Chanock, Stephen Jacob

Abstract Pancreatic Ductal Adenocarcinoma (PDAC) is the third leading cause of cancer-related deaths in United States with a 5-year survival rate of only 8%. Inherited predisposition plays an important role in PDAC risk. Rare, moderately to highly penetrant mutations in hereditary cancer and pancreatitis genes, identified in families with a high incidence of disease, account for a small fraction of cases. At the other end of the spectrum, common risk variants of low penetrance have been discovered through genome-wide association studies (GWAS). Currently, over 20 common PDAC risk signals have been identified through GWAS in populations of European ancestry. However, these loci explain only a small fraction of genetic heritability for PDAC, and the causal genes for the large majority of these are unknown. Most susceptibility alleles discovered through GWAS are located in noncoding regions of the genome and are likely to function through allele specific regulation of gene expression. Transcriptome-wide association studies integrate expression quantitative trait loci (eQTL) with GWAS data to discover genes whose cis-regulated expression is associated with complex traits. To identify novel susceptibility loci and likely causal genes, we applied this approach to PDAC. Firstly, we used two reference datasets generated from histologically normal pancreatic tissue samples from individuals of European ancestry, the DCEG/LTG (n=95) and Genotype-Tissue Expression (GTEx v7, n=174), to build robust predictors for gene expression (n=2,872-5,906; r2>0.01), using 5 different models (Top1, Lasso, Enet, Blup and Bslmm) in FUSION and PrediXcan. Gene expression predictors (n=2,043-21,422) were also generated by Enet in 48 different tissues (n=74-421 samples) from 608 individuals of European ancestry (GTEx). We then applied these predictor models to PDAC GWAS summary statistics from 9,040 PC cases and 12,496 controls. We discovered 24 genes whose genetically predicted expression is significantly associated with PDAC risk (FDR < 0.05), including 13 candidate genes at 10 novel loci (CELA3B, SMC2, SMC2-AS1, RP11-80H5.9, SMUG1, BTBD6, RCCD1, PNMT, CDK12, PGAP3, SUPT4H1, RP11-888D10.3 and PGPEP1) and 11 at 5 known risk loci (TERT, CLPTM1L, ZDHHC11B, INHBA, ABO, PDX1, KLF5, WDR59, CFDP1, BCAR1 and TMEM170A). Pathway enrichment analysis for genes differentially expressed in the top vs. bottom quartile of CELA3B expression highlighted higher expression of inflammatory and immune response genes in both the DCEG/LTG and GTEx pancreatic transcriptome datasets, indicating that low expression of this gene may be associated with an inflammatory state in the pancreas. Our findings showcase the utility of integrating multiple datasets to generate hypotheses and provide novel clues for the genetic mechanisms underlying pancreatic cancer risk. Citation Format: Jun Zhong, Ashley Jermusyk, Lang Wu, Jason W. Hoskins, Irene Collins, Mingfeng Zhang, Song Lei, Charles C. Chung, Tongwu Zhang, Wenming Xiao, Rachael Stolzenberg-Solomon, Alison P. Klein, Brian M. Wolpin, Xiao-Ou Shu, Stephen J. Chanock, Sara Olson, Nilanjan Chatterjee, Jill Smith, Jianxin Shi, Peter Kraft, Gloria M. Petersen, Wei Zheng, Laufey T. Amundadottir. Large-scale transcriptome-wide association study (TWAS) identifies novel candidate susceptibility genes for pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1591.

Abstract Genome-wide association studies (GWAS) have so far identified 25 loci associated with glioma risk, with most showing specificity for either glioblastoma (GBM) or non-GBM tumors. The majority of these GWAS susceptibility variants reside in noncoding regions and the causal genes underlying the associations are largely unknown. Here we performed a transcriptome-wide association study to search for novel risk loci and candidate causal genes at known GWAS loci using Genotype-Tissue Expression Project (GTEx) data to predict cis-predicted gene expression in relation to GBM and non-GBM risk in conjunction with GWAS summary statistics on 12,488 glioma cases (6,183 GBM and 5,820 non-GBM) and 18,169 controls. Imposing a Bonferroni-corrected significance level of P < 5.69 × 10−6, we identified 31 genes, including GALNT6 at 12q13.33, as a candidate novel risk locus for GBM (mean Z = 4.43; P = 5.68 × 10−6). GALNT6 resides at least 55 Mb away from any previously identified glioma risk variant, while all other 30 significantly associated genes were located within 1 Mb of known GWAS-identified loci and were not significant after conditioning on the known GWAS-identified variants. These data identify a novel locus (GALNT6 at 12q13.33) and 30 genes at 12 known glioma risk loci associated with glioma risk, providing further insights into glioma tumorigenesis. Significance: This study identifies new genes associated with glioma risk, increasing understanding of how these tumors develop.

