumors. The need for stability analysis in every clinical metabolomic study is warranted. MC13-0045 Genes from the TP53 network and their role in ovarian cancer prognosis and response to chemotherapy A. Podgorska, L. Szafron, A. Felisiak-Golabek, J. Kupryjanczyk. Department of Pathology, Maria Sklodowska-Curie Memorial Cancer Centre and Institute of Oncology, Warsaw, Poland Background: The TP53 gene is one of the most frequently mutated genes in human cancers and is thought to play a crucial role in malignant transformation. The key role of TP53 as a tumour suppressor is to block cell cycle progression and/or to induce apoptosis in response to cellular stresses. Within the TP53 pathway there are different up/downstream effectors regulating its stability and activity. Purpose/Objective: The aim of this study was to evaluate clinical significance of the expression of 5 different TP53-related genes (MDM2, MDM4, HUWE1, GADD45A and CDKN2A) in advanced ovarian cancers, with respect to the TP53 status. Materials and Methods: Ovarian carcinomas from 106 patients treated with taxane-platinum (TP, n=74) or platinum-cyclophosphamide (PC/PAC, n=32) were chosen for real-time PCR analysis. mRNA expression levels of 5 genes were screened using TaqMan® Gene Expression Assays (Life Technologies). Data were analysed using the relative quantitation method with HGPRT as a reference gene. Univariate and multivariate statistical analyses (Cox proportional hazards model and logistic regression model) were performed to evaluate overall survival, disease-free survival and response to chemotherapy. Analyses were made in the entire group, as well as in subgroups of tumours with (TP53+) and without TP53 accumulation (TP53−). Results: Within the TP-treated group of ovarian cancer patients elevated expression of MDM2, MDM4, HUWE1 and GADD45A mRNA was associated with enhanced risk of death in the TP53− subgroup (HR 1.30, p=0.002; HR 1.83, p=0.005; HR 6.27, p=0.001; HR 5.44, p=0.028; respectively). High GADD45A mRNA expression increased also the risk of recurrence in the TP53+ subgroup (HR 3.49, p=0.048). Elevated MDM4 mRNA levels decreased the probability of complete remission in the TP53− subgroup (OR 0.33, p=0.046). High expression of HUWE1 might have also an influence on platinum sensitivity in the TP53− subgroup (OR 0.024, p=0.052). CDKN2A mRNA expression did not show any statistically significant effect on clinical endpoints. In the PC/PAC treated group none of the analysed gene expression levels had an influence on clinical outcome. Conclusions: In the present study we have demonstrated the clinical importance of the expression of 4 TP53-related genes. Our results suggest that high mRNA levels of MDM2, MDM4, HUWE1 and GADD45A are negative predictive and/or prognostic factors in ovarian cancers treated with taxane-platinum agents. Poster Presentations S25 MC13-0046 Volatile organic compounds as an early diagnostic tool for malignant pleural mesothelioma K. Lamote 1 , J. Van Cleemput 2 , K. Nackaerts 3 , J.P. van Meerbeeck 4 . 1 Respiratory Medicine, Ghent University Hospital, Ghent, Belgium; 2 Occupational Health Service, Eternit N.V., Kapelle-op-den-Bos, Belgium; 3 Respiratory Medicine, University Hospital Gasthuisberg, Leuven, Belgium; 4 Thoracic Oncology/MOCA, Antwerp University Hospital, Antwerp, Belgium Background: Early diagnosis of malignant pleural mesothelioma (MPM) can improve patients’ outcome but is hampered by non-specific symptoms and investigations, which delay diagnosis and result in advanced stage disease [van Meerbeeck JP, 2011]. An accurate non-invasive test allowing early stage diagnosis in asbestos-exposed persons is lacking. Purpose/Objective: Breathomics aims at a non-invasive analysis of volatile organic compounds (VOCs) reflecting the cells’ metabolism. The breathogram obtained by the electronic nose does however, not allow identification of MPM-related VOCs [Chapman EA 2009, Dragonieri S 2011]. Ion mobility spectrometry (IMS) combines the advantages of online direct sampling with the possibility of VOC identification and linking to MPM pathogenesis [Baumbach JI 2009]. We investigated which VOCs could play a role in MPM pathogenesis in order to build a possible diagnostic MPM tool. Materials and Methods: 10 MPM patients and 10 healthy professionally asbestos-exposed individuals were included after refraining from eating, drinking and smoking for at least 2 hours before sampling. They breathed tidally for 3 minutes through a mouthpiece connected to a bacteria filter. Ten ml alveolar air was sampled via a CO2 -controlled ultrasonic sensor and analyzed using the BioScout Multicapillary Column/Ion Mobility Spectrometer (MCC/IMS, B&S Analytik, Dortmund, Germany) [Westhoff M 2009], by using N2 as a carrier gas. Per subject a background sample was taken. Peaks of interest were visually selected and their intensity (V) was analyzed and compared between background and breath samples via on-board VisualNow 3.2 software and SPSS v21 (IBM) using Mann-Whitney-U tests. Results: Out of 41 peaks of interest, three show a significantly higher intensity in the exhaled breath of MPM patients than healthy controls [Table]. The high AUCROC of resp. P12 (0.877) and P24 (0.863) suggests a possible role of these associated VOCs in MPM pathogenesis and as a diagnostic marker in discriminating MPM patients from asbestos-exposed healthy controls. Peak MPM Intensity (V)* AEx Intensity (V)* p-value AUCROC P6 P9 P11 P12 P19 P20 P24 P26 P27 P33 P36 0.018 [0.013–0.027] 0.106 [0.085–0.124] 0.089 [0.068–0.120] 0.078 [0.054–0.168] 0.044 [0.038–0.048] 0.044 [0.029–0.059] 0.102 [0.085–0.123] 0.010 [0.007–0.016] 0.012 [0.008–0.021] 0.006 [0.004–0.007] 0.012 [0.009–0.016] 0.026 [0.019–0.031] 0.116 [0.104–0.145] 0.082 [0.064–0.093] 0.027 [0.022–0.060] 0.045 [0.042–0.055] 0.072 [0.043–0.078] 0.079 [0.069–0.089] 0.011 [0.006–0.013] 0.013 [0.010–0.052] 0.010 [0.007–0.019] 0.012 [0.009–0.016] 0.41 0.17 0.45 <0.01 0.76 0.17 0.03 0.76 0.60 0.03 0.82 0.490 0.763 0.750 0.877 0.777 0.620 0.863 0.567 0.618 0.237 0.517 *Median [IQR]. 1/K0 : inversed reduced ion mobility. AEx: healthy asbestos exposed individual. AUCROC : Area under the receiver operator characteristic curve (accuracy in diagnosing MPM). MPM: Malignant Pleural Mesothelioma patient. RT: retention time. Conclusions: Several VOCs of interest were obtained in the breath of MPM patients. Two peaks were significantly discriminating between both populations. GC-MS analysis and further large cohort studies are ongoing in order to validate the accuracy of IMS as a diagnostic tool for MPM. MC13-0047 Serum and tissue biomarkers for cancer biospecimen integrity L. Agrawal, H. Moore, J. Vaught. BBRB Cancer Diagnosis Program, National Cancer Institute, Rockville, USA Background: High quality cancer biospecimens with appropriate clinical annotation are critical in the era of biomarker discovery in personalized medicine. Numerous pre-analytical factors affect human biospecimen integrity for biomarker research in cancer. This situation is applicable to a variety of biospecimens including plasma/serum and fixed cancer tissues used for biomarker analysis. Purpose/Objective: The purpose of the current work performed via Biospecimen Research Network (BRN) is to enable research to provide the foundation for evidence-based practices and to develop serum and tissue biomarkers for human biospecimen integrity. BRN is a program eastablished by the U.S. National Cancer Institute (NCI) Biorepositories and Biospecimen Research Branch (BBRB) to coordinate NCI’s biospecimen resource activities and address those issues that affect access to the high quality specimens for biomarker research. Materials and Methods: We summarize here the development of assays and identification of biomarkers that may be used as sentinel markers of plasma/tissue stability in biobanks using mass-spectroscopy proteomics, analysis of circulating miRNA, and immunostaining of FFPE tissues (AQUA technology). Results: In one of the projects, identification of protein biomarkers using mass-spectrometry and Illumina arrays in serum obtained from breast cancer and matched normal subjects has led to development of potential biomarker panel for serum/plasma integrity and these biomarkers are currently being validated. A second project studied effects of pre-analytical variables on circulating miRNA and identified and validated new and improved housekeeping miRNA and biomarkers associated with breast cancer. In another study, a series of biomarkers have been validated by construction of tissue microarray (TMA) from 93 breast cancer specimens with known time to fixation as a pre-analytical variable. A tissue quality index (TQI) model was generated to predict the time to fixation and tissue quality by studying a subset of biomarker proteins in breast cancer tissues using AQUA scores. Conclusions: This presentation will outline the progressive efforts taken by BRN investigator-led projects to identify and validate biomarkers for cancer biospecimen integrity and help establish standards for serum and cancer tissue quality for clinical trials and biomarker assay development. MC13-0048 Disruption of CD9 affects adhesion, migration and actin polymerization in ETV6/RUNX1 pre-B lymphocytes M.P. Arnaud 1 , A. Vallee 1 , G. Robert 1 , E. Dondi 2 , B. Jacinthe 3 , C. Leroy 2 , N. Varin-Blank 2 , M.D. Galibert 1 , M.B. Troadec 1 , V. Gandemer 1 . 1 CNRS UMR6290, Institut de Génétique et Développement de Rennes, Rennes, France; 2 Université Paris 13, UMR978 INSERM, Rennes, France; 3 Unité d’Hémato-pédiatrie, Centre Hospitalier Universitaire, Rennes, France Background: CD9, a membrane protein member of the tetraspanin family, is implicated in hematopoietic stem cell engraftment. Moreover CD9 expression has been correlated with the risk of metastasis and is associated with a poor clinical outcome in various types of cancer. Notably, CD9 is under expressed in the ETV6/RUNX1 pre-B acute lymphoblastic leukaemia (ALL). Purpose/Objective: The purpose of our study is to investigate the effect of CD9 expression on the motility process in vitro and to identify the regulation pathways related to CD9. Materials and Methods: The CD9-positive REH cell line (from an ETV6/RUNX1 ALL relapse) was used. By lentiviral transduction of shRNA targeting CD9 mRNA, we generated REH cells depleted in CD9. We also used an anti-CD9 antibody. Ability of cells to adhere on fibronectin and to migrate in response to CXCL12 were measured in vitro. The activation of PI3K/AKT pathway according to CD9 expression was studied by western blotting. We also investigated the presence of membrane villi on cell surface by scanning electron microscopy. Finally F-actin polymerization after CXCL12 stimulation was measured by rhodamin-phalloidin labelling and localization of CD9 and CXCR4 was observed by immunofluorescence. Results: We demonstrated a higher adhesion on fibronectin in CD9 overexpressing cells and with anti-CD9 antibody. Western Blot analysis revealed that AKT activation positively correlate with cell’s adhesion rate. Preliminary results with AKT inhibitors reinforced PI3K/AKT pathway role in adhesion of REH. As well, the more CD9 is expressed, the more the migration rate in response to CXCL12 is high. Migration rate is also increased with anti-CD9 antibody. By immunofluorescence we showed first that CD9 and CXCR4 are colocalized on cell membrane and secondly co-capped with actin in the presence of anti-CD9 antibody. Without any stimulation, CD9-positive REH had longer villi than ShCD9 transducted cell lines. After CXCL12 stimulation F-actin polymerization is increased in CD9-positive cells. Conclusions: Our data show that CD9 is a key actor in pre-B lymphoblasts adhesion and CXCL12 dependant migration. CD9 expression interacts with actin remodelling. We are now investigating a potential link between CD9 and RAC1 known to be activated in response to CXCL12.Therefore, the expression level of CD9 could impact leukemic blasts abilities to spread and be a major actor of relapses. S26 MC13-0050 Proteomic markers in early buccal mucosa squamous cell cancers S. Nair 1 , S. Malgundkar 2 , A. Patil 3 , S. Kane 3 , K. Sadhana 4 , A. D’Cruz 1 , S. Zingde 2 . 1 Head and Neck Service, Tata Memorial Center, Mumbai, India; 2 Zingde lab, ACTREC TMC, Mumbai, India; 3 Pathology, Tata Memorial Center, Mumbai, India; 4 Bio-Statistics, ACTREC TMC, Mumbai, India Background: Carcinoma of buccal mucosa is common in South Asia due to large-scale usage of chewable tobacco. However, clinically these cancers behave similar to tongue cancers, which are common in western countries. Like any other head and neck squamous cell cancers (HNSCC), their prognosis is affected by established clinical factors like nodal metastasis and T stage. Presence of regional lymph node metastasis is one of the important factors for poor clinical outcome and nodal status of the neck plays a decisive role in the choice of treatment. An assessment of the cervical lymph node metastatic status in oral cavity cancer not only helps to predict the prognosis of patients, but also helps surgeons to choose the appropriate treatment. About 70% of patients in T1 (<2 cm) and 30% in T4 primary tumour stage may not have lymphatic metastasis at the time of treatment. Similarly, even after using all the available technologies to detect lymph nodes in the neck, the rate of occult metastasis is significantly high in OSCC. Many literature reports points to several molecular markers expressed by the primary tumour in predicting its metastasising nature. It is important that these markers should be such that their use can be effectively translated into easy to implement techniques in any molecular pathology laboratory. While gene expression profiles are increasingly used for prognostication, their relative costs and time requirements make them difficult for routine use in clinical service. Proteomic markers can be detected using IHC and will be easier to use clinically. Purpose/Objective: This study aims in developing a set of markers in early buccal mucosa cancers using previously established set of proteomic markers differentially expressed in buccal mucosa cancers. Materials and Methods: We have analyzed formalin fixed paraffin embedded blocks from 90 patients with early stage (T1/T2) buccal mucosa cancers treated surgically at Tata Memorial Hospital, Mumbai. All patients were clinically node negative at the time of surgery. Tissue microarrays was prepared and the sectons stained with antibodies for 19 markers. The stained slides were graded and quantified based on the localization, extent of staining and intensity of staining. Results: Median expressions of P53 and EGFR between node negative and positive tumors were not significant. However, we observed significant differences in the expression of the following markers among node negative versus node positive tumor samples. 1. Higher expression of SFN is associated with lower risk of nodal metastasis (p 0.03), 2. Higher expression of TCTP is associated with lower risk of nodal metastasis (p 0.003). These markers along with 14-3-3-zeta also showed significant differences in expression between well differentiated tumors and others. Conclusions: Proteomic markers have potential to predict nodal metastasis in early stage buccal mucosa cancers. MC13-0051 Molecular characterization of anaplastic astrocytoma and glioblastoma multiforme: PTEN loss in conjunction with EGFR alterations predicts response to therapy M. Donovan 1 , A. Colomer 2 , P. Puig 2 , N. Erill 2 , I. Alarcon 2 , N. Vidal 3 , I. Ferrer 3 . 1 Pathology, Mt. Sinai Medical Center, New York, USA; 2 Pathology, Althia Health S.L., 3 Pathology, Bellvitge Hospital, Baracelona, Spain Background: Anaplastic Astrocytoma (AA) and Glioblastoma multiforme (GBM) represent the most common and malignant of adult brain tumors. Despite best standard of care including surgical resection, radiation and chemotherapy (e.g. temozolomide), the median survival remains approximately 12–15 months, with fewer than 25% of patients surviving up to 2 years and fewer than 10% of patients surviving up to 5 years. Previously we had observed that EGFR amplification/polysomy in conjunction with PTEN LOH by FISH was associated with improved outcome in patients with GBM. Purpose/Objective: We sought to confirm and validate these initial findings by expanding the patient cohort size, examining the protein levels of the individual genes and to determine if AA patients have a similar predicted phenotype. Materials and Methods: Tissue microarrays (TMAs) with up to 4 cores/tumor specimen from 174 patients (124 GBM and 50 AA) with a median follow up of Poster Presentations 8 years were evaluated using dual-color DNA FISH for both PTEN and EGFR amplification (AMP) and or high polysomy (POLY). In parallel, TMA arrays were interrogated with a previously described multispectral immunofluorescent (IF) assay (Cordon-Cardo et al., 2007) which included: GFAP, EGFR, PTEN and Ki67. Feature performance was univariately assessed using the concordance index and Kaplan-Meier survival curves, individually controlled for age and type of surgery (i.e. biopsy vs resection). Results: The median survival for AA patients was 377 days vs. 175 days for the GBM patients (p<0.001), median age of AA was 52 years and GBM, 62 years (P≤0.001). By IF, tumor specific KI67 was not predictive whereas high levels of EGFR (p=0.014) and low PTEN (p=0.008) were associated with poor outcome in the AA only group. In contrast, higher levels of PTEN in the GBM cohort were associated with poor outcome (p=0.022). EGFR AMP but not EGFR POLY was consistently associated with poor survival and this finding was most significant when PTEN LOH was <40% for either AA (p=0.001) or GBM (p=0.002), respectively. By comparison, the combination of PTEN LOH >40% and EGFR polysomy in GBM when controlled for surgical type, was associated with improved survival (p-value 0.002), while those rare GBM patients with both EGFR AMP and PTEN LOH >40% also had improved survival (p=0.012). The data supports a role for accurately assessing both PTEN loss and EGFR status when predicting response to treatment, especially for patients with EGFR POLY. Conclusions: The incorporation of EGFR/PTEN DNA FISH and quantitative IF is useful for developing a biologic baseline for understanding disease course and response to treatment in malignant gliomas. PTEN LOH and EGFR polysomy/amplification in GBM suggests a mechanism for favorable response and should help guide subsequent multi-modal treatment regimens. MC13-0052 Accuracy evaluation of a novel companion diagnostic method for BRAF V600E/V600K identification in GSK clinical trials L. O’Donnell 1 , W. Sweet 1 , X. Lu 1 , F. Meynier 2 , A. Derome 3 , L. Ganee 4 , F. Poyet-Gelas 3 , A. Martin 5 , M. Casey 6 , N. Kertesz 7 . 