Wei, Peng

AbstractIt is of great scientific interest to identify interactions between genetic variants and environmental exposures that may modify the risk of complex diseases. However, larger sample sizes are usually required to detect gene‐by‐environment interaction (G × E) than required to detect genetic main association effects. To boost the statistical power and improve the understanding of the underlying molecular mechanisms, we incorporate functional genomics information, specifically, expression quantitative trait loci (eQTLs), into a data‐adaptive G × E test, called aGEw. This test adaptively chooses the best eQTL weights from multiple tissues and provides an extra layer of weighting at the genetic variant level. Extensive simulations show that the aGEw test can control the Type 1 error rate, and the power is resilient to the inclusion of neutral variants and noninformative external weights. We applied the proposed aGEw test to the Pancreatic Cancer Case–Control Consortium (discovery cohort of 3,585 cases and 3,482 controls) and the PanScan II genome‐wide association study data (replication cohort of 2,021 cases and 2,105 controls) with smoking as the exposure of interest. Two novel putative smoking‐related pancreatic cancer susceptibility genes, TRIP10 and KDM3A, were identified. The aGEw test is implemented in an R package aGE.

Background We examined feasibility of a unique approach towards gaining insight into heritable risk for early atherosclerosis: surveying gene expression by endothelial cells from living subjects. Methods and Results Subjects aged <50 years (mean age, 37; range, 22–49) without obstructive coronary artery disease underwent coronary reactivity testing that identified them as having normal or abnormal coronary endothelial function. Cultures of Blood Outgrowth Endothelial Cells (BOEC) from 6 normal and 13 abnormal subjects passed rigorous quality control and were used for microarray assessment of gene expression. Of 9 genes differentially expressed at false discovery rate <0.1%, we here focus upon abnormal subjects having elevated expression of HMGB1 (high mobility group box 1) which we unexpectedly found to be linked to low LAMC1 (laminin gamma 1) expression. This linkage was corroborated by 3 of our past studies and confirmed bio‐functionally. Compared with normal BOEC, abnormal BOEC released 13±3‐fold more HMGB1 in response to lipopolysaccharide; and they deposited one tenth as much LAMC1 into collagen subendothelial matrix during culture. Clinical follow‐up data are provided for 4 normal subjects (followed 13.4±0.1 year) and for 12 abnormal subjects (followed 9.1±4.5 years). Conclusions The known pathogenic effects of high‐ HMGB1 and low‐ LAMC1 predict that the combination would biologically converge upon the focal adhesion complex, to the detriment of endothelial shear responsiveness. This gene expression pattern may comprise a heritable risk state that promotes early coronary atherosclerosis. If so, the testing could be applied even in childhood, enabling early intervention. This approach offers a way to bridge the information gap between genetics and clinical phenotype.

Abstract Background: Obesity and diabetes are major modifiable risk factors for pancreatic cancer. Interactions between genetic variants and diabetes/obesity have not previously been comprehensively investigated in pancreatic cancer at the genome-wide level. Methods: We conducted a gene–environment interaction (GxE) analysis including 8,255 cases and 11,900 controls from four pancreatic cancer genome-wide association study (GWAS) datasets (Pancreatic Cancer Cohort Consortium I–III and Pancreatic Cancer Case Control Consortium). Obesity (body mass index ≥30 kg/m2) and diabetes (duration ≥3 years) were the environmental variables of interest. Approximately 870,000 SNPs (minor allele frequency ≥0.005, genotyped in at least one dataset) were analyzed. Case–control (CC), case-only (CO), and joint-effect test methods were used for SNP-level GxE analysis. As a complementary approach, gene-based GxE analysis was also performed. Age, sex, study site, and principal components accounting for population substructure were included as covariates. Meta-analysis was applied to combine individual GWAS summary statistics. Results: No genome-wide significant interactions (departures from a log-additive odds model) with diabetes or obesity were detected at the SNP level by the CC or CO approaches. The joint-effect test detected numerous genome-wide significant GxE signals in the GWAS main effects top hit regions, but the significance diminished after adjusting for the GWAS top hits. In the gene-based analysis, a significant interaction of diabetes with variants in the FAM63A (family with sequence similarity 63 member A) gene (significance threshold P &lt; 1.25 × 10−6) was observed in the meta-analysis (PGxE = 1.2 ×10−6, PJoint = 4.2 ×10−7). Conclusions: This analysis did not find significant GxE interactions at the SNP level but found one significant interaction with diabetes at the gene level. A larger sample size might unveil additional genetic factors via GxE scans. Impact: This study may contribute to discovering the mechanism of diabetes-associated pancreatic cancer.