Abstract Background: Whether associations between circulating metabolites and prostate cancer are causal is unknown. We report on the largest study of metabolites and prostate cancer (2,291 cases and 2,661 controls) and appraise causality for a subset of the prostate cancer–metabolite associations using two-sample Mendelian randomization (MR). Methods: The case–control portion of the study was conducted in nine UK centers with men ages 50–69 years who underwent prostate-specific antigen screening for prostate cancer within the Prostate Testing for Cancer and Treatment (ProtecT) trial. Two data sources were used to appraise causality: a genome-wide association study (GWAS) of metabolites in 24,925 participants and a GWAS of prostate cancer in 44,825 cases and 27,904 controls within the Association Group to Investigate Cancer Associated Alterations in the Genome (PRACTICAL) consortium. Results: Thirty-five metabolites were strongly associated with prostate cancer (P < 0.0014, multiple-testing threshold). These fell into four classes: (i) lipids and lipoprotein subclass characteristics (total cholesterol and ratios, cholesterol esters and ratios, free cholesterol and ratios, phospholipids and ratios, and triglyceride ratios); (ii) fatty acids and ratios; (iii) amino acids; (iv) and fluid balance. Fourteen top metabolites were proxied by genetic variables, but MR indicated these were not causal. Conclusions: We identified 35 circulating metabolites associated with prostate cancer presence, but found no evidence of causality for those 14 testable with MR. Thus, the 14 MR-tested metabolites are unlikely to be mechanistically important in prostate cancer risk. Impact: The metabolome provides a promising set of biomarkers that may aid prostate cancer classification.

Glioblastoma (GBM) is the most common malignant brain tumor in the United States. Incidence of GBM increases with age, and younger age‐at‐diagnosis is significantly associated with improved prognosis. While the relationship between candidate GBM risk SNPs and age‐at‐diagnosis has been explored, genome‐wide association studies (GWAS) have not previously been stratified by age. Potential age‐specific genetic effects were assessed in autosomal SNPs for GBM patients using data from four previous GWAS. Using age distribution tertiles (18–53, 54–64, 65+) datasets were analyzed using age‐stratified logistic regression to generate p values, odds ratios (OR), and 95% confidence intervals (95%CI), and then combined using meta‐analysis. There were 4,512 total GBM cases, and 10,582 controls used for analysis. Significant associations were detected at two previously identified SNPs in 7p11.2 (rs723527 [p54–63 = 1.50x10−9, OR54–63 = 1.28, 95%CI54–63 = 1.18–1.39; p64+ = 2.14x10−11, OR64+ = 1.32, 95%CI64+ = 1.21–1.43] and rs11979158 [p54–63 = 6.13x10−8, OR54–63 = 1.35, 95%CI54–63 = 1.21–1.50; p64+ = 2.18x10−10, OR64+ = 1.42, 95%CI64+ = 1.27–1.58]) but only in persons >54. There was also a significant association at the previously identified lower grade glioma (LGG) risk locus at 8q24.21 (rs55705857) in persons ages 18–53 (p18–53 = 9.30 × 10−11, OR18–53 = 1.76, 95%CI18–53 = 1.49–2.10). Within The Cancer Genome Atlas (TCGA) there was higher prevalence of ‘LGG’‐like tumor characteristics in GBM samples in those 18–53, with IDH1/2 mutation frequency of 15%, as compared to 2.1% [54–63] and 0.8% [64+] (p = 0.0005). Age‐specific differences in cancer susceptibility can provide important clues to etiology. The association of a SNP known to confer risk for IDH1/2 mutant glioma and higher prevalence of IDH1/2 mutation within younger individuals 18–53 suggests that more younger individuals may present initially with ‘secondary glioblastoma.’