1 Clinical Affairs, BioMerieux Inc., Durham, USA; 2 BioMath, BioMerieux Inc., Grenoble, France; 3 V & V, BioMerieux Inc., Grenoble, France; 4 Product Design & Development, BioMerieux Inc., Grenoble, France; 5 Oncology, GSK, Collegeville, USA; 6 Statistics, GSK, Collegeville, USA; 7 Pharmaceutical Liaison Manager, Response Genetics Inc., Los Angeles, USA Background: The THxID™ BRAF kit is an in vitro diagnostic device intended for the qualitative detection of the BRAF V600E and V600K mutations in DNA samples extracted from formalin-fixed paraffin embedded (FFPE) human melanoma tissue. The THxID™ BRAF kit is a real-time PCR test on the ABI 7500 Fast Dx system and is intended to be used as an aid in selecting melanoma patients whose tumors carry the BRAF V600E mutation for treatment with dabrafenib [Tafinlar® ] and as an aid in selecting melanoma patients whose tumors carry the BRAF V600E or V600K mutation for treatment with trametinib [Mekinist™]. Purpose/Objective: The purpose of this study is to evaluate the accuracy of the THxID™ BRAF assay using retrospective samples from three GSK clinical trials Materials and Methods: A total of 882 specimens from subjects in GSK clinical trials BRF113710, BRF113929 and MEK114267 were retrospectively tested with the THxID™ BRAF assay. Specimens had been screened previously using an investigational use only (IUO) assay from Response Genetics Inc. (RGI). After THxID™ testing, the remaining DNA eluates were then sent to RGI for bidirectional Sanger sequencing. The accuracy of the THxID™ BRAF assay was presented as the concordance with bidirectional Sanger sequencing. Results: For all the specimens combined as a group for samples with valid results, the overall Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA) were 98.1% (403/411) (95% CI [96.2%; 99.0%]) and 93.9% (417/444) (95% CI, [91.3%; 95.8%]), respectively. The PPA for V600E and V600K were 98.0% (341/348) (95% CI, [95.9%; 99.0%]) and 93.7% (59/63) (95% CI, [84.8%; 97.5%]), respectively. No specimen was detected for both V600E and V600K. The overall invalid rate for the THxID™ BRAF assay was 3.1% (27/882) with the following identification by bidirectional Sanger sequencing: 6 samples as V600E (0.7%), 1 samples as V600K (0.1%), and 20 samples as WT (2.3%). Conclusions: The PPA and NPA results, as compared to the bidirectional Sanger sequencing method, indicated a high accuracy of the THxID™ BRAF assay for the identification of BRAF V600E/V600K. The THxID™ BRAF assay Poster Presentations has been demonstrated as an accurate and reliable method to aid in selection of a population with BRAF V600E/V600K mutation for targeted therapy. MC13-0053 High analytical concordance between a novel companion diagnostic assay and an IUO assay for BRAF V600E/V600K detection in melanoma L. O’Donnell 1 , W. Sweet 1 , X. Lu 1 , F. Meynier 2 , A. Derome 3 , L. Ganee 4 , F. Poyet-Gelas 3 , A. Martin 5 , M. Casey 6 , N. Kertesz 7 . 1 Clinical Affairs, BioMerieux Inc., Durham, USA; 2 BioMath, BioMerieux Inc., Grenoble, France; 3 V & V, BioMerieux Inc., Grenoble, France; 4 Product Design & Development, BioMerieux Inc., Grenoble, France; 5 Oncology, GSK, Collegeville, USA; 6 Statistics, GSK, Collegeville, USA; 7 Pharmaceutical Liaison Manager, Response Genetics Inc., Los Angeles, USA Background: The THxID™ BRAF kit is an in vitro diagnostic device intended for the qualitative detection of the BRAF V600E and V600K mutations in DNA samples extracted from formalin-fixed paraffin embedded (FFPE) human melanoma tissue. The THxID™ BRAF kit is a real-time PCR test on the ABI 7500 Fast Dx system intended to be used as an aid in selecting melanoma patients whose tumors carry the BRAF V600E mutation for treatment with dabrafenib [Tafinlar® ] and as an aid in selecting melanoma patients whose tumors carry the BRAF V600E or V600K mutation for treatment with trametinib [Mekinist™]. Purpose/Objective: The objective of the study is to assess the analytical concordance between THxID BRAF assay and an investigational use only (IUO) assay from Response Genetics, Inc. (RGI). Materials and Methods: A total of 926 samples from GSK clinical trials BRF113710, BRF113929 and BRF113683 with results from the RGI IUO were tested with THxID™ BRAF. Analytical concordance between THxID™ BRAF and RGI IUO was assessed using the test results and reported as Positive Percent Agreements (PPA), Negative Percent Agreements (NPA), and Overall Percentage Agreements (OPA). Discordants were tested by bidirectional Sanger sequencing. Results: The PPA between THxID™ BRAF and RGI IUO was 96.2% for V600E, 97.8% for V600K, and OPA was 95.4%. The NPA between the two assays was 96.5%. The same variables excluding the invalid samples were: 98.2%, 97.8%, 97.8%, and 97.5%, respectively. A total of 16 discordant samples (1.7%, 16/926) with valid results from both assays were tested with bidirectional Sanger sequencing. 4 instances were shown as correctly identified by THxID™ BRAF, 9 instances by RGI IUO, and 3 instances were discordant to both assays. There were 22 samples with invalid results reported from both assays, 13 samples reported invalid from THxID™ BRAF and valid from RGI IUO, and 11 samples reported invalid from RGI IUO but valid from THxID™ BRAF. The total invalid rate from THxID™ BRAF and RGI IUO were 3.8% (35/926) and 3.6% (33/926) respectively. Conclusions: THxID™ BRAF assay and RGI IUO assay showed a very high concordance rate, indicating that the prospective clinical outcome with RGI IUO assay can be transposed retrospectively to THxID™ BRAF assay, enabling the selection of patients with BRAF V600E and /or V600K mutation in melanoma tumors for targeted therapy. MC13-0054 High precision of a novel companion diagnostic assay for the detection of BRAF V600E/V600K in formalin-fixed paraffin-embedded melanoma samples L. O’Donnell 1 , W. Sweet 1 , X. Lu 1 , F. Meynier 2 , A. Derome 3 , L. Ganee 4 , F. Poyet-Gelas 3 , A. Martin 5 , M. Casey 6 , N. Kertesz 7 . 1 Clinical Affairs, BioMerieux Inc., Durham, USA; 2 BioMath, BioMerieux Inc., Grenoble, France; 3 V & V, BioMerieux Inc., Grenoble, France; 4 Product Design & Development, BioMerieux Inc., Grenoble, France; 5 Oncology, GSK, Collegeville, USA; 6 Statistics, GSK, Collegeville, USA; 7 Pharmaceutical Liaison Manager, Response Genetics Inc., Los Angeles, USA Background: The THxID™ BRAF kit is an in vitro diagnostic device intended for the qualitative detection of the BRAF V600E and V600K mutations in DNA samples extracted from formalin-fixed paraffin embedded (FFPE) human melanoma tissue. The THxID™ BRAF kit is a real-time PCR test on the ABI 7500 Fast Dx system intended to be used as an aid in selecting melanoma patients whose tumors carry the BRAF V600E mutation for treatment with dabrafenib [Tafinlar® ] and as an aid in selecting melanoma patients whose S27 tumors carry the BRAF V600E or V600K mutation for treatment with trametinib [Mekinist™]. Purpose/Objective: The purpose of this study is to evaluate the precision of the THxID BRAF assay by using a 15-member panel. Materials and Methods: A 15-member panel was formulated to include 3 genotypes [wild type (WT), V600E, V600K], 3 mutant DNA concentrations (LoD claim, Med-High positive, High Melanin content), and 2 tissue types (Skin, Lymph node). The V600E and V600K samples were reconstituted respectively by mixing DNA eluates of mutant and WT samples. The panel was tested at three sites on three lots for three weeks, with each panel member in duplicate. The precision was determined by the percentage of correct identifications of the samples’ mutation status, with the corresponding two-sided 95% score confidence interval (CI). Results: The results showed 100% correct identification of the sample mutation status for fourteen panel members, and 83.3% correct identification for one panel member carrying V600E mutation with mutant DNA concentration close to limit of detection (LoD) claim, in which the low mutation content of the panel member was confirmed by pyrosequencing. The overall combined percentage of correct identification is 98.9%. For one panel member carrying V600K mutation with high melanin content, 100% correct identification was yielded with the diluted samples following the Instructions For Use (IFU) specifications for repeating. The remaining panel members with high melanin content were 100% correctly identified from original DNA extracts. Conclusions: The study demonstrates high precision of the THxID™ BRAF Assay for the detection of BRAF V600E/V600K mutations in FFPE samples from melanoma patients, indicating the assay can be used to aid in the selection of a population with BRAF V600E/V600K mutations for targeted therapy. MC13-0055 Development of a companion diagnostic for pegylated recombinant human PH20 L. Jadin, L. Huang, Q. Zhao, G. Wei, H.M. Shepard, A.B. Gelb, P. Jiang. Research, Halozyme Therapeutics Inc., San Diego, USA Background: The polysaccharide hyaluronan (HA) is a normal constituent of the extracellular matrix. Up-regulation of HA production by tumor and/or associated stromal cells often correlates with tumor progression. A pegylated recombinant human hyaluronidase (PEGPH20) has been shown to deplete tumor HA and lead to increased tumor perfusion, thereby enhancing the activity of anticancer agents in preclinical models. A phase 1b study of gemcitabine plus PEGPH20 in patients with stage IV previously untreated pancreatic ductal adenocarcinoma (PDAC) showed efficacy, particularly when intratumoral HA content was high in pretreatment samples. Therefore, developing a companion diagnostic is important to select patients likely to benefit from therapy by assessing pretreatment biopsies and monitoring the effects of treatment. Purpose/Objective: To evaluate a pseudo-immunohistochemical (IHC) method for the detection of HA in malignancies as a companion diagnostic for PEGPH20 therapy. Materials and Methods: Pseudo-IHC was performed using a biotinylated recombinant protein consisting of a modified human TSG-6 HA-binding domain fused with the Fc fragment of human IgG1a (HTI-601). Quantitative image analysis was performed to determine the percentage of pixels above a defined intensity threshold relative to the total pixels stained and counterstained. The specificity of staining was assessed by pre-incubation of tissues with an HA-degrading enzyme, either Streptomyces hyaluronidase or recombinant human PH20, depending on the specimen type. Results: HTI-601 displayed high binding affinity, with a Kd of approximately 1E−10 M for 150 kDa HA. It proved very sensitive and specific for HA detection in formalin-fixed, paraffin-embedded (FFPE) human normal tissues, human cancers and xenograft tumors by pseudo-IHC. The method was able to detect a significant reduction in HA content of PC3 human prostate cancer and BxPC3 human pancreatic cancer xenograft tumors following PEGPH20 treatment. Conclusions: This study demonstrates that pseudo-IHC using HTI-601 is a specific, sensitive and accurate method for evaluating HA content in FFPE human normal tissues and neoplasms. The value of this method as a companion diagnostic for PEGPH20 is currently under investigation in Phase 2 clinical trials of PDAC. S28 MC13-0057 A comprehensive IT platform to support GTEx operation C. Shive 1 , L. Qi 1 , D. Tabor 1 , P. Hariharan 1 , S. Wu 1 , K.S. Um 1 , J. McLean 1 , N. Lockhart 2 , P. Guan 3 . 1 BBRB Support Program, SAIC-F, Bethesda, USA; 2 Division of Genomics & Society, National Human Genome Research Institute, Bethesda, USA; 3 BBRB/CDP/DCTD, National Cancer Institute, Bethesda, USA Background: Designed to build upon Genome Wide Association Study (GWAS) findings, the NIH Common Fund’s Genotype-Tissue Expression (GTEx) project aims to study gene expression and regulation across multiple human tissues (30+ tissue types) from approximately 1000 healthy normal donors. It is expected to provide valuable insights into gene regulation and its tissue specificity, identify correlation between genetic variations and variations in gene expression levels as expression quantitative trait loci (eQTLs), and help to understand inherited susceptibility to diseases. Purpose/Objective: To meet the challenge of GTEx requirements for collecting and tracking high quality biospecimen samples, a custom-built software system named Comprehensive Data Resources (CDR) was developed to support sample collection work flow, clinical data entry, case management, and review and curation of study data. Materials and Methods: CDR is built with combination of technologies from Grails, Oracle, Groovy, jQuery, Apache Solr. Results: The CDR provides secure user access to case and sample data based on pre-defined roles and privileges. Personally Identifiable Information (PII) and Protected Health Information (PHI) are restricted to a limited data set (LDS) and to authorized users through dynamic content redaction. Intuitive graphic user interfaces for the Biopecimen Source Sites (BSS) streamline data entry workflow by strictly following SOPs for sample collection and processing. Contextual automated data checks and business rule validations confirm data integrity and SOP adherence simultaneously. Web services APIs allow the Pathology Resource Center to access digital imaging data from tissue slides housed remotely at the Comprehensive Biospecimen Resource (CBR). API’s connect to CBR’s LIMS systems for real-time sample inventory data. De-identified GTEx data is provided via a private API with the Broad Institute (LDACC) before the final release into dbGaP. The reporting and analytics module supports data analysis and aggregation, report generation and real-time operational data snapshots. Conclusions: CDR is a distributed web-based system designed to support GTEx operation from pilot phase to full scale-up stage. It manages and maintains multi-dimensional data models around each donor case (average 500+ data elements/case). As an efficient case management tool capable of connecting to various remote informatics systems, CDR could be adapted to the broader biobanking community with the flexibility of building user-defined work flows in the system. MC13-0058 Inflammation-mediated epigenetic background for switching from normal program to cancer growth V. Halytskiy. Molecular Immunology Department, Palladin Institute of Biochemistry of the National Academy of Sciences of Ukraine, Kiev, Ukraine Background: Although inflammation is closely associated with tumor growth, molecular basis of this interrelation remains unclear, especially when pathogens do not damage the DNA. However, the inflammation entails regular changes in the expression of cell microRNAs (miRNAs). Expression of miRNAs miR-155, miR-21, miR-146a, miR-125b, miR-31, miR-34c, miR-200, miR-203 and miR-205 is usually up-regulated whereas expression of miRNAs miR-7a/b, miR-34a, miR-143, miR-145, miR-320a, miR-375, miR-379 and miR-434-3p is down-regulated. Purpose/Objective: This investigation aims to identify in what way the shifts in miRNA expression pattern contribute to the cell transformation and tumor growth. Materials and Methods: miRNA targets within gene transcripts were predicted in silico using TargetScan software. Results: miRNA miR-143 can silence abl2, bcl-2, erbB3 and MYST2 genes. miR-145 targets E2F3, RASA1/2, CDK6, erbB3/B4, ESR1, ACTB/G1, KDM1B/2B and Elp3 genes. miR-320 suppresses E2F1/3/7, RASA1, CDK6, p57, ESR1, ITGB5, KDM1B and KAT6B genes. Down-regulation of these miRNAs causes derepression of genes encoding histone acetyltransferases, histone demethylases, key elements of proliferative and antiapoptotic signal pathways as well as genes responsible for cell motility and abnormal adhesion. Up-regulated miRNA miR-155 silences HDAC2/4/9, SIRT1, EZH1, Poster Presentations SETD7, CLDN1, CGN, OCLN, F11R (JAM-A) and TGFBR2 genes and genes coding α-actinins. miR-21 can target DNMT3B, HDAC2, SIRT5, SETD6/8, SUV39H2, CLDN1, CGN, CADM1, VCL and TGFBR2 genes. Conclusions: Inflammation is associated with miRNA expression shifts that lead to increasing of cell proliferation and survival as well as to silencing of antiproliferative and proapoptotic genes. Also, up-regulated miRNAs suppress genes encoding components of cytoskeleton and intercellular junctions. This results in alterations in cell–cell adhesion, impairs contact inhibition, facilitates cell motility and migration. Furthermore, up-regulated miRNAs silence genes encoding the histone deacetylases, histone methyltransferases and de novo DNA methyltransferase. This causes increasing of overall level of chromatin acetylation and expression and, therefore, makes possible the reactivation of silent oncogenes as well as transposons, which can rapidly lead to dramatic increase of DNA damage level and genome destabilization. Thus, inflammation creates epigenetic background for cell transformation as well as for tumor promotion and metastatic spread. MC13-0059 T-cell infiltration (TCI) observed on whole liver colorectal metastases (LCM) resected after preoperative treatment is a prognostic survival factor M. Van den Eynde 1 , B. Mlecnik 2 , J.P. Machiels 3 , D. Debetancourt 4 , A. Mourin 5 , J.F. Gigot 6 , N. Haicheur 7 , F. Marliot 7 , F. Pagès 7 , J. Galon 2 . 