Citations

Abstract Background: Next generation sequencing (NGS) has revealed that the genetic profiles of individual tumors are highly heterogeneous due to their subclonal composition. Tumor heterogeneity has profound clinical implications affecting differences in treatment response and therapeutic resistance. Methods: An estrogen receptor positive, Her2 normal breast cancer patient underwent definitive surgical treatment with modified radical mastectomy. Multiple samples were obtained from the primary cancer and metastatic lymph nodes. Flow cytometry based methods were used to isolate and classify distinct cell subpopulations according to their ploidy, and samples were processed for whole exome sequencing (WES). Comparative Genomic Hybridization (CGH) methods were used to identify somatic unique copy-number alterations (CNAs). WES-Single Nucleotide Variations (SNVs) were used to infer subclonal phylogenetic relationships using Maximum Parsimony methods and annotated with variation-cluster-barcodes (vc-barcodes) generated using a variation Bayesian mixture model (VBMM) approach that focuses on copy-number neutral sequence variations for subclone identification (SciClone). Our comparative phylogenetic-VBMM clustering method was used to identify CNA and/or SNV drivers that are key in differentiating subclonal populations at multiple tumor sites. Results: CNA-neutral WES-SNVs from 25 subclonal populations (isolated from 4 distinct sites) clustered into 6 major groups and were used to generate vc-barcodes. A phylogenetic reconstruction of subclonal architecture annotated with vc-barcode information expedited identification of key variations underlying the differentiation of subclones at distinct tumor sites. For example, we identified an amplification event involving the Anaplastic Lymphoma Kinase (ALK) gene that was more frequent in primary biopsy subclones (12/17, ∼71%) in comparison to metastatic subclones (2/6, ∼34%). As the ALK amplification is lost, we find that most of the metastatic subclones also acquire a predicted damaging mutation in this oncogene (4/6, ∼67%). In addition, we identified a potential driver metastatic cell lineage that carries the ALK amplification in the absence of the nonsynonymous mutation. Conclusions: We developed a hybrid clustering methodology and used it to reconstruct the subclonal architecture in primary and metastatic tumors from a single breast cancer patient. Our methods have identified several CNAs and/or SNVs underlying subclonal differentiation, as well as potential biomarkers of disease progression and indicators of emerging resistance. Application of these methods to larger cohorts and types of tumors should be conducted to ascertain more precise estimates of the predictive accuracy of our processes. Citation Format: Mia D. Champion, Princy Francis, Barbara A. Pockaj, Michael T. Barrett. Hybrid clustering methodologies to distinguish CNAs and/or SNVs that drive subclonal differentiation in samples from a breast cancer patient primary tumor and metastatic lymph nodes. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4861. doi:10.1158/1538-7445.AM2015-4861

Abstract Epithelial ovarian cancer (EOC) is the fifth leading cause of cancer death among women in the United States (5% of cancer deaths). The standard treatment for patients with advanced EOC is initial debulking surgery followed by carboplatin-paclitaxel combination chemotherapy. Unfortunately, even with modern chemotherapy, most patients relapse and die with the five-year overall survival around 45%. In addition, those patients that initially respond to taxane-platinum therapy eventually develop platinum-resistant tumors and relapse. Thus, finding novel therapeutics for treating EOC is essential. One approach that has been used widely in cancer drug discovery is Connectivity Mapping (CMAP) using gene expression data and drug phenotype data assessed on cancer cell lines (CCL) (i.e. “sensitive” or “resistant” based on dose-response drug data). However, recent publications have highlighted some issues with the use of these standard CCL sets that are often used in these drug screening studies. Therefore, instead of CMAP analysis based on gene expression signatures developed on drug response data collected on a small set of EOC cell lines, we took a novel CMAP approach based on information collected on EOC patients and clinical endpoints. That is, we determined the gene expression signature (e.g., set of genes) that were associated with time to recurrence in data collected on EOC patients within TCGA (n = 518) and a Mayo