Abstract Multiple myeloma (MM) is twice as common in African Americans (AA) compared to European Americans (EA). The reported familial clustering and the elevated MM risk among first-degree relatives of cases implicate genetic susceptibility. Previous genome-wide association studies (GWAS) in EA have identified 16 novel risk loci. In this study, we tested the generalizability of the established risk alleles to AA and conducted a meta-GWAS analysis using two sets of AA to identify additional novel common MM risk variants. In the first study, we genotyped 1,305 incident AA MM cases from the African American Multiple Myeloma Study (AAMMS) using the Illumina HumanCore GWAS array and compared them to 7,078 AA controls from the African Ancestry Prostate Cancer Consortium (AAPC) and African Ancestry Breast Cancer Consortium (AABC) using the Illumina 1M-Duo. In the second study, 95 additional AAMMS cases and 435 AA MM cases from the University of Arkansas for Medical Sciences (UAMS) were genotyped using the Illumina MegaBead Chip and compared to 2,390 AA controls from the Multiethnic Cohort. The Haplotype Reference Consortium (HRC) was used to impute the overlapping typed SNPS from each GWAS case and control set together. Per-allele risk associations were tested for 8,715,278 overlapping genotyped and imputed variants with >1% frequency and >0.8 imputation score using unconditional logistic regression in both sets, and the combined effects were estimated using a fixed-effect meta-analysis. Of the 16 reported risk loci discovered in EA, directional consistency was present for 15 variants; eight of these replicated at nominal significance p<0.05, with the most statistically significant variant being rs4487645 at 7p15.3 (OR=1.38, p=3.56×10-6). AA individuals with polygenic risk scores from these 16 variants (PRS) in the top 10% stratum had a 1.44-fold increased MM risk compared to those with a PRS in the 25th -75th percentiles. Additionally, we identified three suggestive novel loci located at 12q12, 9p24.3 and 9p13.1 at p<1×10-6, with ORs ranging from 1.25-1.55, but none reached genome-wide significance. The variant at 9p24.3 is located in an intron in the KANK1 gene and a correlated SNP in EAs (r2=0.5) is strongly associated with gene expression in neoplastic plasma cells (unpublished, Weinhold and Morgan). Our study replicated most of the reported risk loci discovered among EA, demonstrated that a PRS constructed using the 16 reported risk alleles was associated with MM risk, and provides suggestive evidence for additional loci associated with MM risk in AAs. Citation Format: Zhaohui Du, Chi Song, Kristin Rand, Niels Weinhold, David Van Den Berg, Amie Hwang, Xin Sheng, Victor Hom, Sikander Ailawadhi, Ajay K. Nooka, Seema Singhal, Karen Pawlish, Edward S. Peters, Cathryn Bock, Ann Mohrbacher, Alexander Stram, Sonja I. Berndt, William Blot, John David Carpten, Antoinette Stroup, Andrew Olshan, Wei Zhang, African Ancestry Breast & Prostate Consortium, Stephen Chanock, Jayesh Mehta, Graham A. Colditz, Jeffrey Wolf, Thomas G. Martin, Michael Tomasson, Mark A. Fiala, Howard Terebelo, Nalini Janakiraman, Laurence Kolonel, Loic LeMarchand, Elad Ziv, Daniel Stram, Ravi Vij, Leon Bernal-Mizrachi, Gareth J. Morgan, Jeffrey A. Zonder, Carol Ann Huff, Sagar Lonial, Robert Z. Orlowski, David V. Conti, Christopher A. Haiman, Wendy Cozen. A meta-analysis of genome-wide association studies of multiple myeloma among African Americans [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 223.