1 Oncology, Université catholique de Louvain, Brussels, Belgium; 2 Cancer immunology, Centre de Recherche des Cordeliers, Paris, France; 3 Oncology, 4 Centre du Cancer, 5 Pathology, 6 Digestive Surgery, Université Catholique de Louvain, Brussels, Belgium; 7 Immunology, Hopital Européen Georges Pompidou, Paris, France Background: Colorectal cancer TCI is a strong prognostic factor for survival after primary tumor resection. Curative surgery of LCM is the only hope for cure of metastatic patients (pts). Nevertheless, 70% of them will relapse. Purpose/Objective: TCI analysis of LCM is poorly characterized and could be a prognostic factor as in primary tumor. Materials and Methods: Pts engaged for curative liver surgery after preoperative treatment with available FFPE blocks for all resected LCM, were included. An immunoscore (IS), defined by the TCI in the center (CT) and the invasive margin (IM) for each LCM was determined using whole-slide quantitative immunohistochemistry (markers: CD3, CD8, CD45RO). The mean value of the 3 most infiltrated fields (0.8 mm2 ) for each markers was defined in the CT and IM for all LCM. The total number of high densities (Hi, above the cut-off at the median density) in CT and IM for each marker was used to stratify pts for the IS. The markers were combined 2 by 2 in CT and IM (CD3-CD8, CD3-CD45RO, CD8-CD45RO) and finally regrouped to an IS of 0–2 Hi (IS0–2: low TCI) or 3–4 Hi (IS3–4: high TCI). For pts with multiple LCM; the median value of all densities, the least and the most infiltrated LCM/pt were analyzed. Cumulative DFS/OS analyses were performed using the Kaplan-Meier estimator. OS/DFS analyses were made using univariate Cox regression and compared by log-rank tests (IS0–2 vs 3–4). Results: 59 patients (M/F 1.1, 203 LCM, mean 3.4/pt, synchr/metachr 5.4) were included. IS3–4 in the least infiltrated metastasis is significantly associated with OS and DFS for all markers combinations. LCM/pt Median of all Markers Survival CD3-CD8 DFS OS DFS OS DFS OS DFS OS DFS OS DFS OS DFS OS DFS OS DFS OS CD3-CD45RO CD8-CD45RO Least infiltrated CD3-CD8 CD3-CD45RO CD8-CD45RO Most infiltrated CD3-CD8 CD3-CD45RO CD8-CD45RO HR Log-rank Months (IS0–2 vs 3–4; p-value (IS0–2 vs 3–4) 95% CI) 1.2 (0.7–2.3) 2.4 (0.7–7.3) 1.5 (0.8–2.9) 2.2 (0.8–5.9) 1.0 (0.6–2.0) 1.0 (0.4–2.9) 1.8 (1.0–3.4) 8.8 (2.0–39.1) 2.5 (1.3–4.9) 4.0 (1.2–14.2) 2.0 (1.0–3.7) 2.6 (0.9–7.6) 1.0 (0.6–1.9) 2.1 (0.7–6.6) 1.7 (0.9–3.2) 3.7 (0.8–16.2) 1.6 (0.8–3.2) 2.4 (0.7–8.4) 0.48 0.12 0.16 0.11 0.87 0.17 0.05 0.0007 0.004 0.01 0.03 0.06 0.93 0.17 0.12 0.06 0.16 0.16 8.0 vs 14.9 27.9 vs NR 8.0 vs 17.0 31.8 vs NR 8.4 vs 16.0 Poster Presentations Conclusions: The TCI score (IS0–2 vs 3–4) determined in the least infiltrated LCM/pt is a prognostic factor. MC13-0062 Glioblastoma patients with arterio-venous normalization during anti-angiogenic therapy have prolonged survival K. Emblem 1 , T.T. Batchelor 2 , E.R. Gerstner 2 , D.G. Duda 3 , M.C. Pinho 1 , M. Ancukiewicz 3 , P.Y. Wen 4 , B.R. Rosen 1 , A.G. Sorensen 5 , R.K. Jain 3 . 1 Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, USA; 2 Department of Neurology, Massachusetts General Hospital, Boston, USA; 3 Department of Radiation Oncology, Massachusetts General Hospital, Boston, USA; 4 Center for Neuro-Oncology, Dana-Farber Cancer Institute, Boston, USA; 5 Siemens Healthcare North America, Siemens, Boston, USA Background: Vessel architectural imaging (VAI) is a new paradigm in magnetic resonance imaging (MRI) of cancer for assessment of vessel type and relative oxygen saturation (SO2 ) levels [1]. Purpose/Objective: We used VAI to monitor vascular remodeling in newly diagnosed glioblastoma (nGBM) patients treated with the anti-angiogenic agent cediranib and chemo-radiation. Materials and Methods: In this IRB-approved phase I/II study of cediranib (NCT00662506), an oral pan-VEGF inhibitor [2], 40 patients with nGBM were imaged weekly for 6 weeks during therapy and then monthly thereafter. Our MRI protocol included a gradient-echo and spin-echo perfusion sequence for VAI [1]. Here, a temporal shift in the gradient-echo signal relative to that of spin-echo is used to assess in vivo information on vessel type (fast-inflow arterioles or slow-inflow venules) and SO2 levels. As a control group, we included 14 patients with nGBM enrolled in a parallel study at the same S29 institution (NCT00756106). These patients underwent identical MRI and chemo-radiation, but without cediranib. Results: Twenty-one patients were identified as responders by the tumor arterio-venous ratios overlapping that of reference tissue at a minimum of two time points during therapy (Fig. 1). Responding patients had prolonged overall survival (OS) compared to 19 non-responding patients (median OS = 696 d vs. 381 d, Cox; P <0.01, corrected for MGMT methylation status). At day +21, the median SO2 level of non-responders was 20% higher than pre-treatment, compared to −10% for responders (Mann-Whitney; P <0.05), indicating more hypoxia in non-responding tumors. Conclusions: Using VAI we show for the first time in humans that antiangiogenic therapy induce vascular normalization in nGBM by arterio-venous remodeling and improved oxygenation. Patients who responded to therapy by mimicking the vessel architecture of normal tissue had prolonged survival. References: [1] Emblem KE, et al. Nature Med 2013; in press. [2] Batchelor TT, et al. J Clin Oncol 2010;28:2817–23. MC13-0063 MicroRNA-mRNA networks for identification of biomarkers for breast cancer A.F. Evangelista, T. Macedo, R.J.S. Oliveira, M.M.C. Marques. Molecular Oncology Research Center, Research and Teaching Institute, Barretos Cancer Hospital, Barretos, Brazil Background: MicroRNAs (miRNAs) are highly conserved small noncoding molecules (∼22 nt) involved in post-transcriptional regulation. These molecules are playing important role in a range of disease, especially in cancer. Considering the importance of miRNAs as non-invasive biomarkers, the identification of mRNA targets is important to elucidate molecular mechanisms for future target therapies. Purpose/Objective: Our aim was to identify microRNAs biomarkers by integration of mRNA and miRNAs expression profiles using bioinformatics tools. Materials and Methods: In this study we performed mRNA and miRNA Agilent microarrays of breast cancer cell lines, classified in different molecular subtypes, such as luminal (T47D, MCF-7 and MCF-7/AZ), Her2 overexpression (SK-BR3 and BT-20), triple negative (MDA-MB-231 and Hs578T) and normal phenotype (HB4A). All the microarray data analyses were performed using the R environment. The anti correlated expression profiles of differentially expressed miRNAs and mRNAs were considered and integrated networks were constructed using GenMiR algorithm. The targets identified were compared with those predicted by miRDip tool and functional analyses were performed using DAVID and Panther databases. Results: Unsupervised analysis revealed separation according to molecular subtype in both mRNA and miRNA expression profiles. It was constructed miRNA-mRNA networks for each one of the 70 miRNAs differentially expressed, identified according to ANOVA p<0.01 bonferroni-corrected, some of them previously associated with breast cancer progression, such as miR-10, miR-21 miR-31, miR-221/222 and let7a. Their targets could be grouped according to biological mechanisms, pathways in cancer and breast cancer molecular subtype. Conclusions: In conclusion, the present study identified several miRNA biomarkers and targets of interest in miRNA-mRNA networks. Financial support: FAPESP. MC13-0064 Identification of a putative precursor lesion of papillary thyroid carcinoma by cyclin D1 overexpression and p38 MAPK phosphorylation M. Lamba Saini 1 , B. Weynand 1 , J. Rahier 1 , M. Mourad 2 , M. Hamoir 3 , E. Marbaix 1 . 1 Pathology, Université catholique de Louvain, Brussels, Belgium; 2 Surgery, Université catholique de Louvain, Brussels, Belgium; 3 Otorhinolaryngology, Université catholique de Louvain, Brussels, Belgium Figure 1. Vessel architectural imaging. (A) Pre-treatment MRI including VAI for identification of vessel type. (B) Corresponding MRI during therapy. Note the change in the arterio-venus ratio compared to pre-treatment. (C) Average arterio-venous ratios for responding and non-responding patients. Responders mimic the arterio-venous ratios of healthy tissue during therapy. Similar to non-responders, a collective vascular normalization is not observed in the control study. (D) Patients with arterio-venous normalization have prolonged OS. Background: Papillary thyroid cancer (PTC) is the commonest endocrine malignancy. Though significant progress has been made to understand the pathways involved in the tumorigenesis of PTC, no precursor lesion has been identified yet. Purpose/Objective: The present study aims to identify and understand the precursor lesion of PTC and its molecular markers. Materials and Methods: Thirteen cases of metastatic PTC, papillary microcarcinoma and follicular variant of PTC (FVPTC) were identified from S30 a histological review of 510 cases. In addition, 13 cases of a subset of follicular adenomatoid nodules with focal areas showing nuclear features characteristic of PTC, identified as putative PTC precursor lesion, were also analyzed. Immunohistochemical analysis of galectin-3, HBME-1, CK 19 and the proliferation markers Ki 67 and cyclin D1 was performed. Lesions were analyzed for cyclin D1 gene amplification by fluorescent in situ hybridization. PTC also frequently carries several genetic alterations in genes coding for proteins that activate the mitogen-activated protein kinases (MAPK) signaling pathway, which plays a key role in the regulation of cell growth and differentiation. The role of MAPK pathway activity in PTC was investigated by immunohistochemical labelling of phosphorylated ERK, JNK and p38. Results: All putative precursor lesions showed immunolabelling of cyclin D1 and Ki 67; 11/13 cases showed immunolabelling of CK 19; 10/13 cases showed immunolabelling of HBME-1 and 4/13 cases showed immunolabelling of galectin-3. Surrounding adenomatoid areas showed no to faint focal staining of cyclin D1, HBME-1 and galectin-3 in all thirteen cases. A low rate of cyclin D1 gene amplification (Figure) was identified in a significant proportion of cells in the putative precursor lesion as compared to surrounding benign adenomatoid areas. ERK and JNK activation was seen in 50 and 35 percent of PTC cases with immunolabelling in less than 10 percent of cells. p38 MAPK phosphorylation was seen as abundant cytoplasmic immunolabelling in 55% of PTC cases and 60% of putative precursor lesion cases. A one way ANOVA test showed significant difference between the ERK, JNK and p38 phosphorylation (p<0.01). Poster Presentations (GEP) to sub-classify patients and predict outcome. Although promising, these assays require fresh frozen tissue and for most health centers remain prohibitively out of reach due to cost. Purpose/Objective: We sought to confirm our original observation that increasing levels of VEGF and BCL6 were associated with reduced outcome using an expanded patient cohort evaluated with quantitative immunofluorescence (QIF). In addition, we wanted to assess whether QIF, using biomarkers currently employed in the Colomo and Hans algorithm, is comparable to GEP data, for risk stratification. Materials and Methods: Complete clinical and QIF data for 118 patients from three different sites in Spain were evaluated to predict response to standard R-CHOP. Multiplex QIF with CRI Nuance imaging software was performed for selected markers including: MUM1, CD10, CD20, BCL6, BCL2, VEGF, and CMYC. The Colomo and Hans algorithm’s using QIF were compared with previously obtained GEP data on 36 patients classified as either ABC (n=14) or GCB (n=22). Kaplan-Meier survival function curves, concordance index (CoI) and multivariate models were employed to associate marker expression with OS and PFS. Results: For all three cohorts, both low IPI and female gender predicted good overall survival (p<0.001, respectively). We validated our previous VEGF cutpoint and confirmed high levels were associated with poor PFS (0.042). We then re-calibrated the VEGF cut-point on 118 patients and further improved this association for both OS (p=0.003) and PFS (p=0.004). We also validated our prior BCL6 cut-point and confirmed high levels were associated with poor PFS (p=0.001) and after re-calibration the results were even more significant (OS, p=0.008; PFS, p=0.001). In addition, increasing amounts of CMYC were associated with poor PFS and OS, respectively (p=0.014, 0.021). Mum1 by QIF alone was able to risk stratify patients classified by GEP (p<0.05) suggesting a role for IF and other markers in future algorithms. Best multivariate model to predict OS or PFS included the IPI score and BCL6. Clinical heterogeneity impacted on selected marker performance between cohorts. Conclusions: VEGF and BCL6 may be useful markers for predicting likely patient-specific response to current therapies. QIF with additional biomarkers and integrated with clinical data should be comparable with GEP to risk stratify patients in the future. MC13-0066 ORILAB a functional and molecular imaging corelab for cancer research Figure 1. Amplification of cyclin D1 gene seen in the putative precursor lesion using FISH. Cyclin D1 gene appears in red whereas chromosome centromeric repeat region is green (bar = 20 μm). Conclusions: Increased expression of cyclin D1 and amplification of its gene along with immunolabelling of HBME-1 and p38 phosphorylation in areas showing cytological features of PTC within follicular adenomatoid nodules suggest that these areas could correspond to a precursor lesion of follicular variant of PTC. Increased expression of p38-MAPK cascade in PTC variants indicate that it is functional in PTC. p38-MAPK hyper-expression in the precursor lesion can act as a potential complementary marker. However, its role in the tumorigenesis of PTC needs to elucidated. MC13-0065 Quantitative assessment of VEGF, BCL6 and CMYC predicts outcome in patients with DLBCL treated with R-CHOP: Implications for guiding future treatment selection M. Donovan 1 , N. Erill 2 , P. Puig 2 , A. Colomer 2 , L. Colomo 3 , E. Campo 4 . 1 Pathology, Mt. Sinai Medical Center, New York, USA; 2 Pathology, Althia Health S.L., Barcelona, Spain; 3 Pathology, Hospital Clinic, Barcelona, Spain; 4 Pathology, Hospital Clinic, University of Barcelona, Barcelona, Spain Background: The pathogenesis of DLBCL is both complex and heterogeneous, and pathogenetic mechanisms remain largely unknown. Although the addition of Rituximab (R) to CHOP has significantly improved outcome, approximately 30–40% of patients ultimately die of their disease supporting the need to improve risk stratification and understand response to therapy. The international prognostic index (IPI) remains the “gold standard” for progression free survival (PFS) and overall survival (OS); however, more recently, several groups have reported the utility of gene expression profiles L. Belenguer-Querol 1 , T. Guiot 2 , C. Garcia 2 , S. Chao 3 , S. Drisis 3 , M. Lemort 3 , P. Flamen 2 . 1 Oncology Related Imaging coreLAB, Institut Jules Bordet, Brussels, Belgium; 2 Nuclear Medicine, Institut Jules Bordet, Brussels, Belgium; 3 Radiology, Institut Jules Bordet, Brussels, Belgium Background: Functional and molecular imaging techniques allow the identification of additional cancer imaging biomarkers for prognosis or prediction of patient response to a specific treatment. Biomarkers technical feasibility and diagnostic accuracy are available yet, however large scale multicentric trials are required to validate them before implementation in clinical routine. Setting up these studies involving functional and molecular imaging becomes cumbersome due to image acquisition intra and inter-center variability and differences in image quality and analysis. Purpose/Objective: An Oncology Related Imaging coreLAB platform, ORILAB has been established within a university dedicated cancer hospital environment to facilitate the use of imaging biomarkers in multicentric studies through standardization and harmonization of imaging data acquisition, analysis and storage. Materials and Methods: ORILAB exploits available in-house academic resources from multiple disciplines including physicists, engineers and physicians. The platform entails definition of imaging protocols, a Clinical Data Management System (CDMS) to electronically capture imaging trial data from participating sites or from image reviewers, a Quality Management System (QMS) to assure quality of source data and a centralized review (CR) for image interpretation. The CR is constituted by Nuclear Medicine and Radiology physicians with extensive experience in the development of new cancer biomarkers and with a special interest in translating these imaging procedures into clinical trials which demands greater standardization endeavors. Results: Imaging protocols describe guidelines for standardized acquisition and transmission of imaging data reducing intra and inter-center variability. Electronic encoding of imaging trial data results in faster and more reliable input of trial data in multicentric studies by an earlier detection of noncompliance and queries. The QMS allows the adaptation of the CDMS workflow to Poster Presentations imaging data characteristics, such as automatic extraction of relevant imaging information and visualization of trial images. New analysis tools developed using accepted algorithms and models from the literature but absent on commercial platforms allow more efficient and reproducible extraction of relevant imaging parameters. Imaging data analyzed with identical tools and in a centralized review approach diminishes inter-observer variability. Conclusions: Regardless of generalization of morphology-based response evaluation criteria, a large panel of other functional and molecular imaging techniques for evaluating new drugs in oncology is underused due to a lack of standardization. We created a qualified multidisciplinary team to manage all clinical trial stages involving imaging data under strict standardization and quality procedures, reducing intermediaries and contributing to a faster adoption of innovative image biomarkers in multicentric trials. MC13-0067 Association between Kras mutation in circulating cell-free tumor DNA and clinical progression of colorectal cancer in patients receiving cetuximab M. Del Re 1 , P. Ulivi 2 , F. Belcari 1 , A. Passardi 2 , W. Zoli 2 , D. Amadori 2 , R. Danesi 1 . 