Clinic study (n = 474). The set of genes associated with time to recurrence (with and without adjustment for additional clinical covariates) was completed separately for TCGA and Mayo Clinic studies, restricting all analyses to EOC tumor of high-grade serous histology. Each of these 4 sets of genes, where genes with hazard ratios (HR) > 1 coded as “positively” associated and genes with HR < 1 coded as “negatively” associated with the phenotype, were inputted into CMAP software (Broad Institute) to determine a set of drugs for which our signature “matches” the “reference” signature. The drugs that overlapped between the CMAP analyses on the two studies were carried forward for validation studies involving drug screens on a set of 10 EOC cell lines. Of the 11 drugs carried forward, we a priori hypothesized that 5 of these drugs would kill EOC cells (mitoxantrone, podophyllotoxin, wortmannin, doxorubicin, and 17-AAG), for which all 5 showed an ability to kill EOC cells in vitro (cell lines were treated using serial dilutions of the drugs (20-0 μM) for 72 h followed by measuring cell viability using the CellTiter-Blue assay). Further research is on-going to understand the functional mechanism by which these candidate drugs might treat EOC and to apply this bioinformatics approach to other cancer drug screening studies. Note: This abstract was not presented at the meeting. Citation Format: Brooke L. Fridley, Rama Raghavan, Gottfried E. Konecny, Chen Wang, Ellen L. Goode, Harsh B. Pathak, Stephen Hyter. A bioinformatics approach to drug discovery: patient based connectivity mapping. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4853. doi:10.1158/1538-7445.AM2015-4853

Abstract Introduction: Invasive lobular carcinoma (ILC) comprises approximately ∼10-20% of breast cancers (BC) and appears to have a distinct biology from invasive ductal carcinoma (IDC). ILC is known to have a higher incidence of multifocal (MF) disease when compared to IDC (Biglia N, EJSO, 2013) and in general, MF BC has been associated with an increased rate of regional lymph node metastases (Cabioglu N, JACS, 2009). Treatment decisions are most often defined by the largest of the MF tumors without targeting potentially relevant biologic heterogeneity of the disease. We conducted a genomic analysis of MF ILC to help address this knowledge void. Methods: We characterized gene expression in multiple foci from 11 patients (pts) with MF ILC (all ER+ and HER2-, mean number of foci = 2.62) and 13 benign breast tissues. RNA was extracted from 3×1.5mm cores from each foci and gene expression was measured in 730 known cancer genes using the NanoString PanCancer pathways panel. Differential gene expression analyses were performed between 1) MF tumors and benign tissue (BT) accounting for pairing of pts and 2) different foci of MF tumors from the same ILC pt by ANOVA. The association of clinical covariates (age, number of tumors, grade and stage) in the top genes that were differentially expressed between multiple foci was assessed by linear mixed models accounting for pt and tumor. Results: A total of 20 genes (including MET and EGFR, top 5 genes: SPRY2, SOX17, FIGF, PDGFA, IL6) were differentially expressed (p<0.05) between malignant foci from the same pt, after correction by false discovery rate (FDR). EGFR expression was also negatively associated with the number of positive lymph nodes (N1a vs N0 absolute fold change = 0.51, p = 0.043). We observed an excess of genes that were down-regulated in MF ILC compared to BT. 277 genes were differentially expressed at p<0.05 after correction by FDR. The top 5 differentially expressed genes (by fold change) between ILC and BT were COL11A1, COL1A1, WIF1, FIGF and PROM1. CDH1, a known marker of ILC, was down-regulated in ILC compared to benign (absolute fold change = 2.54, p = 6.02×10−6), ranked only 39/277 genes by fold change. MET and EGFR ranked 29 and 42/277 genes (p = 5.99×10−9 and 3.32×10−8 respectively). Conclusion: We observed significant heterogeneity for 20/730 genes between individual foci of MF ILC within the same pt suggesting that MF synchronous ILC lesions may have different biologic behavior. This in turn may have important prognostic implications with potential therapeutic utility. Our analyses also shows significant down-regulation of gene expression between ILC tumors and BT for many genes other than CDH1 suggesting that additional biomarkers can be used to further characterize ILC Study funded in part by Mayo Clinic Cancer Focus Research Team, the 26.2 with Donna Foundation and the Serene M. and Frances C. Durling endowment. Citation Format: Pooja Advani, Nadine Norton, Daniel J. Serie, Xochiquetzal Geiger, Julia E. Crook, Aubrey E. Thompson, Edith A. Perez, Alvaro Moreno Aspitia. Characterization of individual foci of multifocal invasive lobular breast cancer using gene expression profiling. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4761. doi:10.1158/1538-7445.AM2015-4761