Abstract Introduction: Obesity and diabetes are two major modifiable risk factors for pancreatic cancer. Genetic variation affecting the risk of obesity- and diabetes-related pancreatic cancer have yet to be comprehensively investigated at the genome-wide level. Aims: To identify novel gene-environment interactions (GxE) of obesity/diabetes on pancreatic cancer risk. Methods: We conducted a multi-stage genome-wide gene-diabetes and gene-obesity interaction study using GWAS data of 7095 cases and 10710 controls from Pancreatic Cancer Consortium (PanScan I, PanScan II and PanScan III) and Pancreatic Cancer Case Control Consortium (PanC4). About 870,000 SNPs, having MAF ≥ 0.005 and having been genotyped in at least one study, were analyzed. Missing genotypes were imputed using IMPUTE2 and the Haplotype Reference Consortium (HRC) reference panel. Case-control (CC) and case-only (CO) methods were used for SNP-level GxE analysis. Fixed-, random- and joint-effect models built in “rareGE” R package were used for gene-based GxE analysis. About 20,000 genes that have at least one SNP genotyped were included in the analysis. Age, gender and principal components accounting for subpopulation structure were adjusted in both SNP-level and gene-level analyses. Obesity was defined as body mass index ≥30 mg/m2. Diabetics with &lt;3-year duration were excluded for gene-diabetes interaction analysis. The PanScan I, II, III and PanC4 datasets were analyzed separately followed by meta-analysis of the summary statistics. Fixed-effect and Fisher’s p-value based meta-analysis was applied for SNP-level and gene-level GxE analysis, respectively. Results: No genome-wide significant (P&lt;5 x 10-8 ) interaction at the SNP level was detected in any of the PanScan or PanC4 datasets or in the meta-analysis. A significant interaction of diabetes and FAM63A (family with sequence similarity 63 member A) gene (P&lt;2.5 x 10-6) was observed in the meta-analysis (PFixed= 3.4 x 10-5, PRandom= 1.9 x 10-6, PJoint= 6.8 x 10-7) . This interaction showed nominal significance in all individual dataset analyses. The corresponding P values for this interaction were 4.2 x 10-2, 3.8 x 10-2 and 6.8 x 10-2 in PanScan I, 2.8 x 10-2, 3.9 x 10-2 and 6.5 x 10-2 9.1 x 10-3, 2.2 x 10-4 and 8.8 x 10-6 in PanScan III, and 3.1 x 10-3, 3.4 x 10-3 and 8.8 x 10-3 in PanC4, respectively. Conclusions: In the largest genome-wide GxE analysis for pancreatic cancer to date, we identified a significant diabetes interacting gene in pancreatic cancer, which was consistently replicated in 4 individual GWAS studies. This observation identifies FAM63A as a candidate for future functional studies to understand its role in pancreatic cancer etiology. Further increasing the sample size may unveil additional genetic loci that contribute to the susceptibility to pancreatic cancer via G x E. Citation Format: Hongwei Tang, Lai Jiang, Laufey T. Amundadottir, Harvey A. Risch, Rachael S. Stolzenberg-Solomon, Alison P. Klein, Brian M. Wolpin, Gloria Petersen, Nilanjan Chatterjee, Donghui Li, Peter Kraft, Peng Wei. Genomewide gene-diabetes and gene-obesity interaction scan in 7095 cases and 10710 controls from Pancreatic Cancer Cohort Consortium and Pancreatic Cancer Case Control Consortium [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1295. doi:10.1158/1538-7445.AM2017-1295

Abstract Introduction Pancreatic cancer is a highly lethal malignancy due to late diagnosis and aggressiveness of the tumor. Characterization of high-risk populations for monitoring, intervention and early detection remains challenging. Obesity and diabetes are potentially modifiable contributors to pancreatic cancer. Genetic factors modifying the risk of obesity- and diabetes-induced pancreatic cancer have previously not been fully investigated at the genome-wide level. Hypothesis There are potential genetic factors modifying the associations between obesity/diabetes and pancreatic cancer. Methods Integrating genotype with risk factor data from a GWAS for the Pancreatic Cancer Case Control Consortium, we performed a genome-wide gene-environment interaction (G x E) scan for 457,688 SNPs in a discovery study of 2,028 cases and 2,109 controls using state-of-the-art methods including case-only (CO), case-control (CC), Empirical Hierarchical Bayes (EHB), Empirical-Bayes test (EB) and three 2-step approaches (DG2, EG2, DGEG2). Results We detected a genome-wide significant (P &lt; 1.09 x 10-7 ) interaction of diabetes with rs13061928 (in CNTN4) in CO (P = 8.8 x 10-8 ), which was independent of diabetes in control group (P = 0.08). The minor allele was in a synergism with diabetes on the cancer risk: diabetics carrying GA/AA genotype had a 3.39-fold (95% CI: 2.47-4.65) increased disease risk compared with non-diabetics carrying GG genotype. Consistently, the SNP-diabetes interaction had a high ranking statistic in EHB (No.1) and high p-value ranks for EG2 and DGEG2 methods (No.1, P= 5.44 x 10-5). Conclusions These observations, once validated/confirmed, may help define at-risk subpopulation and initiate targeted intervention and prevention of pancreatic cancer. Genome-wide G x E analysis requires larger sample size than genetic main effects scan; combining the CO method with alternative methods may be the optimal scheme for genome-wide G x E analysis. Citation Format: Hongwei Tang, Eric J. Duell, Harvey A. Risch Risch, Sara H. Olson, H. Bas Bueno-de-Mesquita, Steven Gallinger, Elizabeth A. Holly, Gloria M. Petersen, Paige M. Bracci, Robert R. McWilliams, Mazda Jenab, Elio Riboli, Anne Tjønneland, Marie Christine Boutron-Ruault, Rudolf Kaaks, Dimitrios Trichopoulos, Salvatore Panico, Malin Sund, Petra H. M Peeters, Kay-Tee Khaw, Christopher I Amos, Donghui Li, Peng Wei. Genome-wide gene-diabetes and gene-obesity interaction scan in the pancreatic cancer case control consortium. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2214. doi:10.1158/1538-7445.AM2014-2214