1 Clinical and Experimental Medicine, University of Pisa, Pisa, Italy; 2 Biosciences Laboratory, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST) IRCCS, Meldola, Italy Background: Cell-free tumor DNA (cftDNA) is released into the circulation and its recovery from plasma is a minimally invasive alternative to tumor biopsy for selection of appropriate treatments based on molecular profiling. The periodic monitoring of cftDNA for the identification of molecular changes associated with resistance to target-specific treatments is a potentially powerful approach to the optimization of drug sequences based on the evolving mutational status of the tumor. Purpose/Objective: Since Kras-mutated cell clones may arise under the selective pressure exerted by cetuximab on a wild type tumor, the present study screened Kras mutations in a small cohort of patients resistant to cetuximab treatment. Materials and Methods: Three Kras wild-type patients at diagnosis with radiologic evidence of disease progression (PD) while on cetuximab were screened for the developent of Kras mutations as a mechanism of secondary resistance. Primary tumor at diagnosis and peripheral blood samples (6 ml) were drawn from patients at pre-cetuximab time-point (baseline) and at PD during cetuximab treatment. DNA was extracted from plasma with QIAamp Circulating Nucleic Acid Kit to recover DNA fragments of ≤1000 bp. PCR amplification was carried out with a QX100™ ddPCR™ System (Bio-Rad) on 20-μL samples containing cftDNA and TaqMan probes for KRAS G12D (35G>A) and G12V (35G>T) labeled with FAM/VIC. Samples were then loaded into a droplet reader, which discriminates the difference in fluorescence amplitudes on the basis of target gene amplification. Results: Testing of the cftDNA samples at PD in patient 1 and 2 showed the presence of the Kras mutation G12V and the G12D in patient 3. Interestingly, pre-cetuximab cftDNA was positive for patients 2 and 3, but was not sufficient to test the mutational status of Kras in patient 1. Conclusions: ddPCR is a third-generation PCR technique for highly sensitive detection of DNA fragments. Detection of mutations in cftDNA by advanced technological platforms has important applications in the monitoring of patients for the occurrence of secondary mutations that render their tumors resistant to target-specific anticancer agents. Acknowledgments: This study was funded in part by MIUR/PRIN 2011-2012 (Rome, Italy). MC13-0068 National Cancer Institute (US) clinical assay development program B. Conley 1 , B.A. Conley 1 , T.G. Lively 1 , J. Rohan 2 , S. Bharti 1 , M.M. Cavenagh 1 , C. Lih 2 , P.M. Williams 2 . 1 DCDT/CDP, National Cancer Institute, Bethesda, MD, USA; 2 DCTD/FNLCR, National Cancer Institute, Bethesda, MD, USA Background: Discoverers of predictive or prognostic molecular features for cancer often do not have the resources to analytically validate a “locked down” assay. Purpose/Objective: The Clinical Assay Development Program (CADP) was created in 2011 to provide accessible resources to assay developers to transition promising markers/signatures into validated assays for use in clinical trials (http://cadp.cancer.gov/). S31 Materials and Methods: Clinical (CLIA accredited) laboratory services are provided through the Clinical Assay Development Network of 8 conracted laboratories and/or by a research and CLIA accredited laboratory at FNLCR. Tissue resources are also provided. Applicants from academia, industry or government are eligible and must provide one intended clinical use, a prototype assay, and relevant marker prevalence information. Applicants must also describe the clinical need, the current state of the assay, and future plans for assay development (e.g., use in a clinical trial). Examples of services provided include platform migration, standard operating procedure (SOP) development, cutpoint validation, statistical assistance, etc. Applications recommened after evaluation by a panel of outside experts are reviewed internally to ensure availability of appropriate resources. The successful application is overseen by a project management team: project manager, National Cancer Institute subject matter experts, expertise from contracted resources and the assay submitter. Intellectual property remains with the assay submitter. After validation, assay perfomance specifics and SOPs are returned to the assay submitter. Results: To date, two assays have been validated (in a 14 month time frame for both) and 5 are in project management. Platforms include gene expression analysis, ELISA, IHC/IFA, DNA sequencing, FISH, PCR. The success rate for applications is about 25%. Additional applicants have used the advice of the CADP team to improve development strategy. Conclusions: There is continued need for education and assistance in assay development for assays intended for use in clinical trials. Each assay presents unique challenges and solutions. Creativity is needed in finding appropriate specimens for assay validation. To date, the initial results of CADP are promising. MC13-0069 Obtaining maximal information from limited FFPE tissue specimens: Development and validation of a 4-gene custom NGS panel for a GIST clinical trial P. Fang 1 , Z. Yan 1 , J. Kristof 1 , J. Staha 1 , K. Pelak 1 , M.R. Palmer 2 , C. Tribouley 3 , C. Spittle 1 , C. Galderisi 1 , J. Li 1 . 1 Assay Development, MolecularMD Corp, Portland, USA; 2 Oncology Translational Medicine, Novartis Pharmaceuticals Corporation, Cambridge, USA; 3 Oncology Correlative Sciences, Novartis Pharmaceuticals Corporation, Florham Park, USA Background: Limited tissue availability and FFPE sample quality present challenges for cancer mutation profiling, particularly for solid tumors such as GIST. Maximizing the data from specimens requires advanced profiling techniques; focused NGS panels provide higher sensitivity and more comprehensive sequence coverage than other methods and enable multiplexed examination of several genes related to a given therapy. Purpose/Objective: A custom GIST panel was designed to profile 4 drug pathway related genes using minimal DNA from FFPE tissue, which may aid in the interpretation of drug responses. Materials and Methods: The region of interest (ROI) includes 20 exons across c-KIT, PTEN, PIK3CA and PDGFRα, a total of 2.1 kb. Primers were designed using Ion AmpliSeq Designer and MolecularMD’s proprietary primer design method. 28 FFPE specimens representing primarily GI tumors were analyzed. Sequencing was performed on the Ion PGM, and Data were analyzed with Torrent Suite 3.4.2 and the MolecularMD analysis pipeline. Results: Assay Specificity: No critical variants (non-synonymous and nonintronic) were detected in 6 normal FFPE DNA samples. LOD: The LOD of the assay is 5% for SBS and indels, as determined by sequencing cell line dilutions. Precision: 100% of the critical variants in the DNA used in the precision study were reproducibly detected in 3/3 runs. Accuracy: 47 critical variants with frequencies ranging from 7% to 79% (28 SBS, 11 deletions, 3 deletions with insertions, and 5 insertions) were detected in 28 tumor FFPE samples. The maximum observed deletion and insertion sizes were 51 bp and 45 kb, respectively. Variants were observed in all 4 of the targeted genes. All variants with sufficient DNA quantity (n=46) were confirmed by direct Sanger sequencing (variants with frequency >10%) or Sanger sequencing following mutation enrichment (variants with frequency <10%). Conclusions: This validation study demonstrates that with only 30 ng DNA input, the KIT-PDGFRα-PIK3CA-PTEN custom NGS Panel is capable of accurately detecting mutations, including SBS and small indels, in 20 exons across 4 genes. This custom panel is tailored to the demands of a targeted drug trial, and was designed to identify mutations in genes and regions most relevant to the interpretation of trial results. A custom NGS panel is scalable, in terms of ROI size and number of samples to be tested, and S32 can be employed to maximize the breadth and depth of profiling information generated from a given tissue specimen. MC13-0070 Whole blood RNA signature as prognostic and predictive biomarker in genitourinary malignancies W. Oh 1 , U. Chippada Venkata 1 , L. Wang 2 , E. Reese 1 , T. Yee 1 , T. Kochukoshy 1 , C. Tsao 1 , M. Galsky 1 , J. Zhu 2 , Y. Gong 1 . 1 Hematology/Medical Oncology, Mt. Sinai Medical Center, New York, USA; 2 Genetics and Genomic Sciences, Mt. Sinai Medical Center, New York, USA Background: We developed a whole-blood RNA transcript-based 6-gene signature (consisting of ABL2, SEMA4D, ITGAL, and C1QA, TIMP1, CDKN1A) which separated patients into high and low risk survival groups in castration resistant prostate cancer (CRPC) (Ross et al. Lancet Oncology, 2012). Purpose/Objective: In this study, we validate the prognostic signature on an updated qPCR platform and examine dynamic changes in the signature during disease progression and with treatment. The application of the signature across other genitourinary malignancies is also explored. Materials and Methods: Whole blood was collected in PAX gene RNA tubes from prostate (n=53), bladder (n=18) and renal (n=13) cancer patients in different stages of the disease and on various treatment regimens. A subset of patients had serial blood draws at multiple time points. Pre and post treatment blood was also drawn from CRPC patients on an oral satraplatin clinical trial (NCT01289067; PI: Oh). The 6-gene survival score for each blood draw was derived as previously described. Results: The 6-gene score generated on a new qPCR platform is highly reproducible (CV <3% between triplicates). The signature remains highly prognostic, separating CRPC patients into high and low risk groups of death (p-value of logrank test <0.001). Interestingly, patients in earlier disease stages had significantly lower scores compared to patients with CRPC (p-value of t-test = 0.036). Patients (n=15) who had serial blood draws showed different trends of scores with time, with a subset of patients demonstrating rises preceding death. In a pilot study to determine the impact of treatment on the prognostic score, we observed that 2 satraplatin responders showed stable scores over the course of one year of therapy, whereas 2 progressing non-responders showed significantly increased scores after 2 and 4 months of treatment, indicating the potential use of the signature to assess treatment outcome. Further, preliminary analysis showed that the 6-gene signature may have prognostic value in bladder and renal cancers, which will be confirmed in a larger cohort. Conclusions: Our results suggest that the 6-gene whole blood RNA signature has compelling prognostic ability independent of platform, provides crucial prognostic information and is possibly indicative of an individual’s response to disease progression and therapy. MC13-0072 Tumor perfusion during bevacizumab and irinotecan in recurrent glioblastoma: A multimodal approach M. Eoli 1 , A. Di Stefano 2 , D. Aquino 3 , A. Scotti 3 , E. Anghileri 1 , L. Cuppini 1 , E. Prodi 3 , G. Finocchiaro 1 , M.G. Bruzzone 3 . 1 Neuro-Oncology, Fondazione IRCCS Istituto Neurologico C. Besta, Milan; 2 General Neurology, Fondazione IRCCS Istituto Neurologico Nazionale C. Mondino, Pavia; 3 NeuroRadiology, Fondazione IRCCS Istituto Neurologico C. Besta, Milan, Italy Background: Angiogenesis is a requirement for progression of glioblastoma (GBM) and Bevacizumab (Bev), an antibody directed to vascular endothelial growth factor (VEGF), was recently used to treat GBM. However in vivo modifications induced by treatment are under active investigation. Purpose/Objective: To analyze tumor changes induced by Irinotecan (Ir) and Bev, we use two different methodologies: relative CBV variation (rCBV) and Difference Perfusion Maps (DPMs). Materials and Methods: 42 recurrent GBM patients underwent Bev (10 mg/kg) and Ir (125 or 340 mg/m2 ) every 2 weeks and were followed up with a radiological protocol, including Dynamic Susceptibility Contrast MRI every 8 weeks. Radiological responses were assessed based on RANO criteria (Wen et al, 2012). Two methods were used to assess perfusion changes. In method A, relative CBV variation after 8 weeks of treatment was calculated through semi-automatic ROI placement in the same anatomic region as in baseline. In method B, relative CBV variations with respect to baseline values were calculated into the evolving tumor region by voxel-by-voxel difference. Poster Presentations DPMs were created showing where rCBV significantly increased, decreased or remained unchanged. Results: Method A showed a significant decrease of rCBV in patients with stable disease or partial response after 8 weeks of treatment (p=0.01) while patients with tumor progression maintained elevated levels of rCBV (p=0.38). Method B, based on DPMs, showed rCBV was increased in 35% (±16%) of tumor volume (8 week-increased blood volume, 8w-IBV) and was decreased in 17% (±14%) (8 week-decreased blood volume, 8w-DBV). Patients presenting 8w-IBV higher than 18% (first quartile) showed a significantly longer PFS (p=0.045) and OS (p=0.016). Using DPM we observed that early increase in global perfusion is related to better survival. Conclusions: Further studies will be necessary to confirm the potential of DPMs in predicting response to antiangiogenic therapy. MC13-0073 MiR-183 and miR-494 as risk biomarkers for breast cancer development T. Macedo, A.F. Evangelista, A.F.L. Marino, R.A.C. Vieira, A.F. Longatto, V.A.O. Silva, D.O. Vidal, M.M.C. Marques. Molecular Oncology Research Center, Research and Teaching Institute, Barretos Cancer Hospital, Barretos, Brazil Background: MicroRNAs (miRNAs) are small non-coding molecules (∼22 nt) considered as new important biomarkers due to their stability and distribution in virtually all tissues. Recent studies have suggested that microRNAs (miRNAs) are involved in breast cancer initiation and progression by silencing the expression of their target genes. However, miRNAs deregulation during breast cancer development is poorly understood. Purpose/Objective: Our aim was to identify miRNA biomarkers for breast cancer development. Materials and Methods: We performed Agilent miRNA microarray (∼866 human miRNAs) from 64 non-metastatic vs. metastatic patients. All the microarray data analyses were done in R statistical environment. ROC and relative risk curve were applied to miRNA expression values identifying two important biomarkers, miR-183 and miR-494. To evaluate the role of these microRNAs, we performed in vitro functional assays in metastatic breast cancer cell lines. Overexpression and silencing interventions were used to evaluate the role of miRNAs miR-183 and miR-494 in MDA-MB-468, MDA-MB-231 and MCF7 breast carcinoma cell lines. Cellular behavior (proliferation, migration and invasion) was tested by XCelligence system. Finally, target prediction was done using nine different bioinformatics algorithms and the technics of real-time PCR and Western blot were applied to confirm the identified targets. Results: The initial miRNA screening in breast cancer patients pointed two important biomarkers for breast cancer development, miR-183 and miR-494. The expression of these miRNAs were confirmed by real-time PCR. Further functional assays confirmed their role in breast cancer, especially by their influence in cell behavior (proliferation, invasion and migration). It was identified several targets by bioinformatics tools and we confirmed the regulation of targets of importance, such as PTEN and specific matrix metalloproteinases. Conclusions: We identified two miRNAs, miR-183 and miR-494, as important miRNA biomarkers for breast cancer development. Financial support: FAPESP. MC13-0074 Correlation between p53 immunohistochemistry and mutational status in human melanoma tumors M. Palmer, K. Horrigan, Y. Wang, Y. Yu, D.A. Sirko-Osadsa, B.H. Lee. OTM, Novartis Institutes for BioMedical Research Inc., Cambridge, USA Background: Oncogenic p53 alterations can affect the efficacy of novel therapeutic molecules that target genes including HDM2 which negatively regulate normal p53 function. Therefore, proper characterization of p53 status in tumor samples is important for appropriate patient selection and clinical trial enrollment for these novel therapies. Purpose/Objective: This study examines the utility of an IHC H-score for p53 for identifying TP53 mutations in human melanoma samples. By direct comparison of these technologies we hope to determine the fraction of mutation positive melanoma that can be detected by IHC. Materials and Methods: Here we characterized a set of human primary melanoma tumors using both immunohistochemistry (IHC) for p53 protein expression and full gene sequencing. An IHC H-score, calculated for each tumor, was correlated to mutational status. Poster Presentations Results: Tumors with p53 mutations were found to have H-scores of 120 or greater, except for truncating p53 mutations which had H-scores of 0. Tumors with no detectable p53 mutations had H-scores ranging from 0 to 120. Conclusions: The findings support that segregation of p53 mutant versus wild-type melanomas is possible by p53 IHC, although careful consideration must be made for tumors that harbor p53 truncating mutations, which may be completely negative for p53 expression by IHC. MC13-0076 CHFR silencing and microsatellite instability as predictors of sensitivity to docetaxel and gemcitabine in colorectal cancer L. Pelosof 1 , S. Yerram 2 , N. Ahuja 1 , A. Delmas 2 , L. Danilova 3 , J. Herman 1 , N. Azad 4 . 1 Cancer Biology Program, 2 Gastrointestinal Cancer Program, 3 Bioinformatics, 4 Gastrointestinal Cancer Program and Chemical Therapeutics Program, Johns Hopkins University, Baltimore, USA Background: It is increasingly recognized that even within a particular tumor type there is heterogeneity and that epigenetic differences likely contribute to the varying responses to standard treatment observed among patients. By combining tumors’ epigenetic alterations with existing knowledge of drugs’ mechanisms, DNA methylation alterations can serve as predictive biomarkers for the efficacy of particular chemotherapeutic agents which can direct patients’ treatment. CHFR methylation is associated with taxane sensitivity in multiple tumor types while microsatellite instability (MSI) as a predictive marker for therapeutic effect has had conflicting results though it is possibly associated with gemcitabine sensitivity. Purpose/Objective: To examine CHFR methylation and MSI as predictors of chemotherapy sensitivity to docetaxel and gemcitabine in colorectal cancer (CRC). Materials and Methods: Five MSI cell lines and five MSS cell lines were assessed for their CHFR methylation status and CHFR mRNA expression levels. Growth curves were used to assess these lines’ sensitivity or resistance to increasing concentrations of docetaxal and gemcitabine. To assess if CHFR re-expression results in decreased sensitivity to docetaxel, multiple cell lines were treated with the demethylating agent 5-azacitidine and then exposed to increasing concentrations of docetaxel. In vivo treatment of human xenografts of RKO, CACO2, and COLO205 were also examined for their sensitivity to gemcitabine, docetaxel, and the combination. Results: We observed increased sensitivity to gemcitabine in cell lines with MSI and to docetaxel in cell lines with CHFR inactivation via DNA methylation. Cells lines with silenced CHFR that demonstrated CHFR re-expression with 5-azacitidine treatment showed decreased sensitivity to docetaxel. Cell lines with silenced CHFR that did not have significant CHFR re-expression showed no change in sensitivity to docetaxel. In the xenograft model, the MSI-High/CHFR -methylated line RKO had tumor growth inhibition to each agent, and at least additive tumor growth inhibition with combination therapy. The MSS-CHFR-unmethylated line, CACO2, was resistant to single and combination therapy. Conclusions: CHFR methylation in CRC cell lines predicted for sensitivity in vitro and in vivo to docetaxel, while MSI-High cell lines were more sensitive to gemcitabine. These data suggest that a subset of CRC patients may be selectively sensitive to the combination of gemcitabine and docetaxel, and are the basis for an ongoing clinical trial of this combination in a biomarker-selected patient population. MC13-0077 Array CGH analysis of paired metastatic biopsies obtained pre-treatment and at resistance to FOLFOX-bevacizumab in metastatic CRC patients Z. Diaz 1 , E. Przybytkowski 2 , C. Lan 2 , S. McNamara 3 , A. Aguilar-Mahecha 2 , E. Camlioglu 4 , A. Gologan 5 , G. Batist 2 , M. Basik 6 . 