Abstract Background: In United States there will be an estimated 74,690 new diagnoses and 15,580 deaths in 2014 due to bladder cancer. Lack of reliable markers of early disease and limited chemotherapy options have resulted in poor patient outcomes. A better approach is to treat patients with early intervention using targeted therapy. Arginine metabolic pathway has been identified to be deregulated in bladder cancer. Arginine is either derived from the microenvironment or synthesized by cells containing arginosuccinate synthase I (ASS1). Arginine undergoes subsequent conversion to nitric oxide (NO), which which can activate numerous intracellular signaling mediators, including guanylate cyclase (GC). We evaluated dual targeting of this metabolic pathway by arginine deprivation (ADI-PEG 20, Polaris Pharmaceuticals, San Diego, CA) and NO chelation (Cobinamide) as a therapy for bladder cancer. Experimental Design: Expression of arginine-NO pathway components in FFPE tissue sections was assessed by IHC and correlated with histopathology, outcomes and progression to identify pathway activation during early and late disease and stratified based on moderate to strong expression (2+, 3+) compared to absent to weak expression (0, 1+). In addition, a panel of bladder cell lines was screened for arginine-NO pathway components by immunoblotting and representative ASS1 negative and positive cells were used to evaluate the effects of arginine deprivation and/or NO chelation in vitro on colony formation and cell viability by clonogenic and MTT assays. Effects of NO production on bladder cancer survival and motility was assessed utilizing siRNA gene silencing and small molecule inhibitors. Effects of arginine deprivation by ADI-PEG 20 treatment in vivo was assessed by a mouse xenograft model. Results: Analysis of human tumors showed that ASS1 loss and increase in NOS occurred early in the course of bladder neoplastic progression. We identified loss of the ASS1 enzyme in bladder cancer cells and compensatory dependence on extracellular arginine, which can be successfully targeted using arginine removal to induce bladder cancer cell death, both in vitro and in vivo. We also identified increased nitric oxide synthase (NOS; the enzyme that catalyzes NO production) and GC activation in invasive bladder cancer cell lines. We characterized the effects of NO production on bladder cancer cell survival, growth and motility. Disruption of the NO pathway through gene silencing and small molecule inhibitors resulted in decreased cell survival, migration and invasion. Conclusions: Thus, dual targeting of this metabolic pathway using arginine chelation by the arginine degrading enzyme ADI-PEG20 and NO scavenging by a novel compound Cobinamide can be a potential mechanism to prevent development and/or progression of early bladder neoplasia. Citation Format: Divya Sahu, Sounak Gupta, Andrew M. Hau, Paul Elson, John Bomalaski, Gerry R. Boss, Donna E. Hansel. Targeting the arginine-nitric oxide pathway to arrest development and progression of invasive bladder cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4712. doi:10.1158/1538-7445.AM2015-4712

Abstract Panel-based germline genetic testing of established breast and ovarian cancer predisposition genes is widely used in high-risk clinics to inform on risks of breast and ovarian cancer. While most protein truncating mutations identified in predisposition genes are associated with moderate to high risks of these cancers, variants of uncertain significance (VUS), in the form of missense and intronic variants, have an undefined relevance to cancer. Many hundreds of VUS in BRCA1 and BRCA2 identified by clinical genetic testing over the last 20 years still have uncertain clinical relevance. In addition, VUS in other known predisposition genes are being identified by current panel-testing in up to 10% of individuals without BRCA1 or BRCA2 mutations. These VUS cause problems in cancer risk estimation and clinical management of patients undergoing clinical genetic testing. In an effort to evaluate the clinical significance of VUS in cancer predisposition genes we have developed functional assays that characterize