Abstract Background: We previously identified 5 SNPs at 4 susceptibility loci for diffuse large B-cell lymphoma (DLBCL) in individuals of European ancestry through a large genome-wide association study (Cerhan et al, Nat Gen 2014;46:1233-8); however, much of the heritability remains unexplained. To further elucidate genetic susceptibility to DLBCL, we sought to validate 2 loci at 3q13.33 and 3p24.1 that were suggestive in the original report. Methods: We selected two SNPs rs9831894 (3q13.33) and rs6773363 (3p24.1) for de novo replication from studies at the Mayo Clinic, MD Anderson, and Memorial Sloan Kettering. Logistic regression was used to estimate odds ratios (ORs), using a log-additive model, and adjusting for age, gender and significant eigenvectors for each genotyping center separately. For both SNPs, we then conducted a meta-analysis of all replication studies (n=3) and our original GWAS studies (n=4), encompassing in total 5662 cases and 9237 controls for rs9831894, and 5510 cases and 12,817 controls for rs6773363. Meta-analysis was conducted using the fixed-effects inverse variance method based on the β estimates and standard errors from each study. Results: In a meta-analysis of the 4 original GWAS scans (3856 cases and 7666 controls), rs9831894 (MAF=0.40) was associated with DLBCL risk (OR=0.84, P=4.52 x 10-9) and this replicated in a meta-analysis of the 3 studies with de novo genotyping (OR=0.80, P=4.17 x 10-5); the overall meta-analysis showed a strong association with DLBCL risk (OR=0.83, P=3.62 x 10-13). This locus maps near a plausible candidate gene, CD86, a protein coding gene and a member of the immunoglobulin superfamily that encodes a type I membrane protein. Binding of CD86 with cytotoxic T-lymphocyte-associated protein 4 negatively regulates T-cell activation and diminishes the immune response, while binding of CD86 with CD28 antigen is a costimulatory signal for activation of the T-cell. In a meta-analysis of the same 4 GWAS scans, rs6773363 (MAF=0.45) was tentatively associated with DLBCL risk (OR=1.17, P= 3.68 x 10-7) and this replicated in a meta-analysis of the studies with de novo genotyping (OR=1.27, P=3.78 x 10-7); the overall meta-analysis showed a strong association with DLBCL risk (OR=1.20, P=2.31 x 10-12). This locus also maps near a plausible candidate gene, eomesodermin (EOMES), a transcription factor crucial for embryonic development of the central nervous system and also putatively involved in T-cell differentiation in viral infection defense. Conclusion: In this follow-up analysis of our initial GWAS, we have identified two additional loci associated with risk of DLBCL, the most common lymphoma subtype. These loci provide additional evidence for the role of immune function in the etiology of DLBCL. Citation Format: Geffen Kleinstern, Michelle Hildebrand, Vijai Joseph, Sonja I. Berndt, Hervé Ghesquières, James McKay, Sophia S. Wang, Alexandra Nieters, David Cox, Alain Monnereau, Angela R. Brooks-Wilson, Qing Lan, Mads Melbye, Rebecca D. Jackson, Lauren R. Teras, Mark P. Purdue, Claire M. Vajdic, Demetrius Albanes, Anne Zeleniuch-Jacquotte, Simon Crouch, Yawei Zhang, Susan L. Slager, Xifeng Wu, Karin E. Smedby, Gilles Salles, Christine F. Skibola, Nathaniel Rothman, Stephen J. Chanock, James R. Cerhan. Inherited variants at 3q13.33 and 3p24.1 influences risk of diffuse large B-cell lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 221.

Abstract Objective Genome-wide association studies (GWAS) identify associations of individual SNPs with cancer risk but usually only explain a fraction of the inherited variability. Pathway analysis of genetic variants has been shown for many disorders to be a powerful tool in discovering novel networks of susceptibility genes. Design We have conducted a large agnostic pathway-based meta-analysis of GWAS data using the summary-based adaptive rank truncated product (sARTP) method to identify novel gene sets and pathways associated with pancreatic ductal adenocarcinoma (PDA) in 9,040 cases and 12,495 controls. We performed expression quantitative trait loci (eQTL) analysis and functional annotation of the top SNPs in genes identified in PDA-associated gene sets and pathways. Results We identified 14 gene sets and pathways associated with PDA at FDR < 0.05. Five of the strongest signals (P-value ≤ 1.3x10-5 Bonferroni corrected) were in genes from KEGG maturity onset diabetes of the young, Reactome regulation of beta cell development, Biocarta role of epidermal growth factor (EGF) receptor transactivation by G-protein-coupled receptors in cardiac hypertrophy pathway, Nikolsky breast cancer chr17q11-q21 amplicon gene set, and Pujana ATM Pearson correlation coefficient (PCC) network gene set. We identified and validated rs876493 and three correlated SNPs (PGAP3) and rs3124737 (CASP7) as eQTLs in two normal derived pancreas tissue datasets. Conclusion Our agnostic gene set and pathway analysis integrated with functional annotation and eQTL analysis provides novel insight into genes and pathways that may be biologically relevant for risk of PDA. Citation Format: Naomi Walsh, Han Zhang, Paula L. Hyland, Qi Yang, Evelina Mocci, Mingfeng Zhang, Erica J. Childs, Zhaoming Wang, Stephen Chanock, Patricia Hartge, Robert Hoover, Peter Kraft, Donghui Li, Eric J. Jacobs, Gloria M. Petersen, Brian M. Wolpin, Harvey A. Risch, Laufey T. Amundadottir, Kai Yu, Alison P. Klein, Rachael Z. Stolzenberg-Solomon. Large pathway and gene set analysis of GWAS data identifies novel associations for pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 242.