1 Q-CROC, Jewish General Hospital, Montreal, Canada; 2 Department of Oncology, Jewish General Hospital, Montreal, Canada; 3 Translational Research Program, Quebec Clinical Research Organization in Cancer, Montreal, Canada; 4 Department of Radiology, Jewish General Hospital, Montreal, Canada; 5 Department of Pathology, Jewish General Hospital, Montreal, Canada; 6 Department of Oncology and Surgery, Jewish General Hospital, Montreal, Canada Background: Biopsy-driven clinical trials are essential to develop personalized therapeutics, since many critical questions are best addressed in the tumor tissue being treated. Currently, first-line treatment regimen for S33 metastatic colorectal cancer (CRC) consists of either FOLFOX or FOLFIRI in combination with bevacizumab. However, there are no known biomarkers to predict which patients are likely to respond or be resistant to either treatment. Purpose/Objective: Our objective was to identify biomarkers of clinical resistance to FOLFOX-bevacizumab. To do so, we designed a prospective biopsy-driven study in patients with metastatic CRC (NCT00984048) receiving first-line treatment with FOLFOX-bevacizumab. Materials and Methods: Eligible patients had confirmed metastatic CRC, measurable disease, and consented to tumor biopsy (three needle-core biopsies (NCBs) of a non-resectable liver metastasis) before treatment and at resistance. This study was approved at several Canadian hospitals, and recruitment is still ongoing. Patient biopsy samples were analyzed for DNA copy number alterations using the 244K Agilent platform. Results: Thus far, sixty patients agreed to partake in this ongoing multicenter trial and to provide NCBs from liver metastases. Using standard operating procedures developed for this trial, we were able to both preserve morphology and obtain high-quality genomic material from biopsy tissue. Histology was verified on every biopsy. Of the patients recruited to the study, 3% withdrew consent before the first biopsy procedure, 3% of biopsies were non-neoplastic, 3% were neuroendocrine, and 5% were mixed neuroendocrine/adenocarcinoma. Not all patients agreed to, or were eligible for a biopsy at acquired resistance to treatment. Some patients remain on FOLFOX-bevacizumab and have not yet developed resistance. There were 10 paired biopsies that met our criteria (>60% neoplastic cells and <20% necrosis) for downstream molecular profiling. We have already demonstrated in a previous pilot study that the biopsies are suitable for DNA analysis (array comparative genomic hybridization, methylation profiling) and RNA analysis (gene expression profiling, splicing isoforms variants and micro RNA profiling). Here we report on array CGH analysis from 10 paired liver metastatic biopsies obtained pre-treatment and at resistance to FOLFOX-bevacizumab in metastatic colorectal cancer patients. Conclusions: We conclude that obtaining serial liver NCBs is challenging, but safe and feasible in metastatic CRC patients with unresectable liver disease. This study will provide insight on the relevance of metastatic tissue to identify genomic markers of clinical resistance. MC13-0078 PARP1 variants associated with response to Temozolomide in metastatic melanoma patients A. Sedgewick 1 , M. Romkes 2 , S. Buch 2 , L. Villaruz 2 , I. Abecassis 3 , M. Saul 3 , J.M. Kirkwood 3 , P.V. Benos 4 , H. Tawbi 3 . 1 Joint Carnegie Mellon University-University of Pittsburgh Program in Computational Biology, University of Pittsburgh, Pittsburgh, USA; 2 University of Pittsburgh Cancer Institute, University of Pittsburgh, Pittsburgh, USA; 3 University of Pittsburgh Cancer Institute – Melanoma Program, University of Pittsburgh, Pittsburgh, USA; 4 Department of Computational & Systems Biology, University of Pittsburgh, Pittsburgh, USA Background: Temozolomide (TMZ), an alkylating agent, is used in treatment of metastatic melanoma. Response rates from TMZ alone vary (<20%) and there are no known predictive biomarkers for response. Purpose/Objective: We aimed to identify candidate biomarkers (SNP, methylation and mRNA and miRNA expression) for response to TMZ in patients with metastatic melanoma. Materials and Methods: A selected SNP panel and whole-genome mRNA/miRNA expression, and methylation microarray data were collected from tissues from 69 patients with metastatic melanoma treated with TMZ. Patients were labeled as responder (n=22) or non-responder (n=47) based on radiographic assessments. Potential biomarkers were identified using a variety of methods including penalized regression, Mann Whitney U test, and pairwise Markov random fields. We used data from the NCI Developmental Therapeutics Program to find additional support for candidate SNP markers. Results: We identified SNPs in PARP1 (rs1805407) and PIK3R1 (rs251402) by multiple feature selection tests as markers for response to TMZ. There were no responders in the group of 21 patients with one or more variant PARP1 alleles (p=5.4e−5). We examined the NCI-60 cell lines and identified additional PARP1 SNPs that were in linkage disequilibrium with our candidate SNP. Cell lines that had the variant were more resistant to alkylating agents. Cyclophosphamide and Carmustine, were among the agents with the most differential sensitivity scores between groups with and without the variant (uncorrected p=0.014 and 0.021 respectively). Interestingly, this SNP was also associated with increased PARP inhibitor sensitivity. S34 Poster Presentations Conclusions: The role of PARP1 in chemotherapy resistance is well established and PARP inhibitors have been extensively utilized in clinical trials to improve chemotherapy response. Our findings suggest that the PARP1 SNP rs1805407 is linked to a potentially more active form of PARP1 that increases resistance to TMZ. This was supported by findings from the NCI-60 cell lines. We are validating those results in appropriate cell line models. The PIK3R1 SNP highlights the role of altered signaling pathways in melanoma and its functional role is being examined. The PARP1 SNP rs1805407 may serve as a biomarker for response to TMZ chemotherapy and potentially PARP inhibitor sensitivity in metastatic melanoma. MC13-0080 Somatic mutations of the EGFR, KRAS and BRAF genes: Heterogeneity in circulating epithalial tumor cells (CETC) as determined using the Cobas® Z 480 analyzer K. Pachmann, M. Pizon, D. Zimon, E.L. Stein. Transfusion Center, Transfusion Center, Bayreuth, Germany Background: Targeted therapies directed specifically against somatic mutations enhancing the activity of signalling pathways have been shown to improve outcome compared with cytotoxic chemotherapies in patients with advanced tumors carrying the respective mutations. Purpose/Objective: This requires tests to identify such mutations which are mainly performed on formalin fixed material from the primary tumor. However, such material is not always available and, even more importantly, properties of cells with metastatic potential may change during the course of disease. In contrast, using maintrac® , a nondissipative approach avoiding enrichment steps, CETC be detected and individually isolated in almost all patients with lung and colon cancer and melanoma at different times of disease and therefore can provide a liquid biopsy to monitor the course of disease. We, here, report on the successful analysis of such isolated cells for gene mutations in tumor driver genes EGFR, KRAS and BRAF. Materials and Methods: Blood from patients with non-small cell lung cancer, colon cancer and malignant melanoma was analyzed for cells positive for epithelial antigen (EpCAM) using the maintrac® approach, which avoids cell selection, and an image anlysis system or laser scanning cytometry for detection. Between 8 and 20 EpCAM positive cells from each patient were isolated individually using a semiautomated capillary approach and deposited one by one into micro cups. The DNA was subsequently amplified by whole genome amplification and assayed using either the cobas® EGFR Mutation Test, the cobas® KRAS Mutation Test or the cobas® BRAF V600 Mutation Test. Results: DNA could be amplified from all indiviually isolated cells. An EGFR mutation was detected in 12% of isolated tumor cells from a patient with non-small cell lung cancer, the KRAS Mutation was detectable in 28% of cells from a patient with colon cancer and the BRAF Mutation in 100% of evaluable cells from a patient with melanoma. Conclusions: Individually isolating epithelial tumor cells from the peripheral blood from patients with non-small cell lung cancer, colon cancer and melanoma allows not only detect driver mutations in circulating tumor cells but also to determine the frequency of mutated cells. This proves, that at least part of the CETC from the tumor. They can, in the future, be used as markers of response to the action of drugs and contribute insight into how resistance may be acquired. MC13-0081 Cell cycle regulatory proteins in pancreatic cancer Z. Simtniece 1 , I. Strumfa 1 , A. Abolins 1 , A. Vanags 2 , M. Pavars 2 , E. Vasko 1 , J. Gardovskis 2 . 1 Department of Pathology, Riga Stradins University, Riga, Latvia; 2 Department of Surgery, Riga Stradins University, Riga, Latvia Background: Pancreatic ductal carcinoma (PDAC) is one of the most aggressive cancers. Mutations or epigenetic alterations in the tumour suppressor or cell cycle regulatory genes readjust cellular homeostasis and ability of apoptosis (Lang et al., 1998; Angela et al., 2013). Detection of p53, p21, p27 and cyclin D1 are described with predicting value of radiosensitivity and chemosensitivity (Fu et al., 1998; Wang et al., 2012). Purpose/Objective: The aim of this study was to determine the affected cell cycle regulatory proteins and correlation of them with Ki-67 in PDAC. Materials and Methods: Sixty-six consecutive cases of surgically treated PDAC were evaluated retrospectively. The cases were characterised by patient’s age, tumour TNM, stage, histological grade (G1-3), resection margins (R0-1). Expression of Ki-67, p53, p21, p27 and cyclin D1 was detected by immunohistochemistry (IHC) in PDAC and benign pancreatic ducts (BPD). Descriptive statistics was performed by SPSS, version 20. The study was approved by Committee of Ethics. Results: In the whole group, the mean age was 63.5 years [95% confidence interval CI: 60.6–66.0]. The most frequent tumour characteristics were: size >2 cm: 93.9% [83.4–97.8]; T3: 93.9% [85.2–97.5]; N1: 68.8% [56.6–78.8]; stage IIA: 26.6% [17.3–38.5]; stage IIB: 64.1% [51.8–74.7]; G2: 58.5% [46.3–69.7]; R1: 56.5% [44.0–68.1]. The main IHC data are shown in Table 1. Correlation was found between Ki-67 and p21 (p=0.01), p53 and p21 (p=0.01), p53 and p27 (p=0.00), p21 and p27 (p=0.05) expression. Conclusions: Morphologically, pancreatic cancer is characterized by aggressive growth. The expression of the analysed cell cycle regulatory proteins p53, p21 and cyclin D1 is frequently affected in pancreatic carcinogenesis suggesting possible benefit from targeted treatment. The proliferation activity shows direct correlation with p21, but the mutual association between p53, p27 and p21 can involve more complex mechanisms. MC13-0082 Salivary gland cancer in two male BRCA gene mutation carriers P. Bossi 1 , C.B. Ripamonti 2 , S. Manoukian 2 , M. Colombo 2 , B. Peissel 2 , D. Zaffaroni 2 , L.D. Locati 1 , L. Licitra 1 , M.L. Carcangiu 3 , P. Radice 2 . 1 Head and Neck Medical Oncology, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 2 Department of Preventive and Predictive Medicine, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 3 Department of Pathology, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy Background: Germ-line mutations in BRCA genes predispose to breast and ovarian cancers in women. In males, BRCA1 and BRCA2 mutations carriers have a lifetime risk of breast cancer of 2% and 7% respectively and they also exhibit an increased risk for prostate cancers. However, involvement of BRCA mutations in other malignancies is still under investigation. No data are available about the correlation between BRCA mutations and increased risk of salivary gland cancers (SGC). Purpose/Objective: To assess SGC in a cohort of patients undergoing genetic cancer risk assessment. Materials and Methods: We reviewed all the cases evaluated in our Institution, for whom BRCA mutations assessment was available. Patients with malignant SGC were retrieved. Results: We evaluated 2260 families registered in our Institutional databases from 1996 to 2012; 637 families were BRCA mutation positive, whereas 1623 index cases tested negative for BRCA mutations. Within mutated families, we identified 801 carriers of BRCA1 mutation, 456 carriers of BRCA2 mutation and 2 carriers with mutations in both genes. The review revealed two males with SGC among the BRCA mutation carriers and no case in the non-mutated group. A 56 year-old BRCA1 mutation carrier developed an high-grade mucoepidermoid carcinoma of the parotid gland. He was treated with surgery and postoperative radiochemotherapy, and died 4 years later for distant metastases. A 66-year-old BRCA2 mutation carrier developed a submandibular salivary gland papillary cistoadenocarcinoma treated with surgery and radiation. At 68 years, a sovraclavear node recurrence appeared and was treated with radiotherapy, followed by parapharyngeal relapse with no treatment opportunities. At the same age, he developed a ductal breast cancer. Abstract MC13-0081 – Table 1. Expression of Ki-67, p53, p21, p27, cyclin D1 in PDAC and BPD Parameter Target Ki-67 p53 p21 p27 Cyclin D1 Positive cases (%) PDAC BPD PDAC BPD PDAC BPD 100 [94.3–100] 21.8 [12.9–34.4] 1–55 0–4 20.8 [17.7–24.4] 0.4 [0.2–0.7] 69.4 [56.9–79.4] 14.0 [7.3–25.4] 0–97 0-10 30.8 [22.8–39.1] 0.65 [0.2–1.2] 93.6 [84.5–97.4] 37.5 [25.9–50.7] 0–65 0–13 23.4 [18.9–27.9] 1.5 [0.8–2.4] 100 [94.3–100] 100 [94.3–100] 6–82 26–90 34.7 [30.6–39.2] 51.5 [47.3–55.6] 80.9 [69.5–88.7] 18.6 [10.8–30.4] 0–65 0–16 19.9 [16.1–23.7] 0.9 [0.3–1.6] Range of positive cells (%) Mean count of positive cells (%) Poster Presentations Loss of heterozygosity (LOH) in tumor DNA extracted from paraffin-embedded tissues of SGC was assessed by comparison with constitutional DNA from matched peripheral blood leukocytes, using polymorphic markers linked to BRCA1 and BRCA2 loci. We detected BRCA1 allele loss for the BRCA1 carrier and BRCA2 loss for the BRCA2 carrier. By sequencing analysis, we defined that the losses affected the wild-type alleles in both cases. This indicated somatic bi-allelic inactivation of BRCA genes, a well-documented mechanism of cancer progression in cancers from BRCA mutation carriers. Conclusions: BRCA1 and BRCA2 mutations carriers seem to have an increased risk of SGC in comparison with BRCA negative cases. Further confirmation is needed in larger cohorts. MC13-0084 Tumor-specific DNA methylation in high-risk prostate cancer K. Litovkin 1 , E. Lerut 2 , S. Joniau 3 , O. Geveart 4 , S. Isebaert 5 , M. Spahn 6 , B. Kneitz 7 , K. Haustermans 5 , A. Van Eynde 1 , M. Bollen 1 . 