S28 Poster Presentations 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 collect- ing 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, 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 pre- operative 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 mm 2 ) 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 Markers Survival HR Log-rank Months (IS0–2 vs 3–4; p-value (IS0–2 vs 3–4) 95% CI) Median of all CD3-CD8 DFS 1.2 (0.7–2.3) 0.48 OS 2.4 (0.7–7.3) 0.12 CD3-CD45RO DFS 1.5 (0.8–2.9) 0.16 OS 2.2 (0.8–5.9) 0.11 CD8-CD45RO DFS 1.0 (0.6–2.0) 0.87 OS 1.0 (0.4–2.9) 0.17 Least infiltrated CD3-CD8 DFS 1.8 (1.0–3.4) 0.05 8.0 vs 14.9 OS 8.8 (2.0–39.1) 0.0007 27.9 vs NR CD3-CD45RO DFS 2.5 (1.3–4.9) 0.004 8.0 vs 17.0 OS 4.0 (1.2–14.2) 0.01 31.8 vs NR CD8-CD45RO DFS 2.0 (1.0–3.7) 0.03 8.4 vs 16.0 OS 2.6 (0.9–7.6) 0.06 Most infiltrated CD3-CD8 DFS 1.0 (0.6–1.9) 0.93 OS 2.1 (0.7–6.6) 0.17 CD3-CD45RO DFS 1.7 (0.9–3.2) 0.12 OS 3.7 (0.8–16.2) 0.06 CD8-CD45RO DFS 1.6 (0.8–3.2) 0.16 OS 2.4 (0.7–8.4) 0.16