the impact of the variants on the functions of the proteins encoded by the predisposition genes. In particular, a homology directed repair reporter assay providing a quantitative measure of homologous recombination repair of DNA double strand breaks has been developed for characterization of VUS in homologous recombination associated genes. When applied to 13 known pathogenic and 30 known neutral variants from the DNA binding domain of BRCA2, this assay exhibits 100% specificity and 100% sensitivity. Based on these results a model estimating the probability of pathogenicity of each BRCA2 VUS was developed. The HDR assay and probability model has now been applied to 230 BRCA2 VUS identified through genetic testing. Here we identify 80 BRCA2 variants that are deficient in HDR activity. We further combined these results with prior probabilities of pathogenicity derived from the sequence based Align GVGD in silico prediction model to estimate the posterior probability of pathogenicity for each VUS. Our results suggest that in the absence of family based segregation information, results from the HDR functional assay and in silico prediction models can be used to predict the pathogenicity of BRCA2 VUS. The assay has also been used to characterize VUS in the PALB2 and RAD51C predisposition genes, leading to identification of a subset of missense variants that are deficient in HDR repair activity and may be associated with increased risk of breast and ovarian cancer. These results will prove useful for improved risk assessment of carriers of these BRCA2, PALB2 and RAD51C variants and their family members and for validation of other approaches to mutation classification. Citation Format: Fergus J. Couch, Hermela Shimelis, Andreas Schroeder, Chunling Hu, Emily Hallberg, Gary Lipton, Edwin Iversen, Noralane M. Lindor. Characterization of variants of uncertain significance (VUS) in breast and ovarian cancer predisposition genes. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4685. doi:10.1158/1538-7445.AM2015-4685

Abstract Background: We investigated the role of multifactor interactions and epithelial ovarian cancer (EOC) risk using data collected within the Collaborative Oncological Gene-environment Study and the Ovarian Cancer Association Consortium containing 18,000 EOC cases and 26,000 controls. To date, researchers have identified 18 EOC susceptibility loci. However, it has been estimated that many more common variant loci exist; we hypothesized that some of the unexplained variation is due to gene-environment interactions. Similar to breast and endometrial cancers, many EOC risk factors relate to estrogen exposure, and increased levels of estrogen has been associated with obesity in post-menopausal women. Therefore, we hypothesized that gene-environment interactions dealing with hormone-related risk factors could differ between obese and non-obese women. Methods: We considered multifactor interactions between single nucleotide polymorphisms (SNPs) and six hormone-related factors: oral contraceptive use; parity; endometriosis; tubal ligation; hormone replacement therapy (HRT); and estrogen use; and assessed whether these GE interactions differed between obese and non-obese women. The SNPs included in this analysis included the top associated markers within the 18 confirmed risk loci, as well as, 36,811 SNPs that lie within 84 candidate gene regions related to hormone biosynthesis & metabolism and insulin-like growth factors. For the candidate SNP analyses, we assessed multifactor interactions across environmental factors using logistic regression models including age, study site, and population substructure. For the candidate gene analyses, a two-stage screening and testing procedure was utilized. All analyses were completed including: (1) all histologies; and (2) only serous histology. Results: For the 18 known EOC risk loci, no hormone-related risk factors showed statistically significant interaction at a threshold of 10-4. In contrast, the two-stage analysis of the candidate genes found several significant gene-environment-environment effects, after adjusting for multiple testing. For analysis of all cases, the SNPs in the following candidate genes showed a multifactor interaction with a hormone-related risk factor and obesity: IGFBP5 (HRT, rs729597, p = 1.3×10-13), CGA (HRT, rs58124219, p = 1.0×10-5), and SSTR1 (parity, rs77266093, p = 1.1×10-5). Similarly, the serous histology analyses detected a multifactor interaction with HRT, obesity and IGFBP5 (rs729597, p = 8.8×10-5). Additionally, serous only analyses also detected an interaction involving ESR1 (tubal ligation, Chr6:152246911(indel), p = 8.0×10-6). Conclusions: In a very large case-control collection, assessment of multifactor gene-environment interaction is feasible, and, here, suggest that the relationship between SNPs in candidate genes and hormone-related risk factors in EOC susceptibility may vary. Citation Format: Joseph Usset, Brooke Fridley, Ellen Goode, Joellen Schildkraut, Paul Pharoah. Assessment of multifactor gene-environment interactions and ovarian cancer risk: SNPs, obesity, and hormone-related risk factors. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4684. doi:10.1158/1538-7445.AM2015-4684