1 Cellular & Molecular Medicine, KU Leuven, Leuven, Belgium; 2 Pathology University Hospitals Leuven &, Department of Imaging and Pathology KU Leuven, Leuven, Belgium; 3 Urology University Hospitals Leuven &, Department of Development and Regeneration KU Leuven, Leuven, Belgium; 4 Stanford Center for Cancer Systems Biology, Stanford University School of Medicine, Stanford, USA; 5 Radiation Oncology University Hospitals Leuven &, Department of Oncology KU Leuven, Leuven, Belgium; 6 University Hospital Bern, Department of Urology, Bern, Switzerland; 7 Department of Urology and Paediatric Urology, University Hospital Würzburg, Würzburg, Germany S35 charide; however recent reports describe that paclitaxel could trigger the TLR4 signalling pathway. Purpose/Objective: Our aim was to evaluate influence of genes involved in the TLR4 pathway in the distant-recurrence free survival in breast cancer patients treated with neoadjuvant Taxane-Anthracycline chemotherapy. Materials and Methods: We retrieved normalized gene expression data from a public database (GSE25066) that includes 508 samples of breast cancer patients profiled with the U133A microarray and treated with neoadjuvant Taxane-Anthracycline chemotherapy. We choose randomly three cohorts contained in that dataset as discovery group (I-SPY, LBJ/IN/GEI and MDACC cohort1; n=244) and two cohorts as validation group (MDACC cohort2 and USO; n=264). We screened the influence of the levels of expression of the follow genes in the disease-free survival: TLR4, IRAK1, IRAK3, IRAK4, TRAM, TRIF, TRAF3, TRAF6, IRF3, IRF7, SARM, SOCS3, SOCS1, DUBA, SHP_2, PIN1, TAK1, TAB1, TAB2, NFKB1, NFKB2, IL-12P40, TNF, TTP, TAN, TRIM38, CYLD, USP4, A20, NLRX1, IKKA, IKKB, NEMO, TRIM5, SHP-1, SYK and CBL-B. Results: Patients were grouped in tertiles according to the level of gene expression for each cohort. In the discovery set, significant genes for DFS were IRAK1 (P=0.005); IRAK4 (P=0.033); TRAF3 (P=0.032); USP4 (P<0.001); A20 (P=0.042); TRIM5 (P=0.042) and TNF (P=0.016). TLR4 was not related to DFS. We assigned a score according to the level of expression of each gene and its influence in the survival, (0, 0.5 and 1 according to the tertile, where 1 was for the poorest prognostic group). An overall score was obtained by the sum of all scores and patients were stratified in three groups: Low-score (LS), intermediate-score (IS) and high score (HS). Background: About a quarter of non-metastatic prostate cancers (PCa) are classified as high-risk, based on preoperative prostate-specific antigen (PSA) levels above 20 ng/ml, a biopsy Gleason score of 8–10 and/or an advanced clinical stage (≥T3). Purpose/Objective: The aim of this study was to evaluate promoter CpG island hypermethylation in patients with high-risk PCa at five genes that have previously been reported to be hypermethylated in PCa. Materials and Methods: A two-step quantitative multiplex methylationspecific PCR (QM-MSP) was developed to assess methylation at the promoters of APC, CCDN2, GSTP1, PTGS2 and RARB. Formalin-fixed paraffin-embedded (FFPE) radical prostatectomy samples from 280 patients with high-risk PCa, classified in two Belgian and one German cohort, were analyzed. In addition, we examined non-cancerous FFPE prostate tissue from 42 patients and lymph-node metastases from 16 patients. Our analysis followed strategic steps: First, we assessed differential methylation patterns between BPH, high-risk PCa and lymph node metastasis. Second, we studied the associations between quantitative methylation and adverse pathological features (pathological stage, Gleason score, biochemical recurrence and clinical failure). Results: We found that the combined DNA-methylation pattern of four genes (APC, CCND2, GSTP1 and RARB loci) can distinguish high-risk PCa from non-cancerous tissue with a specificity of 98% and a sensitivity of 96–100%. Significant associations between RARB methylation and pathological stage, and between GSTP1 methylation and tumor volume were found in two cohorts. Univariate analysis revealed that the high methylation levels of RARB and GSTP1 correlated with a higher risk for biochemical recurrence. The hypermethylation of RARB was associated with a higher risk for clinical failure. Interestingly, an increased methylation level at lymph node metastasis, as compared to that of the primary tumor, was associated with a higher risk for clinical failure. Finally, methylation levels of the selected markers were highly correlated (Pearson correlation coefficients 0.45–0.82, P ≤0.001), hinting at a common underlying mechanism. Conclusions: DNA hypermethylation of APC, CCND2, GSTP1 and RARB detected PCa with high sensitivity and specificity. Hypermethylation of RARB and GSTP1 were associated with adverse pathological and clinical outcomes. An increased DNA methylation in lymph-node metastatic sites, relative to that of the primary tumor, was associated with a higher risk for clinical failure. MC13-0085 Genes involved in TLR4 signal transduction as prognostic factors for survival following Taxane-Anthracycline neoadjuvant chemotherapy for invasive breast cancer J. Pinto, P. Valdiviezo, A. Aguilar, C. Flores, R. Velazco, J. Suazo, C. Vallejos, H. Gómez. Division de Investigación, Oncosalud-Auna, Lima, Peru Background: TLR4 receptor is typically activated by bacterial lipopolysac- Figure 1 S36 5-years DFS rates for de discovery set were LS=89%; IS=64.3%; HS=48.4% (P=0.001) and for the validation set were LS=89.8%, IS=77.2%; HS=67.9% (P=0.007). In the validation set, patients with residual disease had 5-years DFS rates of LS=85.4%, IS=73.7%; HS=58.7% (P=0.004). Differences were also observed in the groups with clinical stages I/II (P=0.031) and clinical stages III (P=0.005), but not for PAM50 subtypes. Conclusions: Levels of expression of genes involved in TLR4 signal transduction plays an important role in response for Taxanes-Anthracyclines therapy. We were able to identify a subgroup of patients with low score of signalling genes who could benefit from taxane-based therapies. MC13-0086 Acquired resistance to Hedgehog inhibitor through smoothened mutation in basal cell carcinoma P. Bossi 1 , F. Perrone 2 , B. Cortelazzi 2 , L. Licitra 1 . 1 Head and Neck Medical Oncology, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 2 Experimental Molecular Pathology, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy Background: Inappropriate activation of hedgehog (Hh) pathway is involved in the pathogenesis of skin basal cell carcinoma (BCC). Loss of function of transmembrane receptor Patched (PTCH1), or gain of function of Smoothened (SMO ), Sonic Hh (SHH ) and GLI family transcription factors can be ligand-independent oncogenic drivers of this disease. Vismodegib is a small molecule inhibitor of the Hh pathway that binds and inhibits SMO, demonstrating activity in clinical trials against advanced BCC. Purpose/Objective: To evaluate mechanisms of acquired resistance to Vismodegib in a BCC case. Materials and Methods: We report a case of BCC having a dramatic response to Vismodegib treatment, further developing treatment resistance. Formalin-fixed specimens of pre-treatment primary tumor and recurrence arisen during vismodegib regimen were investigated for the mutational status of the serpentine receptor SMO by sequencing (3500DX Genetic Analyzer). Results: From April 2012, a 78 years old man with a large, ulcerated and bleeding BCC of the left scapula, not amenable to surgery or radiotherapy was treated with Vismodegib 150 mg/day. The patient experienced a rapid regression of disease, with a clinical complete remission after 6 months of treatment. However, 11 months after treatment start, 2 subcutaneous nodules within the previous tumor field appeared and were surgically removed, with histological diagnosis of BCC. In the recurrence specimen SMO sequencing revealed the presence of a G to C missense mutation at position 1697 leading to the change D473Y within the exon 8. This nucleotide substitution was not detected in the primary tumor specimen analysis. Conclusions: The D473Y mutation in SMO could represent a mechanism of acquired resistance to vismodegib in BCC. This finding is sustained by a similar report in medulloblastoma in which the D473H mutation has been demonstrated to confer SMO inhibitor resistance impairing the receptor’s ability to bind the drug without affecting the ability of SMO to transmit the Hh signal (Yauch et al. Science 2009, 326:572–574). Further elucidation of such mechanisms can help in developing therapeutic strategies able to overcome Hh inhibitors resistance. MC13-0087 Epigenetic assay detects early stage non-small cell lung cancer in sputum L. Van Neste 1 , F. Moreau 1 , M.J. Kelley 2 , S.A. Belinsky 3 , J. Bigley 4 , W. Van Criekinge 5 . 1 R&D, MDxHealth, Liège, Belgium; 2 School of Medicine, Duke University, Durham, USA; 3 Lovelace Respiratory Research Institute, Duke University, Albuquerque, USA; 4 Clinical Affairs, MDxHealth, Irvine, USA; 5 R&D, MDxHealth, Irvine, USA Background: Similar to all other major cancers, lung cancer incidence and mortality recently started to decline in the US. However, it is still the main cause of cancer-related death. Improved surgical techniques and combined therapies have increased the 1-year survival to ∼45%. When detected early, the 5-year survival rate in localized (stage I or II) disease is ∼50%. Unfortunately, only a minority of cancers are detected early, resulting in a 5-year survival of 17% overall. These survival rates emphasize the importance of early detection. Recently, a study showed that the use of spiral CT screening for lung cancer resulted in a 20% reduction in mortality, but a high false positive rate (96%) was observed. Poster Presentations Purpose/Objective: Complementing traditional screening methods with molecular triage markers has proven beneficial for diagnostic purposes. Epigenetic markers, i.e. aberrant DNA-methylation, are observed early during oncogenesis. Materials and Methods: Sputum of a test set of 92 patients, 40 cancer cases and 52 cancer-free controls, was evaluated as a non-invasive source to detect the presence of non-small cell lung cancer. The methylation status of 11 candidate markers (HOXA9, TAC1, DPSYL4, SOX17, HOXD1, CDO1, RASSF1, JAM3, SFRP2, GPNMB andGREM1) was tested using a quantitative methylation-specific PCR approach. Results: Cutoffs for single markers were optimized and performance was calculated. To demonstrate clinical utility, the minimal specificity was fixed at 90%. Four genes showed adequate sensitivity (HOXA9 : 53%, CDO1: 58%, RASSF1: 75% and SFRP2 : 55%) as single markers. Combinations of two markers were explored using all 11 genes in search of marker complementarity. RASSF1 was present in all marker panels, with HOXA9, TAC1, SOX17, HOXD1, CDO1, SFRP2 or GREM1 as second gene, reaching sensitivities of 85%. Twelve stage I or II lung cancers in this study were used for an early detection subset analysis. Single marker assays of four genes were able to sensitively identify such cases (HOXA9 : 67%, CDO1: 58%, RASSF1: 83% and SFRP2 : 67%). When RASSF1 was paired with HOXA9, SFRP2 or GREM1, all early stage cases were successfully identified. Conclusions: This feasibility study demonstrates that aberrant DNAmethylation can be specifically detected in sputum samples of cancer patients. Panels of RASSF1 with either one of HOXA9, SFRP2 and GREM1 are particularly promising for the early detection of lung cancer, offering therapeutic advantages and potentially better clinical outcomes. MC13-0088 Controls for quantification of total cell free DNA qPCR analyses in plasma N. Pallisgaard 1 , R. Fredslund Andersen 1 , K.L. Garm Spindler 2 , I. Brandslund 1 , A. Jakobsen 2 . 1 Clinical Biochemistry, Vejle Hospital, Vejle, Denmark; 2 Oncology, Vejle Hospital, Vejle, Denmark Background: Analyses and quantification of total cell free DNA (cfDNA) in plasma samples has potential as source for detection of cancer mutations and to monitor treatment response. Since the concentration of cfDNA may be very low a substantial fraction may easily be lost during the purification process, or contaminated with DNA from lysated lymphocytes could result in fasely increased levels. Hemolysis may occur during drawing as a result of the pressure drop from vein to collection tube, lysis due to prolonged storage times prior to plasma isolation or to profound pipetting when isolating the plasma supernatant. Consequently, including of controls for preanalytical pitfalls are of utmost importance. Purpose/Objective: Develop qPCR control reactions to qualify the analysis of cell free DNA in plasma. Materials and Methods: Plasma DNA from 70 healthy individuals was purified on a Maxwell purification robot and cfDNA from another 100 healthy individuals was purified on a Qiasymphony robot. A 182 bp fragment from the soya bean gene CPP1 was constructed from oligonucleotides and amplified by PCR. The CPP1 fragment was used as a spike in control and added the plasma samples prior to purification and subsequently measured in purified cfDNA by qPCR. To identify lymphocyte DNA a multiplex qPCR assay (PBC) detecting a substantial fraction of the unique immunoglobulin rearrangements in B-cell DNA was developed. A qPCR targeting the single copy gene beta-2-microglobulin (B2M) was used to measure the number of genomic alleles in the samples and in parallel the CPP1 spike in and lymphocyte control. Results: In normal blood samples the number of measure B-cell alleles by the PBC assay was 0.5% (standard deviation 0.44) of total number of alleles. The variation was ascribed to individual genetic background and infection history since 3 other IgH multiplex assays tested gave similar results. In plasma samples the average number of alleles was 4700 per mL (standard deviation 2400) in samples where Lymphocyte DNA was not detected. Based on the B-cell control we found that plasma samples contaminated with lymphocyte DNA ranged between 10 and 50%. Conclusions: We have developed two qPCR control reactions to qualify the analysis of cell free DNA in plasma samples which may identify contaminating lymphocyte DNA and/or the loss of sample DNA during the purification. Poster Presentations MC13-0089 Epigenome-wide discovery of ovarian and breast cancer specific DNA methylation markers H. Lempiainen 1 , D. Mertens 1 , A. Brandenburg 1 , M. Remmert 1 , J. Hayward 2 , A. Jones 2 , S. Anjum 2 , M. Widschwendter 2 , M. Flesch 1 , J. Hoefkens 1 , T. Rujan 1 , T. Wittenberger 1 . 1 Expressionist, Genedata, Basel, Switzerland; 2 Department of Women’s Cancer, University College London Elizabeth Garrett Anderson Institute for Women’s Health, London, United Kingdom Background: Breast and ovarian cancers pose huge and unsolved challenges to the medical profession. Breast cancer is the most common cancer in women in the EU: more than 332,000 women are diagnosed with breast cancer each year and a woman dies every 6 minutes from this disease. Ovarian cancer, whilst far less common than breast cancer, is often diagnosed when the disease is at an advanced stage and has spread to other areas of the body. More than 60% of ovarian cancer patients die within the first 5 years after diagnosis. Implementation of successful screening programs has dramatically reduced the number of women dying from cervical cancer. Similarly, the EU FP7 consortium EpiFemCare aims to reduce the number of women diagnosed with late stage breast or ovarian cancer by 50%, reduce the number of women who receive unnecessary long-term chemotherapy by 50%, and reduce the number of women dying from these cancers by 20%. Purpose/Objective: EpiFemCare will establish and clinically validate a series of blood tests based upon DNA methylation technology that will facilitate both early detection and prediction of therapeutic outcome. The project consists of three phases: (1) Epigenome-wide discovery of ovarian/breast cancer specific DNA methylation markers. (2) Development of serum based assays for cancer specific markers. (3) Validation of the serum test performance in thousands of serial samples from prospective clinical trials. Materials and Methods: In phase 1 Illumina Infinium Human Methylation450 BeadChip Array technology is used to assess the methylation status of ∼485,000 sites in cancer and control tissues. In parallel Reduced Representation Bisulfite Sequencing (RRBS) is used to identify & confirm cancer specific methylated circulating DNA in matching serum samples. Results: Using Genedata Expressionist® for Genomic Profiling, we have established an automated bioinformatics pipeline for the detection of cancer specific differentially methylated regions (DMRs) that are most likely to fulfill the strict specificity criteria of a serum based test. Conclusions: The most promising DMRs are taken forward for the serum based clinical assay development and validation. MC13-0090 Epigenetic silencing promotor gene of ESR1 in DNA circulating and not expression of estrogen receptor in tumors of patients with breast cancer J. Martinez-Galan 1 , J.R. Delgado 1 , R. Del Moral 2 , B. Torres-Torres 3 , S. Ríos 4 . 1 Medical Oncology, Hospital Universitario Virgen de las Nieves, Granada, Spain; 2 Radiotherapy Oncology, Hospital Universitario Virgen de las Nieves, Granada, Spain; 3 Molecular Biology, Centro de Investigaciones Biomédicas, Granada, Spain; 4 Molecular Biology, Departamento de Radiología y Medicina Física, Granada, Spain Background: Cell-free circulating DNA carries not only tumor-specific changes in its sequence but also distinctive epigenetic marks, namely DNA methylation in island CpG region promotor. Recent data suggest that benign diseases have very specific methylation patterns within cell-free circulating DNA, which are different from the pattern of a malignant tumor of the same organ. Purpose/Objective: To determine whether Estrogen Receptor 1 (ESR1)(+) and ESR1(−) status relates to epigenetic changes in breast cancer-related genes and to correlate with expression in tumor of estrogen receptor RE(+) and RE(−). Materials and Methods: We quantified methylation levels ERS1 gene in serum of 92 pts breast cancer. A PCR quantitative technique was used to analyze levels of methylation gene. We also examined and correlationed the expression of ER in tumors by immunohistochemistry. Results: Median age was 58 years (32–88); 69% were postmenopausal women. Nodal involvement (N0, 63%; N1, 30%; N2, 7%), tumor size (T1, 58%; T2, 35%; T3, 4%; T4, 4%) and grade (G1, 20%; G2, 37%; G3, 30%). The methylated ESR1 in serum was significantly associated with ESR1(−) in breast tumors >80% (p=0.0179). Methylation ESR1 was preferably associated with triple negative (80%) and HER2+ (60%) subtype. S37 Nevertheless unmethylation ESR1 was found more frequently in LA (71%) and LB (59%) phenotype. With a median follow up of 6 years, we found worse overall survival (OS) with more frequent ESR1 methylation gene (p>0.05), Luminal A; ESR1 methylation OS at 6 years 81% vs 93% when was ESR1 unmethylation. Luminal B; ESR1 methylation 86% SG at 6 years vs 92% in unmethylation ESR1. Triple negative; ESR1 methylation SG at 6 years 75% vs 80% in unmethylation ESR1. HER2; ESR1 methylation SG at 6 years was 66.7% vs 75% in unmethylation ESR1. Conclusions: Gene promoter region hypermethylation is a significant event in primary breast cancer. However, its impact on tumor progression and potential predictive implications remain relatively unknown. Our study identifies the presence of variations in global levels of methylation promoters ESR1 genes in breast cancer with different phenotype classes and shows that these differences have clinical significance. MC13-0091 MicroRNA in biofluids – Robust biomarkers for disease P. Mouritzen 1 , T. Blondal 1 , D. Andreasen 1 , M. Wrang Teilum 2 , A. Thomsen 1 , N. Tolstrup 3 . 