Abstract Purpose: Breast biopsies with benign findings (benign breast disease [BBD]) outnumber biopsies revealing in-situ or invasive cancer 4:1, providing an opportunity to identify histopathological features that predict breast cancer risk. Greater degrees of involution of terminal duct lobular units (TDLUs), the structures within the breast that produce milk and the primary source of breast cancer precursors, have been inversely associated with breast cancer risk among women with BBD. We recently developed measures to quantitate levels of TDLU involution, which demonstrate high inter/intra pathologist reproducibility. Here we assessed whether TDLU counts/100mm2 and median TDLU span (microns), two measures inversely related to degree of TDLU involution, are associated with subsequent breast cancer risk among women with BBD. Methods: From the Mayo BBD cohort (n = 9,376), we evaluated benign biopsies from 99 women who later developed breast cancer (cases) and 145 age-matched controls who did not develop breast cancer. Digitized images of biopsy sections were reviewed to enumerate TDLUs/mm2 and measure median TDLU span (microns) for up to ten normal TDLUs. Breast cancer risk factors were available from questionnaires or medical records. To assess associations with breast cancer risk, subjects were categorized into quartile levels of TDLU counts and median TDLU spans, based on data from controls. Odds ratios (ORs) and 95% confidence intervals (CI) were estimated using multivariable conditional logistic regression models, adjusted for histologic type of BBD, family history of breast cancer and menopausal hormone use. Results: Compared to controls, women who developed breast cancer had higher median number of TDLU counts/100mm2 (28 vs 20, p = 0.03) and larger TDLU spans (300 vs 267 microns, p = 0.14). In multivariable models higher TDLU counts (quartile4 vs. quartile1, OR = 2.44, 95%CI = 0.96-6.19, ptrend = 0.02) and larger TDLU span measures (quartile4 vs. quartile1, OR = 2.83, 95%CI = 1.13-7.06, ptrend = 0.03) were associated with subsequent diagnosis of breast cancer. Combinatorial metrics of TDLU counts with median TDLU span measures identified women at higher risk; specifically, women above the median for both TDLU span and TDLU counts had an OR = 3.75 (95%CI = 1.40-10.00, ptrend = 0.008), compared with women below the median for TDLU span and TDLU counts. Conclusion: These data show that lack of TDLU involution, as measured by increased persistence of TDLU counts, larger median TDLU spans and cross-classification using these measures was associated with increased breast cancer risk among women with BBD, extending prior work in this cohort based on other TDLU involution metrics. Future studies to identify determinants of TDLU involution, its association with breast cancer risk and its potential as an intermediate endpoint in prevention studies warrant consideration. Citation Format: Jonine D. Figueroa, Ruth Pfeiffer, Maya Palakal, Amy C. Degnim, Derek Radisky, Lynn C. Hartmann, Marlene Frost, Melody L. Stallings Mann, Louise A. Brinton, Daphne Papathomas, Daniel Visscher, Mark E. Sherman. Standardized measures of lobular involution and subsequent breast cancer risk among women with benign breast disease. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4682. doi:10.1158/1538-7445.AM2015-4682

Abstract Background Although a number of cancer chemopreventive agents have now been evaluated in phase II and phase III clinical trials that assess their efficacy and tolerability in both high risk and general populations, few have shown clear efficacy and uptake of approved agents has been disappointingly low. Therefore, chemopreventive agents should undergo rigorous screening prior to clinical trial evaluation in order to improve the selection of suitable agents for testing in large, expensive multi-center phase III trials. The aim of this study is to develop a descriptive method for prioritizing candidate agents for further development. Methods This framework was developed through a systematic literature review of Pubmed/Medline, Embase and The Cochrane Library. A semi-quantitative assessment focusing on efficacy and toxicity/tolerability was established based on expert input. Data from clinical trials, cancer epidemiological, observational and preclinical studies, including in vivo and in vitro basic research, was used to retrospectively evaluate the clinical application and effectiveness of cancer chemopreventive agents. Two chemopreventive agents with either a positive (aromatase inhibitors (AIs) for breast cancer) or negative (selenium for lung cancer) phase III clinical trial outcome were subsequently evaluated to determine the usefulness of the framework. Results Chemopreventive agents must reduce or reverse the process of carcinogenesis in normal or pre-malignant tissue with minimal toxicity. Evaluation of efficacy included data from preclinical and animal models, epidemiological studies, and clinical trials (both early phase trials in the organ of interest as well as secondary endpoints from other relevant studies). Agent suitability evaluation included assessment of toxicities, tolerability, and agent specific attributes (PK/PD, bioavailability, dosing, etc.). Examination of selenium and AI historical data in these categories revealed strong concordance with phase III outcomes. The detailed descriptive evaluation of the agents will be presented. Conclusion This framework provides a descriptive paradigm for cancer chemopreventive agent evaluation through systematic and rigorous interrogation. Thorough, standardized evaluation of all available preclinical, epidemiological, and clinical data should be strongly encouraged to better inform the investments required for clinical trials. Citation Format: Naomi Walsh, David S. Alberts, Powel Brown, Paul Limburg, Mark Sherman, Eva Szabo. Framework for clinical evaluation of chemopreventive agents: Defining criteria for future assessment. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4662. doi:10.1158/1538-7445.AM2015-4662

Abstract Background: Recent bioinformatic analyses and high-throughput sequencing efforts have revealed that RNA editing is a widespread post-transcriptional modification that affects non-coding repetitive elements throughout the genome. This modification, which involves the conversion of adenosine into inosine, is mediated by the action of the adenosine deaminases acting on RNA (ADAR) family of enzymes. Based on reports of dysfunctional ADAR expression in epithelial ovarian cancers (EOC), we hypothesized that single nucleotide polymorphisms (SNPs) in five ADAR genes (ADAD1, ADAR1, ADAR2, ADAR3, SND1) may contribute to EOC susceptibility, potentially by altering the expression of corresponding genes in ovarian tissues. Methods: Using directly genotyped and imputed data from 10,891 invasive EOC cases and 21,693 healthy controls of European ancestry represented in the international Collaborative Oncological Gene-environment Study, we evaluated the frequency of 7,133 SNPs in 5 RNA editing genes. Unconditional logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) between genotypes and case status under a log-additive genetic model, with adjustment for the first five principal components representing European ancestry. Subgroup analysis was conducted for the four main histologic sub-types of EOC. We also aggregated SNP-level data and conducted gene-level analyses using the Admixture Maximum Likelihood (AML) test and the Sequence-Kernel Association test for the combined effect of common and rare variants (SKAT-CR). Results: Association analysis revealed a top risk-associated SNP (OR (95% CI) = 1.39 (1.17-1.64)) in ADAR3 with P = 1.3 x10−4, 17 additional SNPs in ADAR1, ADAR3, or SND1 with P = 10−4, and 51 SNPs with P = 10−3; this observed number of statistically-significant SNPs was greater than expected. When restricting to the 6,500 invasive serous adenocarcinomas, the magnitude of association strengthened for numerous SNPs. Exploratory analysis for the less common histologic subtypes (endometrioid (n = 1,439), mucinous (n = 696), and clear cell (n = 660)) revealed SNP-level associations (P<10−3) unique to each sub-type. Gene-level analyses based on AML and SKAT-CR revealed that ADAR1 was globally associated (P<0.10) with susceptibility to invasive EOC, with P = 0.024 and P = 0.023, respectively. Expression quantitative trait locus analysis is underway to evaluate associations between genotype