1 Research & Development, Exiqon, Vedbaek, Denmark; 2 Services, Exiqon, Vedbaek, Denmark; 3 Bioinformatics, Exiqon, Vedbaek, Denmark Background: microRNAs constitute a class of small cellular RNAs (typically 19–23 nt) that function as post-transcriptional regulators of gene expression. Current estimates indicate that more than one third of the human cellular transcriptome is regulated by this small class of RNA (less than 2000 miRNA). The study of extracellular microRNAs and their potential as pathophysiological markers has greatly expanded in the last couple of years. microRNAs have been shown to be actively exported from tissues into the circulation through a variety of mechanisms including exosome and microvesicle transport, and complexing with RNA binding proteins or HDL. Purpose/Objective: The high relative stability of microRNAs in common clinical source materials (FFPE blocks, plasma, serum, urine, saliva, etc.) and the ability of microRNA expression profiles to accurately classify discrete tissue types and specific disease states have positioned microRNAs as promising new biomarkers for diagnostic application. Materials and Methods: We have applied our highly sensitive LNA™-based qPCR platform for detection of microRNAs, which has enabled microRNA profiling in biofluids where levels are extremely low. The platform uses a single RT reaction to conduct full miRNome profiling and allows high-throughput profiling of microRNAs without the need for pre-amplification. Results: Thousands of biofluid samples including serum/plasma and urine have been profiled to determine normal reference ranges for circulating microRNAs as well as to identify biomarkers of disease and toxicology. Extensive data qualification and analysis methods have been developed and are central parameters to secure high quality data from biofluids. Recently, we have compared microRNA profiles of different exosome isolation methods and evaluated the differences between these and standard profiles of whole plasma and serum. Conclusions: Robust microRNA biomarker discovery and validation in serum/plasma and other biofluids depends on a highly sensitive and reliable microRNA detection system as well as careful attention to sample preparation and quality. The use of advanced sample and data qualification methods improve data quality and will be central to the development of circulating microRNAs as diagnostic markers. MC13-0092 The National Cancer Institute (NCI) specimen resource locator J. Demchok, S. Taube, B. Fombonne, I. Lubensky. National Institute of Health, National Cancer Institute, Cancer Diagnosis Program Background: The development of molecular technologies will advance the identification of clinically useful biomarkers and development of diagnostic assays to guide diagnosis and treatment of cancer patients with defined molecular abnormalities. Discovery and validation of markers of cancer prevention, diagnosis and response to therapy requires access to quality clinical biospecimens ranging from relatively small case sets and tissue microarrays to clinically annotated samples from patients uniformly treated in multi-site, randomized clinical trials. The availability of diverse biospecimen resources that procure and distribute such specimens is critical. However, unless researchers can learn about a biospecimen resource, the value of the resource cannot be fully realized. In response to this need the NCI’s S38 Cancer Diagnosis Program has developed, and is expanding, a searchable database: the Specimen Resource Locator (SRL). The SRL’s goal is to make researchers aware of existing specimen resources, and to facilitate the use of available biospecimens. Methods and Materials: The SRL serves as a public comprehensive database of resources where investigators have access to thousands of specimens of different tumor types. Biobanks with specimens from a variety of anatomic sources and diagnoses are available on the SRL. Currently the SRL includes NCI supported cancer specimen repositories, and their url link and contact information. Future plans will include other NIH sponsored and U.S. repositories. Results: Investigators can search the database and retrieve a list of resources that can furnish the specimens for a fee as well as for collaboration. Searches can be performed by tumor type, specimen type, or tissue preservation Poster Presentations method. An advanced search can be performed by requesting annotation criteria and specimen number requirement. If no match of specimen to investigator is available, the SRL refers researchers to the Tissue Expediter (tissexp@mail.nih.gov). The Tissue Expediter is distinct from the SRL in that it is an individual who helps the investigator locate appropriate resources. Conclusion: The availability of specimens and data resources is crucial to increase our understanding of cancer biology and to translate important research discoveries to clinical application. The SRL helps resources to operate more efficiently by allowing researchers to search their database and retrieve a list of resources likely to meet their research needs. The SRL will reduce inappropriate requests to resources and will lead investigators directly to appropriate resource. By funding the SRL and making biospecimens readily available to investigators, the National Cancer Institute facilitates translational research. S39 Author Index Page numbers are followed by abstract number(s) in parentheses. A Abecassis, I., S33 (MC13-0078) Abolins, A., S34 (MC13-0081) Adang, E.M.M., S15 (MC13-0014) Adriaensens, P., S20 (MC13-0032) Agrawal, L., S25 (MC13-0047) Aguilar, A., S35 (MC13-0085) Aguilar-Mahecha, A., S33 (MC13-0077) Ahuja, N., S33 (MC13-0076) Alain, T., S7 (SP024) Alarcon, I., S26 (MC13-0051) Alpana, S., S18 (MC13-0025) Alpy, F., S18 (MC13-0023) Amadori, D., S31 (MC13-0067) Ancukiewicz, M., S29 (MC13-0062) Andersen, R.F., S11 (MC13-0026) Andreasen, D., S37 (MC13-0091) Anghileri, E., S21 (MC13-0033), S32 (MC13-0072) Anjum, S., S37 (MC13-0089) Appelt, A.L., S11 (MC13-0026) Aquino, D., S32 (MC13-0072) Arnaud, M.P., S25 (MC13-0048) Atac, F., S15 (MC13-0010) Azad, N., S33 (MC13-0076) B Banerji, U., S7 (SP025) Barillot, E., S1 (SP002) Basik, M., S33 (MC13-0077) Batchelor, T.T., S29 (MC13-0062) Bathen, T.F., S23 (MC13-0044) Batist, G., S33 (MC13-0077) Becker, K.F., S9 (SP033), S10 (SP036) Becker, R., S4 (SP013) Belcari, F., S31 (MC13-0067) Belenguer-Querol, L., S30 (MC13-0066) Belinsky, S.A., S36 (MC13-0087) Benos, P.V., S33 (MC13-0078) Berry, D., S2 (SP005) Bertolini, F., S21 (MC13-0033) Bharti, S., S31 (MC13-0068) Bie, L., S13 (MC13-0004, MC13-0005, MC13-0006) Bigley, J., S36 (MC13-0087) Blondal, T., S37 (MC13-0091) Boerman, O.C., S21 (MC13-0036) Bogaerts, J., S5 (SP016) Bogers, J.P., S13 (MC13-0001) Bollen, M., S35 (MC13-0084) Bossi, P., S34 (MC13-0082), S36 (MC13-0086) Boutros, P.C., S14 (MC13-0009) Brandenburg, A., S37 (MC13-0089) Brandslund, I., S36 (MC13-0088) Braun, S., S17 (MC13-0021, MC13-0022), S18 (MC13-0024) Bristow, R.G., S14 (MC13-0009) Brünner, N., S19 (MC13-0028) Bruzzone, M.G., S21 (MC13-0033), S32 (MC13-0072) Buch, S., S33 (MC13-0078) Buxton, M., S2 (SP005) C Calleri, A., S21 (MC13-0033) Camlioglu, E., S33 (MC13-0077) Campo, E., S30 (MC13-0065) Campone, M., S2 (SP006) Carcangiu, M.L., S34 (MC13-0082) Cardoso, F., S6 (SP022) Casey, M., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Cavenagh, M.M., S31 (MC13-0068) Chan, K., S23 (MC13-0039) Chao, S., S30 (MC13-0066) Chao, S.L., S19 (MC13-0029), S22 (MC13-0037) Chelsky, D., S8 (SP030) Chen, A., S3 (SP007) Chen, G., S23 (MC13-0040) Chenard, M.P., S18 (MC13-0023) Cheng, Y.Y., S20 (MC13-0031) Chernyaev, V., S21 (MC13-0034) Cheung, A., S23 (MC13-0039) Cheung, I., S6 (SP021) Cheung, N., S6 (SP021) Chippada Venkata, U., S32 (MC13-0070) Choe, G., S19 (MC13-0027) Choi, Y., S23 (MC13-0041) Christensen, I.J., S19 (MC13-0028) Ciniselli, C.M., S8 (SP031) Colombo, M., S34 (MC13-0082) Colomer, A., S26 (MC13-0051), S30 (MC13-0065) Colomo, L., S30 (MC13-0065) Condeelis, J., S16 (MC13-0019) Conley, B., S3 (SP007), S11 (MC13-0060), S31 (MC13-0068) Conley, B.A., S31 (MC13-0068) Corcoran, R.B., S12 (MC13-0075) Cornelis, A., S13 (MC13-0001) Cortelazzi, B., S36 (MC13-0086) Cuppini, L., S21 (MC13-0033), S32 (MC13-0072) D D’Cruz, A., S26 (MC13-0050) Dal Pra, A., S14 (MC13-0009) Danesi, R., S31 (MC13-0067) Danilova, L., S33 (MC13-0076) Darquennes, K., S20 (MC13-0032) Davidovic, R., S15 (MC13-0015) Davis, S., S2 (SP005) Davison, C., S17 (MC13-0022), S18 (MC13-0024) de Geus-Oei, L.F., S15 (MC13-0014) De Greve, J., S23 (MC13-0040) de Vries, E., S5 (SP018) de Wilt, J.H.W., S15 (MC13-0014) Debetancourt, D., S28 (MC13-0059) Del Moral, R., S37 (MC13-0090) Del Re, M., S31 (MC13-0067) Deleu, M., S13 (MC13-0001) Delgado, J.R., S37 (MC13-0090) Delmas, A., S33 (MC13-0076) Delord, J.P., S2 (SP006) Delpous, S., S18 (MC13-0023) Demchok, J., S37 (MC13-0092) DeMichele, A., S2 (SP005) Dempsey, P.W., S17 (MC13-0020) Derome, A., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Di Stefano, A., S21 (MC13-0033), S32 (MC13-0072) Diaz, Z., S33 (MC13-0077) Dondi, E., S25 (MC13-0048) Donovan, M., S26 (MC13-0051), S30 (MC13-0065) Doroshow, J., S11 (MC13-0060) Doroshow, J.H., S3 (SP007) Douillard, J.Y., S17 (MC13-0021, MC13-0022), S18 (MC13-0024) Dowsett, M., S1 (SP003) Drisis, S., S19 (MC13-0029), S30 (MC13-0066) Duda, D.G., S29 (MC13-0062) E Eberhard, D., S3 (SP008) Emblem, K., S29 (MC13-0062) Endo, M., S12 (MC13-0079) Engel, K.B., S9 (SP035) Engelman, J.A., S12 (MC13-0075) Eoli, M., S21 (MC13-0033), S32 (MC13-0072) Erill, N., S26 (MC13-0051), S30 (MC13-0065) Esserman, L., S2 (SP005) Evangelista, A.F., S29 (MC13-0063), S32 (MC13-0073) F Fan, L.Q., S12 (MC13-0071) Fang, P., S31 (MC13-0069) Felisiak-Golabek, A., S24 (MC13-0045) Ferrer, I., S26 (MC13-0051) Finocchiaro, G., S21 (MC13-0033), S32 (MC13-0072) Flaherty, K.T., S12 (MC13-0075) Flamen, P., S30 (MC13-0066) Flesch, M., S37 (MC13-0089) Flores, C., S35 (MC13-0085) Fombonne, B., S37 (MC13-0092) Forbes, T., S11 (MC13-0060) Fredslund Andersen, R., S36 (MC13-0088) Freidlin, B., S11 (MC13-0049) G Galderisi, C., S31 (MC13-0069) Galibert, M.D., S25 (MC13-0048) Galon, J., S28 (MC13-0059) Galsky, M., S32 (MC13-0070) Gandemer, V., S25 (MC13-0048) Ganee, L., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Garcia, C., S30 (MC13-0066) Gardovskis, J., S34 (MC13-0081) Garm Spindler, K.L., S36 (MC13-0088) Gauri, G., S18 (MC13-0025) Gavoille, C., S2 (SP006) Geers, C., S23 (MC13-0040) Gelb, A.B., S27 (MC13-0055) Gelmini, S., S8 (SP031) Gerstner, E.R., S29 (MC13-0062) Gertler, F., S16 (MC13-0019) Geveart, O., S35 (MC13-0084) Gigot, J.F., S28 (MC13-0059) Glass, A., S16 (MC13-0019) S40 Goldkorn, A., S17 (MC13-0020) Golfinopoulos, V., S6 (SP022) Gologan, A., S33 (MC13-0077) Gómez, H., S35 (MC13-0085) Goncalves, A., S2 (SP006) Gong, Y., S32 (MC13-0070) Gravanis, I., S4 (SP012) Groelz, D., S9 (SP033) Grose, R., S6 (SP023) Grunnet, M., S19 (MC13-0028) Guan, P., S28 (MC13-0057) Guendisch, S., S9 (SP033) Guhathakurta, D., S12 (MC13-0071) Guiot, T., S30 (MC13-0066) Guozhen, Z., S14 (MC13-0008) H Haber, D.A., S12 (MC13-0075) Haicheur, N., S28 (MC13-0059) Hall, J., S6 (SP022) Halytskiy, V., S28 (MC13-0058) Hamoir, M., S29 (MC13-0064) Hariharan, P., S28 (MC13-0057) Harrington, R., S11 (MC13-0060) Hartmann, C.C., S8 (SP031) Hartmut, J., S9 (SP032) Haukaas, T.H., S23 (MC13-0044) Haustermans, K., S35 (MC13-0084) Have, C.L., S14 (MC13-0009) Hayward, J., S37 (MC13-0089) Hegi, M., S8 (SP028) Heinrich, M., S10 (SP036) Herman, J., S33 (MC13-0076) Heskamp, S., S21 (MC13-0036) Hidalgo, M., S7 (SP026) Hoefkens, J., S37 (MC13-0089) Hong, X., S13 (MC13-0006) Horrigan, K., S32 (MC13-0074) Hoyer-Hansen, G., S19 (MC13-0028) Huang, E., S11 (MC13-0049) Huang, L., S27 (MC13-0055) Hupé, P., S1 (SP002) Hylton, N., S2 (SP005) I Ignatiadis, M., S19 (MC13-0029) Imtiyaz, A., S18 (MC13-0025) Isaka, M., S12 (MC13-0079) Isambert, N., S2 (SP006) Isebaert, S., S35 (MC13-0084) Ishkanian, A.S., S14 (MC13-0009) J Jacinthe, B., S25 (MC13-0048) Jadin, L., S27 (MC13-0055) Jain, R.K., S29 (MC13-0062) Jakobsen, A., S11 (MC13-0026), S36 (MC13-0088) Jamsheed, J., S18 (MC13-0025) Janssens, J., S13 (MC13-0001) Jeffrey, S., S5 (SP017) Jensen, L.H., S19 (MC13-0028) Jiang, P., S27 (MC13-0055) Jiang, T.A.O., S14 (MC13-0008) Jones, A., S37 (MC13-0089) Jones, J., S16 (MC13-0019) Joniau, S., S35 (MC13-0084) Jurisica, I., S14 (MC13-0009) K Kakkar, N., S16 (MC13-0017) Kamal, M., S1 (SP002), S2 (SP006) Kane, S., S26 (MC13-0050) Kao, S.C., S20 (MC13-0031) Author Index Kap, M., S9 (SP033) Kelley, M.J., S36 (MC13-0087) Kenmotsu, H., S12 (MC13-0079) Kertesz, N., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Khajuria, R., S16 (MC13-0017) Khambata-Ford, S., S3 (SP010) Khurani, N., S18 (MC13-0025) Kim, M.A., S19 (MC13-0027) Kim, S., S23 (MC13-0041) Kim, W.H., S19 (MC13-0027) Kim, Y., S23 (MC13-0041) Kirkwood, J.M., S33 (MC13-0078) Kirschner, M.B., S20 (MC13-0031) Kneitz, B., S35 (MC13-0084) Kochukoshy, T., S32 (MC13-0070) Koh, Y., S12 (MC13-0079) Kovacevic, R., S15 (MC13-0015) Kramer, K., S6 (SP021) Krieger, L., S10 (SP036) Kristof, J., S31 (MC13-0069) Kruhoffer, M., S9 (SP033) Kubista, M., S8 (SP031), S10 (SP036) Kuk, D., S6 (SP021) Kummar, S., S3 (SP007), S11 (MC13-0060) Kupryjanczyk, J., S24 (MC13-0045) Kushner, B., S6 (SP021) L Lacombe, D., S6 (SP022) Lahuerta, J.J., S14 (MC13-0007) Lalonde, E., S14 (MC13-0009) Lam, W.L., S14 (MC13-0009) Lamba Saini, M., S29 (MC13-0064) Lamichhane, S., S23 (MC13-0044) Lamote, K., S25 (MC13-0046) Lan, C., S33 (MC13-0077) Lassen, U., S19 (MC13-0028) Le Tourneau, C., S2 (SP006) Lecumberri, R., S14 (MC13-0007) Lee, B.H., S32 (MC13-0074) Lee, H., S20 (MC13-0030) Lee, H.S., S19 (MC13-0027) Lee, K.H., S20 (MC13-0030) Leenders, W., S21 (MC13-0035) Lemort, M., S19 (MC13-0029), S30 (MC13-0066) Lempiainen, H., S37 (MC13-0089) Leroy, C., S25 (MC13-0048) Lerut, E., S35 (MC13-0084) Lessinger, J.M., S18 (MC13-0023) LeTourneau, C., S1 (SP002) Li, B., S23 (MC13-0039) Li, J., S31 (MC13-0069) Li, M., S13 (MC13-0004) Li, Y., S13 (MC13-0005, MC13-0006) Licitra, L., S34 (MC13-0082), S36 (MC13-0086) Lih, C., S11 (MC13-0060), S31 (MC13-0068) Lih, J., S3 (SP007) Lin, H., S16 (MC13-0019) Lin, S., S22 (MC13-0038) Lin, Y., S14 (MC13-0008) Litovkin, K., S35 (MC13-0084) Liu, H., S22 (MC13-0038) Liu, S.V., S17 (MC13-0020) Lively, T.G., S31 (MC13-0068) Locati, L.D., S34 (MC13-0082) Lockhart, N., S28 (MC13-0057) Longatto, A.F., S32 (MC13-0073) Loueslati, B.Y., S15 (MC13-0010) Louis, E., S20 (MC13-0032) Lu, X., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Lubensky, I., S37 (MC13-0092) Lujic, N., S15 (MC13-0015) Lund, I.K., S19 (MC13-0028) Lyandres, J., S2 (SP005) Lydolph, M., S19 (MC13-0028) M Macedo, T., S29 (MC13-0063), S32 (MC13-0073) Machiels, J.P., S28 (MC13-0059) Maheswaran, S., S12 (MC13-0075) Mak, D.F.Y., S14 (MC13-0009) Malentacchi, F., S8 (SP031) Malgundkar, S., S26 (MC13-0050) Malloff, C., S14 (MC13-0009) Mancuso, P., S21 (MC13-0033) Manoukian, S., S34 (MC13-0082) Marbaix, E., S29 (MC13-0064) Marino, A.F.L., S32 (MC13-0073) Mariyam, Z., S18 (MC13-0025) Marliot, F., S28 (MC13-0059) Marques, M.M.C., S29 (MC13-0063), S32 (MC13-0073) Martens, J., S4 (SP011) Martin, A., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Martinez-Galan, J., S37 (MC13-0090) Martínez-López, J., S14 (MC13-0007) Mateos, M.V., S14 (MC13-0007) Mathelin, C., S18 (MC13-0023) Matveev, V.B., S21 (MC13-0034) McCaughan, B.C., S20 (MC13-0031) McDermott, U., S3 (SP009) McGregor, P., S11 (MC13-0060) McLean, J., S28 (MC13-0057) McNamara, S., S33 (MC13-0077) Mehaffey, M., S11 (MC13-0060) Meng, A., S14 (MC13-0009) Mertens, D., S37 (MC13-0089) Mesotten, L., S20 (MC13-0032) Meynier, F., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Mezlini, A.M.E.L., S15 (MC13-0010) Milosevic, M., S14 (MC13-0009) Mir, R., S18 (MC13-0025) Mitry, E., S2 (SP006) Mlecnik, B., S28 (MC13-0059) Modak, S., S6 (SP021) Moestue, S.A., S23 (MC13-0044) Molkenboer-Kuenen, J.D.M., S21 (MC13-0036) Moon, N., S14 (MC13-0009) Moore, H., S9 (SP035), S25 (MC13-0047) Moreau, F., S36 (MC13-0087) Mori, K., S12 (MC13-0079) Mourad, M., S29 (MC13-0064) Mourin, A., S28 (MC13-0059) Mouritzen, P., S37 (MC13-0091) N Nackaerts, K., S25 (MC13-0046) Nair, S., S26 (MC13-0050) Nakajima, T., S12 (MC13-0079) Nam, K., S19 (MC13-0027) Navis, A., S21 (MC13-0035) Nazarian, R.M., S12 (MC13-0075) Netea-Maier, R.T., S15 (MC13-0014) Nikiforova, Z.N., S21 (MC13-0034) Noeparast, A., S23 (MC13-0040) O O’Donnell, L., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Oelmueller, U., S10 (SP036) Oh, W., S32 (MC13-0070) Ohde, Y., S12 (MC13-0079) Öhrling, K., S17 (MC13-0022) Oktay, M., S16 (MC13-0019) Author Index Oliner, K., S7 (SP027) Oliveira, R.J.S., S29 (MC13-0063) Ostrovnaya, I., S6 (SP021) Oyen, W.J.G., S15 (MC13-0014), S21 (MC13-0036) P Pachmann, K., S34 (MC13-0080) Pagès, F., S28 (MC13-0059) Pal, D., S16 (MC13-0017) Pallisgaard, N., S11 (MC13-0026), S36 (MC13-0088) Palmer, M., S32 (MC13-0074) Palmer, M.R., S31 (MC13-0069) Paoletti, X., S1 (SP001), S2 (SP006) Passardi, A., S31 (MC13-0067) Patil, A., S26 (MC13-0050) Pavars, M., S34 (MC13-0081) Pazzagli, M., S8 (SP031), S10 (SP036) Peeters, M., S17 (MC13-0021, MC13-0022), S18 (MC13-0024) Peissel, B., S34 (MC13-0082) Pelak, K., S31 (MC13-0069) Pellegatta, S., S21 (MC13-0033) Pelosof, L., S33 (MC13-0076) Perlmutter, J., S2 (SP005) Perrone, F., S36 (MC13-0086) Petersen-Baltussen, H., S21 (MC13-0035) Pineda-Lucena, A., S14 (MC13-0007) Pinho, M.C., S29 (MC13-0062) Pintilie, M., S14 (MC13-0009) Pinto, J., S35 (MC13-0085) Piris, A., S12 (MC13-0075) Pizon, M., S34 (MC13-0080) Pizzamiglio, S., S8 (SP031) Podgorska, A., S24 (MC13-0045) Polley, E., S11 (MC13-0049, MC13-0060) Polley, M., S11 (MC13-0049) Postow, M., S5 (SP019) Poyet-Gelas, F., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Prakash, R., S18 (MC13-0025) Prasad, R., S16 (MC13-0017) Prasant, Y., S18 (MC13-0025) Price, T., S17 (MC13-0021), S18 (MC13-0024) Prodi, E., S32 (MC13-0072) Prósper, F., S14 (MC13-0007) Przybytkowski, E., S33 (MC13-0077) Puchades-Carrasco, L., S14 (MC13-0007) Puig, P., S26 (MC13-0051), S30 (MC13-0065) Q Qi, L., S28 (MC13-0057) Qin, W.Q., S15 (MC13-0012) Quinn, D.I., S17 (MC13-0020) R Radice, P., S34 (MC13-0082) Rahier, J., S29 (MC13-0064) Ramnarine, V.R., S14 (MC13-0009) Reekmans, G., S20 (MC13-0032) Reese, E., S32 (MC13-0070) Reid, G., S20 (MC13-0031) Reix, N., S18 (MC13-0023) Remmert, M., S37 (MC13-0089) Renard, M., S23 (MC13-0040) Riegman, P., S10 (SP036) Riegman, P.H.J., S9 (SP033) Rio, M.C., S18 (MC13-0023) Ríos, S., S37 (MC13-0090) Ripamonti, C.B., S34 (MC13-0082) Ristic, D., S15 (MC13-0015) Robert, G., S25 (MC13-0048) Robinson, B., S16 (MC13-0019) S41 Rodon, J., S2 (SP004) Rohan, J., S31 (MC13-0068) Rohan, T., S16 (MC13-0019) Romkes, M., S33 (MC13-0078) Rong, A., S17 (MC13-0021) Rosen, B.R., S29 (MC13-0062) Rothenberg, S.M., S12 (MC13-0075) Rubinstein, L., S3 (SP007) Rujan, T., S37 (MC13-0089) S Sadhana, K., S26 (MC13-0050) Sagar, D., S18 (MC13-0025) Salgado, R., S6 (SP022) San Miguel, J.F., S14 (MC13-0007) Saul, M., S33 (MC13-0078) Sauter, E., S15 (MC13-0012) Scotti, A., S32 (MC13-0072) Sedgewick, A., S33 (MC13-0078) Serizawa, M., S12 (MC13-0079) Servant, N., S1 (SP002) Settleman, J., S12 (MC13-0075) Shak, S., S4 (SP014) Sharma, U., S16 (MC13-0017) Shazia, F., S18 (MC13-0025) Sheikh, N.A., S12 (MC13-0071) Shepard, H.M., S27 (MC13-0055) Shevchenko, V.E., S21 (MC13-0034) Shive, C., S28 (MC13-0057) Sidhu, R., S17 (MC13-0021, MC13-0022), S18 (MC13-0024) Siena, S., S17 (MC13-0021, MC13-0022), S18 (MC13-0024) Silva, V.A.O., S32 (MC13-0073) Simon, R., S11 (MC13-0049, MC13-0060) Sims, D., S11 (MC13-0060) Simtniece, Z., S34 (MC13-0081) Singh, S.K., S16 (MC13-0017) Sirko-Osadsa, D.A., S32 (MC13-0074) Smit, J.W.A., S15 (MC13-0014) Smith, C.S., S15 (MC13-0012) Sopta, J., S15 (MC13-0015) Sorensen, A.G., S29 (MC13-0062) Sorensen, M., S19 (MC13-0028) Spahn, M., S35 (MC13-0084) Sparano, J., S16 (MC13-0019) Spindler, K., S11 (MC13-0026) Spittle, C., S31 (MC13-0069) Squire, J., S14 (MC13-0009) Staha, J., S31 (MC13-0069) Stathopoulos, K., S19 (MC13-0029) Stein, E.L., S34 (MC13-0080) Strauss, W., S17 (MC13-0020) Strumfa, I., S34 (MC13-0081) Suazo, J., S35 (MC13-0085) Sweet, W., S26 (MC13-0052), S27 (MC13-0053, MC13-0054) Sykes, J., S14 (MC13-0009) Symmans, F., S9 (SP034) Symmons, W.F., S2 (SP005) Szafron, L., S24 (MC13-0045) Thoms, J., S14 (MC13-0009) Thomsen, A., S37 (MC13-0091) Tolstrup, N., S37 (MC13-0091) Tomasetto, C., S18 (MC13-0023) Tops, B., S21 (MC13-0035) Torres-Torres, B., S37 (MC13-0090) Trager, J.B., S12 (MC13-0071) Tredan, O., S2 (SP006) Tribouley, C., S31 (MC13-0069) Triche, T.J., S17 (MC13-0020) Troadec, M.B., S25 (MC13-0048) Tsai, H.L., S16 (MC13-0018) Tsao, C., S32 (MC13-0070) Tsao, G., S23 (MC13-0039) Turano, P., S9 (SP033), S10 (SP036) U Ulivi, P., S31 (MC13-0067) Ulloa Montoya, F., S5 (SP020) Um, K.S., S28 (MC13-0057) Umelo, I., S23 (MC13-0040) V Valdiviezo, P., S35 (MC13-0085) Vallee, A., S25 (MC13-0048) Vallejos, C., S35 (MC13-0085) Van Cleemput, J., S25 (MC13-0046) Van Criekinge, W., S36 (MC13-0087) Van den Eynde, M., S28 (MC13-0059) van der Graaf, W.T.A., S21 (MC13-0036) van der Kwast, T., S14 (MC13-0009) Van Eynde, A., S35 (MC13-0084) van Kruchten, M., S5 (SP018) van Laarhoven, H.W.M., S21 (MC13-0036) van Meerbeeck, J.P., S25 (MC13-0046) Van Neste, L., S36 (MC13-0087) Van Raaij, A., S21 (MC13-0035) van Zandwijk, N., S20 (MC13-0031) Vanags, A., S34 (MC13-0081) Vandeurzen, K., S20 (MC13-0032) Vanhove, K., S20 (MC13-0032) Vansteenkiste, J., S23 (MC13-0040) Vant’Veer, L., S2 (SP005) Varin-Blank, N., S25 (MC13-0048) Vasko, E., S34 (MC13-0081) Vaught, J., S25 (MC13-0047) Velazco, R., S35 (MC13-0085) Verderio, P., S8 (SP031) Verdi, H., S15 (MC13-0010) Verjans, M., S13 (MC13-0001) Verrijp, K., S21 (MC13-0035) Vettukattil, R., S23 (MC13-0044) Vidal, D.O., S32 (MC13-0073) Vidal, N., S26 (MC13-0051) Vieira, R.A.C., S32 (MC13-0073) Viertler, C., S9 (SP033), S10 (SP036) Villaruz, L., S33 (MC13-0078) Volkova, M.I., S21 (MC13-0034) Vriens, D., S15 (MC13-0014) Vu, T., S12 (MC13-0071) W T Tabernero, J., S17 (MC13-0021, MC13-0022), S18 (MC13-0024) Tabor, D., S28 (MC13-0057) Takahashi, T., S12 (MC13-0079) Taube, S., S37 (MC13-0092) Tawbi, H., S33 (MC13-0078) Tejpar, S., S6 (SP022) Ter Laan, M., S21 (MC13-0035) Teugels, E., S23 (MC13-0040) Thomas, G., S8 (SP029) Thomeer, M., S20 (MC13-0032) Walsh, W., S11 (MC13-0060) Wang, J.Y., S16 (MC13-0018) Wang, L., S32 (MC13-0070) Wang, Y., S14 (MC13-0008), S32 (MC13-0074) Wargo, J.A., S12 (MC13-0075) Wei, G., S27 (MC13-0055) Weisbuch, S., S8 (SP031) Welsh, A., S18 (MC13-0023) Wen, P.Y., S29 (MC13-0062) Wendling, C., S18 (MC13-0023) Wesseling, P., S21 (MC13-0035) Weynand, B., S29 (MC13-0064) S42 Widschwendter, M., S37 (MC13-0089) Williams, M., S3 (SP007) Williams, P.M., S11 (MC13-0060), S31 (MC13-0068) Wittenberger, T., S37 (MC13-0089) Wrang Teilum, M., S37 (MC13-0091) Wu, S., S28 (MC13-0057) Wyrich, R., S8 (SP031), S9 (SP033) Author Index Y Yamamoto, N., S12 (MC13-0079) Yan, Z., S31 (MC13-0069) Yazici, A., S15 (MC13-0010) Yee, D., S2 (SP005) Yee, T., S32 (MC13-0070) Yerram, S., S33 (MC13-0076) Yilmaz-Yalcin, Y., S15 (MC13-0010) Yu, Y., S32 (MC13-0074) X Xu, T., S17 (MC13-0020) Xu, Y., S17 (MC13-0020) Xue, X., S16 (MC13-0019) Z Zafarana, G., S14 (MC13-0009) Zaffaroni, D., S34 (MC13-0082) Zatloukal, K., S9 (SP033), S10 (SP036) Zhang, J.I.N.G., S14 (MC13-0008) Zhang, X., S13 (MC13-0004) Zhao, Q., S27 (MC13-0055) Zhao, Y., S11 (MC13-0060) Zhu, J., S32 (MC13-0070) Zidi, S., S15 (MC13-0010) Zimon, D., S34 (MC13-0080) Zingde, S., S26 (MC13-0050) Zoli, W., S31 (MC13-0067) S43 Disclosures Last Name First Name Abst ID Category COI Description Abbruzzese James NA Faculty Yes Agrawal Alain Lokesh Tommy MC13-0047 SP024 No No Arnaud Banerji Marie-Pierre Udai MC13-0048 SP025 Becker Robert SP013 Belenguer-Querol Bie Bie Bie Bogaerts Laura Li Li Li Jan MC13-0066 MC13-0006 MC13-0005 MC13-0004 SP016 Bossi Bossi Boutros Carey Chao Chelsky Paolo Paolo Paul Lisa Shih-Li David MC13-0082 MC13-0086 MC13-0009 NA MC13-0037 SP030 Chernyaev Cheung Vitaly Nai-Kong MC13-0034 SP021 POSTER Faculty & Speakers Presentation POSTER Faculty & Speakers Presentation Faculty & Speakers Presentation POSTER POSTER POSTER POSTER Faculty & Speakers Presentation POSTER POSTER POSTER Faculty POSTER Faculty & Speakers Presentation POSTER Faculty & Speakers Presentation Conley Corcoran Dancey de Vries Barbara Ryan Janet Elisabeth MC13-0068 MC13-0075 NA SP018 Del Re Delpous Demchok DeMichele Marzia Stéphanie Joanne Angela MC13-0067 MC13-0023 MC13-0092 SP005 deSouza Diaz Dittrich Nandita Zuanel Christian NA MC13-0077 NA Donovan Donovan Dowsett Michael Michael Mitchell MC13-0051 MC13-0065 SP003 Drisis Eberhard Stylianos David MC13-0029 SP008 Emblem Eoli Eoli Esteva Evangelista Febbo Goldkorn Gorlia Gravanis Kyrre E. Marica Marica Francisco Adriane Philip Amir Thierry Iordanis MC13-0062 MC13-0033 MC13-0072 NA MC13-0063 NA MC13-0020 NA SP012 Grose Grunnet Guan Guhathakurta Richard Mie Ping Debraj SP023 MC13-0028 MC13-0057 MC13-0071 POSTER ORAL Faculty Faculty & Speakers Presentation POSTER POSTER POSTER Faculty & Speakers Presentation Faculty POSTER Faculty POSTER POSTER Faculty & Speakers Presentation POSTER Faculty & Speakers Presentation POSTER POSTER POSTER Faculty POSTER Faculty POSTER Faculty Faculty & Speakers Presentation Faculty POSTER POSTER ORAL No Yes Honoraria/Consultaion fees Merck Advro Biotech, Stock share holders Pharmacydics, Jennerx Research supports/grants from Novartis, AstraZeneca, Johnson & Johnson, Chugai and consultation fees from Novartis, Debiopharm, Chugai, Merck. No No No No No No No No No No No No information received. No Yes No No No Yes No No No Yes No No Yes Yes Yes Yes No Yes Memorial Sloan-Kettering Cancer Center (MSKCC) has a patent application on hu3F8 and 8H9, and Dr. Cheung was named as one of the inventors. MSKCC has licensed the patent on beta-glucan to Biotec Pharmacon and Dr. Cheung was named as one of the inventors. My institution received grants from Roche, Genentech and Novartis for imaging research. Research Funding: Genentech, Pfizer, Roche, Incyte, Wyeth, Millenium Advisory Board: Pfizer. I have a conflict of interest with several pharmaceutical companies, the drigs of which woill be mentioned during my chairmanship, predomantily in form of unrestricted grants dedicated to the research institite I am directing and also honoraria for consulting. I am a consultant to Althia Health, S.L. I am a consultant to Althia Health, S.L. Particpation in company: Astrazeneca. Grants/research supports: Astrazeneca, Roche, Pfizer and consultation fees from Nanostring, Roche, genetix. Receipt honoraria/consultation fees from Onyx, Biodesix, Flagship Biosciences. No No No No No No information received. No No No information received. No No No Yes Sponsorship: All authors are employees of Dendreon Corporation. S44 Disclosures Last Name First Name Abst ID Category COI Description Hall Halytskiy Hartmut Jacqueline Volodymyr Juhl NA MC13-0058 SP032 No No No Hegi Monika SP028 Heskamp Hidalgo Sandra Manuel MC13-0036 SP026 Hilsenbeck Ignatiadis Jadin Janssens Susen Michail Laurence Jaak NA NA MC13-0055 MC13-0001 Faculty POSTER Faculty & Speakers Presentation Faculty & Speakers Presentation POSTER Faculty & Speakers Presentation Faculty Faculty POSTER POSTER Jeffrey Stefanie SP017 Jones Joan MC13-0019 Kaplan Khambata-Ford Richard Shirin NA SP010 Kim Edward NA Kim Kirschner Koh Kummar Seung Michaela Yasuhiro Shivaani MC13-0041 MC13-0031 MC13-0079 SP007 Lacombe Denis SP022 Lamba Saini Lamote Le Tourneau Monika Kevin Christophe MC13-0064 MC13-0046 SP006 Lee Lempiainen LI Li Lih MC13-0030 MC13-0089 MC13-0039 MC13-0069 MC13-0060 Lin Lin Lin Lively LoRusso Loueslati Louis Macedo Martens Ha-young Harri Bin Jin Chih Jian (Jason) Frank Yi Shiu-Ru Tracy Patricia Yacoubi Evelyne Taciane John NA MC13-0008 MC13-0038 NA NA MC13-0010 MC13-0032 MC13-0073 SP011 Martinez-Galan McDermott Joaquina Ultan MC13-0090 SP009 Mir Monzon Moore Rashid Federico Helen MC13-0025 NA SP035 Mouritzen Nair Nam O’Donnell O’Donnell O’Donnell Oelmueller Peter Sudhir Kyung Han Lynellen Lynellen Lynellen Uwe MC13-0091 MC13-0050 MC13-0027 MC13-0052 MC13-0053 MC13-0054 SP036 Oh Oliner William Kelly MC13-0070 SP027 Pachmann Pal Pallisgaard Katharina Deeksha Niels MC13-0080 MC13-0017 MC13-0088 Yes No No No Yes Yes Yes Faculty & Speakers Presentation POSTER No Faculty Faculty & Speakers Presentation Faculty Yes Yes Yes Yes POSTER POSTER ORAL Faculty & Speakers Presentation Faculty & Speakers Presentation POSTER POSTER Faculty & Speakers Presentation POSTER POSTER POSTER POSTER ORAL No No No No Faculty POSTER POSTER Faculty Faculty POSTER POSTER POSTER Faculty & Speakers Presentation POSTER Faculty & Speakers Presentation POSTER Faculty Faculty & Speakers Presentation POSTER POSTER POSTER POSTER POSTER POSTER Faculty & Speakers Presentation POSTER Faculty & Speakers Presentation POSTER POSTER POSTER No No No No No No No No No Consultation fees from Veridex. I am an employee of Halozyme Therapeutics, Inc. I am medical advisor to pharmaceutical and medical device companies. My wife is CEO in a medical device company. Sponsorship: This study was conducted without any sponsorship from the medical device industry or from pharmaceutical industries. I sit on the Advisory Board of MetaStat. I own equity in the company. I am an inventor of intellectual property licensed by MetaStat. Sponsorship: MetaStat. Grant support from Astrazeneca. Employee of Novartis Pharmaceuticals corporation. Receipt grant/research supports from Eli Lilly, Celgen, Novartis. Honoraria/consultaion fees Boerhinger Ingelheim. No No No No No No No No No No No No No information received. No No No No Yes Yes Yes Yes No Yes No No No I work for BioMerieux, Inc. I work for BioMerieux, Inc. I work for BioMerieux, Inc. Employee of QIAGEN GMBH, management committee co-chair of the QIAGEN. I am an employee of Amgen, Inc. Disclosures S45 Last Name First Name Abst ID Category COI Description Palmer Paoletti Michael Xavier MC13-0074 SP001 Yes No Pazzagli Mario SP031 Peeters Marc MC13-0021 POSTER Faculty & Speakers Presentation Faculty & Speakers Presentation POSTER Peeters Marc MC13-0022 POSTER Yes Peeters Marc MC13-0024 POSTER Yes Pelosof Pinto Podgorska Polley Polley Polley Postow Lorraine Joseph Agnieszka Eric Mei Mei Michael MC13-0076 MC13-0085 MC13-0045 NA NA MC13-0049 SP019 No No No No No No Yes PuchadesCarrasco Rodon Leonor MC13-0007 POSTER POSTER POSTER Faculty Faculty ORAL Faculty & Speakers Presentation POSTER Jordi SP004 No Salgado Sauter Sedgewick Servant Roberto Edward Andrew Nicolas NA MC13-0012 MC13-0078 SP002 Shak Steven SP014 Sidransky Simtniece Sopta Spindler NA MC13-0081 MC13-0015 MC13-0026 Stadler David Zane Jelena Karen-Lise Garm Walter Faculty & Speakers Presentation Faculty POSTER POSTER Faculty & Speakers Presentation Faculty & Speakers Presentation Faculty POSTER POSTER ORAL NA Sweep Fred NA Employment and stock ownership. Sponsorship: Novartis. No Yes MP: consultant and participated in advisory boards for Amgen and has also received honoraria and research funding from Amgen, Merck Serono, Ipsen, Novartis, Roche, and Sanofi. JYD: Amgen: participation in Steering Committee, Advisory Board, Symposia, consultant Merck Serono: participation in Advisory Board, Symposia, consultant, research funding Roche: participation in Advisory Board, Symposia, consultant Boerhinger Ingelheim: participation in Advisory Board. SS: member of Advisory Boards for Amgen, AstraZeneca, Sanofi-Aventis, Bayer, Celgene, Health Genomics, Roche. TP: participation in Advisory boards for Amgen, Roche and Merck Serono. JT: advisory role: Amgen, Boehringer, Bristol-Myers Squibb, Genentech, Imclone, Lilly, Merck KGaA, Millennium, Novartis, Onyx, Pfizer, Roche, Sanofi, Celgene. Honoraria for presentations: Amgen, Merck KGaA, Novartis, Roche, Sanofi. RS: Amgen Inc. employee. SB: Amgen GmbH employee. AR: Amgen Inc. employee. MP: consultant and participated in advisory boards for Amgen and has also received honoraria and research funding from Amgen, Merck Serono, Ipsen, Novartis, Roche, and Sanofi. JYD: Amgen: participation in Steering Committee, Advisory Board, Symposia, consultant Merck Serono: participation in Advisory Board, Symposia, consultant, research funding Roche: participation in Advisory Board, Symposia, consultant Boerhinger Ingelheim: participation in Advisory Board. SS: member of Advisory Boards for Amgen, AstraZeneca, Sanofi-Aventis, Bayer, Celgene, Health Genomics, Roche. TP: Advisory boards for Amgen, Roche and Merck Serono. JT: advisory role: Amgen, Boehringer, Bristol-Myers Squibb, Genentech, Imclone, Lilly, Merck KGaA, Millennium, Novartis, Onyx, Pfizer, Roche, Sanofi, Celgene. Honoraria for presentations: Amgen, Merck KGaA, Novartis, Roche, Sanofi. RS: Amgen Inc. employee. SB: Amgen GmbH employee. CD: Amgen Ltd employee. MP: consultant and participated in advisory boards for Amgen and has also received honoraria and research funding from Amgen, Merck Serono, Ipsen, Novartis, Roche, and Sanofi. JYD: Amgen: participation in Steering Committee, Advisory Board, Symposia, consultant Merck Serono: participation in Advisory Board, Symposia, consultant, research funding Roche: participation in Advisory Board, Symposia, consultant Boerhinger Ingelheim: participation in Advisory Board. SS: member of Advisory Boards for Amgen, AstraZeneca, Sanofi-Aventis, Bayer, Celgene, Health Genomics, Roche. TP: participation in Advisory boards for Amgen, Roche and Merck Serono. JT: advisory role: Amgen, Boehringer, Bristol-Myers Squibb, Genentech, Imclone, Lilly, Merck KGaA, Millennium, Novartis, Onyx, Pfizer, Roche, Sanofi, Celgene. Honoraria for presentations: Amgen, Merck KGaA, Novartis, Roche, Sanofi. RS: Amgen Inc. employee. SB: Amgen GmbH employee. AR: Amgen Inc. employee. Bristol-Myers Squibb research funding. No Yes No No No Honoraria/Consultaion fees from Roche, Histogenex. Yes Genomic Health, Inc. employee and stockholder. Yes No No No Consultation fees and stock shareholders “Champions Oncology”. Faculty Yes Consultation fees and grant research support: Active Biotech, Bayer, Bristol-Myers-Squibb, Boerhinger-Ingelheim, Dendreon Exilixis, Novartis, Genentech (Roche), Glaxo-Smith-Kline, Medivation, Pfizer, ImClone (Lilly), Amgen, Takeda (Millenium). Faculty No S46 Disclosures Last Name First Name Abst ID Category Symmans Frazer SP034 Tejpar Thomas Sabine Geraldine NA SP029 Thurin Tops Ulloa Montoya Magdalena Bastiaan Fernando NA MC13-0035 SP020 Umelo Van den Eynde Van Eynde Van Neste Vettukattil MC13-0040 MC13-0059 MC13-0084 MC13-0087 MC13-0044 Vriens Ijeoma Marc Aleyde Leander Muhammad Riyas Dennis Faculty & Speakers Presentation Faculty Faculty & Speakers Presentation Faculty POSTER Faculty & Speakers Presentation POSTER POSTER POSTER POSTER POSTER COI Description MC13-0014 POSTER Yes Wang Welch Williams Zatloukal Jaw Yuan Jack Paul Kurt MC13-0018 NA NA SP033 POSTER Faculty Faculty Faculty & Speakers Presentation No No No No No information received. No Yes Consultation fees AstraZeneca No No Yes GSK Vaccines employee and stock ownership. No No Yes Yes No Patent application filed, but no financial interest. Employee of MDxHealth. No conflicts of interest. Sponsorship: This project was funded in part by the Netherlands Organisation for Health Research and Development (ZonMW). S47 Abstracts Index Speakers’ Presentations SP024 SP025 SP013 SP016 SP030 SP021 SP018 SP005 SP003 SP008 SP012 SP023 SP032 SP028 SP026 SP017 SP010 SP007 SP022 SP006 SP011 SP009 SP035 SP036 SP027 SP001 SP031 SP019 SP004 SP002 SP014 SP034 SP029 SP020 SP033 Alain, T., S7 Banerji, U., S7 Becker, R., S4 Bogaerts, J., S5 Chelsky, D., S8 Cheung, N., S6 de Vries, E., S5 DeMichele, A., S2 Dowsett, M., S1 Eberhard, D., S3 Gravanis, I., S4 Grose, R., S6 Hartmut, J., S9 Hegi, M., S8 Hidalgo, M., S7 Jeffrey, S., S5 Khambata-Ford, S., S3 Kummar, S., S3 Lacombe, D., S6 Le Tourneau, C., S2 Martens, J., S4 McDermott, U., S3 Moore, H., S9 Oelmueller, U., S10 Oliner, K., S7 Paoletti, X., S1 Pazzagli, M., S8 Postow, M., S5 Rodon, J., S2 Servant, N., S1 Shak, S., S4 Symmans, F., S9 Thomas, G., S8 Ulloa Montoya, F., S5 Zatloukal, K., S9 Six Best Poster Abstracts – Oral Presentations MC13-0075 MC13-0071 MC13-0079 MC13-0060 MC13-0049 MC13-0026 Corcoran, R.B., S12 Guhathakurta, D., S12 Koh, Y., S12 Lih, C., S11 Polley, M., S11 Spindler, K., S11 Poster Presentations MC13-0047 MC13-0048 MC13-0066 MC13-0004 MC13-0005 MC13-0006 MC13-0082 MC13-0086 MC13-0009 MC13-0037 MC13-0034 MC13-0068 MC13-0067 MC13-0023 MC13-0092 Agrawal, L., S25 Arnaud, M.P., S25 Belenguer-Querol, L., S30 Bie, L., S13 Bie, L., S13 Bie, L., S13 Bossi, P., S34 Bossi, P., S36 Boutros, P.C., S14 Chao, S.L., S22 Chernyaev, V., S21 Conley, B., S31 Del Re, M., S31 Delpous, S., S18 Demchok, J., S37 MC13-0077 MC13-0051 MC13-0065 MC13-0029 MC13-0062 MC13-0033 MC13-0033 MC13-0063 MC13-0020 MC13-0028 MC13-0057 MC13-0058 MC13-0036 MC13-0055 MC13-0001 MC13-0019 MC13-0041 MC13-0031 MC13-0064 MC13-0046 MC13-0030 MC13-0089 MC13-0039 MC13-0069 MC13-0038 MC13-0008 MC13-0010 MC13-0032 MC13-0073 MC13-0090 MC13-0025 MC13-0091 MC13-0050 MC13-0027 MC13-0052 MC13-0053 MC13-0054 MC13-0070 MC13-0080 MC13-0017 MC13-0088 MC13-0074 MC13-0021 MC13-0022 MC13-0024 MC13-0076 MC13-0085 MC13-0045 MC13-0007 MC13-0012 MC13-0078 MC13-0081 MC13-0015 MC13-0035 MC13-0040 MC13-0059 MC13-0084 MC13-0087 MC13-0044 MC13-0014 MC13-0018 Diaz, Z., S33 Donovan, M., S26 Donovan, M., S30 Drisis, S., S19 Emblem, K., S29 Eoli, M., S21 Eoli, M., S32 Evangelista, A.F., S29 Goldkorn, A., S17 Grunnet, M., S19 Guan, P., S28 Halytskiy, V., S28 Heskamp, S., S21 Jadin, L., S27 Janssens, J., S13 Jones, J., S16 Kim, S., S23 Kirschner, M.B., S20 Lamba Saini, M., S29 Lamote, K., S25 Lee, H., S20 Lempiainen, H., S37 Li, B., S23 Li, J., S31 Lin, S., S22 Lin, Y., S14 Loueslati, B.Y., S15 Louis, E., S20 Macedo, T., S32 Martinez-Galan, J., S37 Mir, R., S18 Mouritzen, P., S37 Nair, S., S26 Nam, K., S19 O’Donnell, L., S26 O’Donnell, L., S27 O’Donnell, L., S27 Oh, W., S32 Pachmann, K., S34 Pal, D., S16 Pallisgaard, N., S36 Palmer, M., S32 Peeters, M., S17 Peeters, M., S17 Peeters, M., S18 Pelosof, L., S33 Pinto, J., S35 Podgorska, A., S24 Puchades-Carrasco, L., S14 Sauter, E., S15 Sedgewick, A., S33 Simtniece, Z., S34 Sopta, J., S15 Tops, B., S21 Umelo, I., S23 Van den Eynde, M., S28 Van Eynde, A., S35 Van Neste, L., S36 Vettukattil, R., S23 Vriens, D., S15 Wang, J.Y., S16