and gene expression in EOC and precursor tissues using data from The Cancer Genome Atlas Project and other sources. Conclusions: Preliminary data from this large-scale collaboration suggest that germline variation involving RNA editing genes may influence susceptibility to EOC. This novel finding warrants further investigation, as inherited and acquired alterations affecting RNA editing may serve as important biological mechanisms that promote the development of EOC and possibly other malignancies. Citation Format: Jennifer Permuth-Wey, Brett M. Reid, Y. Ann Chen, Hui-Yi Lin, Alvaro Monteiro, Zhihua Chen, Andrew Berchuck, Georgia Chenevix-Trench, Jennifer Doherty, Simon Gayther, Ellen Goode, Edwin Iversen, Leigh Pearce, Paul D. P. Pharoah, Catherine Phelan, Susan Ramus, Mary Anne Rossing, Joellen Schildkraut, Thomas Sellers, on behalf of the Ovarian Cancer Association Consortium. Inherited variants affecting RNA editing may contribute to ovarian cancer susceptibility. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4635. doi:10.1158/1538-7445.AM2015-4635

Abstract Background: Cell-type-specific super enhancers, or large clusters of mediator-rich enhancers, are highly transcribed regions of the genome with key functions in the maintenance of cell identity and control. These regulatory elements are easily perturbed and enriched for disease-associated sequence variation. Identification of disease-associated super-enhancers and investigation of their sequence variants may help identify important, biologically relevant associations previously overlooked by genome-wide studies. Thus, we sought to identify variants within ovarian tissue specific super-enhancers and investigate their association with epithelial ovarian cancer (EOC) susceptibility. Methods: We utilized genotype data from a European population of ∼11,000 invasive EOC cases and 22,000 healthy controls from the Ovarian Cancer Association Consortium. A public catalog of 478 super-enhancers, identified by peak histone H3K27ac ChIP-seq signal in ovarian tissue, was used to identify 72,116 variants located within 344 super-enhancer regions. Unconditional logistic regression (log-additive model) was used to assess individual SNP-EOC risk associations. Additionally, we reviewed associations within 500 kb of super-enhancer regions to identify proximal associated variants. When LD structure indicated, we employed conditional analysis to determine independence of the super-enhancer signal. Gene-EOC risk associations were assessed with the Admixture Maximum Likelihood (AML) test and Sequence-Kernel Association Test (SKAT) for the combined effect of common and rare (MAF<0.01) variants. Models were adjusted for population stratification using the first five principal components. To adjust for multiple comparisons, a false discovery rate of 10% was used. Results: The most significant SNP observed was rs147487657 (p = 7.7×10−6) on locus 2q37.3 in the HDAC4 gene and there were an additional 12 SNPs with p<1.0×10−4, although not significant at 10% FDR. Gene-level analyses identified two super-enhancers significantly associated with EOC risk. An enhancer at the previously reported 2q31 locus, containing HOXD8 and HOXD-AS2, associated (SKAT p = 4.3×10−5; AML p<0.001) with the top SNP located within the 5′UTR of HOXD8 transcription factor (rs847163 (T>C); p<2.08×10−5). Conditional analysis did not identify an independent association within the super-enhancer region. The second EOC-associated super-enhancer identified was at locus 1q32 (SKAT p = 0.00052, AML p = 0.001) containing the CSRP1 gene. There were 6 SNPs with p<0.001 and 16 SNPs with p<0.01, all were inversely associated with risk and consistent with the top hit at rs927901 T>C (OR = 0.94; 95% CI: 0.90-0.97; p = 2.1×10−4). Associations varied by histologic subtype, possibly due to variance in disease relevant super-enhancer cell-type.Conclusion: These findings suggest germline variation within super-enhancer regions may contribute to EOC susceptibility Citation Format: Brett M. Reid, Jennifer Permuth-Wey, Y. Ann Chen, Hui-Yi Lin, Alvaro Monteiro, Zhihua Chen, Andrew Berchuck, Georgia Chenevix-Trench, Jennifer Doherty, Simon Gayther, Ellen L. Goode, Edwin Iversen, Leigh Pearce, Paul Pharoah, Catherine Phelan, Susan Ramus, Mary Anne Rossing, Joellen Schildkraut, Thomas Sellers, on behalf of the Ovarian Cancer Association Consortium. Variants within super-enhancer regulatory elements associate with epithelial ovarian cancer risk. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4634. doi:10.1158/1538-7445.AM2015-4634