Amos, Christopher I.

Abstract Background Observational studies indicate that periodontal disease may increase the risk of colorectal, lung, and pancreatic cancers. Using a 2-sample Mendelian randomization (MR) analysis, we assessed whether a genetic predisposition index for periodontal disease was associated with colorectal, lung, or pancreatic cancer risks. Methods Our primary instrument included single nucleotide polymorphisms with strong genome-wide association study evidence for associations with chronic, aggressive, and/or severe periodontal disease (rs729876, rs1537415, rs2738058, rs12461706, rs16870060, rs2521634, rs3826782, and rs7762544). We used summary-level genetic data for colorectal cancer (n = 58 131 cases; Genetics and Epidemiology of Colorectal Cancer Consortium, Colon Cancer Family Registry, and Colorectal Transdisciplinary Study), lung cancer (n = 18 082 cases; International Lung Cancer Consortium), and pancreatic cancer (n = 9254 cases; Pancreatic Cancer Consortia). Four MR approaches were employed for this analysis: random-effects inverse-variance weighted (primary analyses), Mendelian Randomization-Pleiotropy RESidual Sum and Outlier, simple median, and weighted median. We conducted secondary analyses to determine if associations varied by cancer subtype (colorectal cancer location, lung cancer histology), sex (colorectal and pancreatic cancers), or smoking history (lung and pancreatic cancer). All statistical tests were 2-sided. Results The genetic predisposition index for chronic or aggressive periodontitis was statistically significantly associated with a 3% increased risk of colorectal cancer (per unit increase in genetic index of periodontal disease; P = .03), 3% increased risk of colon cancer (P = .02), 4% increased risk of proximal colon cancer (P = .01), and 3% increased risk of colorectal cancer among females (P = .04); however, it was not statistically significantly associated with the risk of lung cancer or pancreatic cancer, overall or within most subgroups. Conclusions Genetic predisposition to periodontitis may be associated with colorectal cancer risk. Further research should determine whether increased periodontitis prevention and increased cancer surveillance of patients with periodontitis is warranted.

Abstract Background Pulmonary embolism (PE) is a life-threatening cardiovascular condition. Studies showed that PE patients were associated with disorders of lipid metabolism and had higher triglyceride and lower HDL-C levels compared with healthy. We conducted the genome-wide association study to identify novel loci contributing to PE. Methods We conducted a large-scale GWAS of PE in 5,466 PE cases and 461,219 controls of European ancestry from the UK Biobank (466,685 participants total). We used genome-wide summary statistics to test for enrichment of functional annotations using ENRICHR. Example pathways included in Enrichr for testing include membership of genes in pathway databases such as the Kyoto Encyclopedia of Genes and Genomes (KEGG), Wikipathway, PANTHER, BioCarta or NCI-Nature pathways. We analyzed the pathways using combined score and p-values which were well validated by comparing to several methods. For pathway analyses, we considered a nominal P-value threshold of 0.05. Results We identified genome-wide significant genetic associations in 63 independent genetic loci for PE (P<5.0x10–7). Our findings for top pathways highlight that lipid metabolism (LIPC, LCAT, NPC2), caffeine metabolism (NAT2), and sudden cardiac death (ABCG8) related genetic loci play an important role in PE alongside genes already associated with coagulation-thrombosis pathway (VWF, THPO, PTPN11, INPP5D, UROS, HMBS) (all p-values p-values <0.05). Conclusion Our findings uncovered unexpected novel factors of PE etiology, suggesting novel mechanistic concepts of PE pathophysiology. Funding Acknowledgement Type of funding source: None

Abstract More Americans die every year of lung cancer than any other cancer. While the environmental risks for lung cancer are well understood, the genetic risk factors for this disease are not, even though it has been shown that lung cancer risk aggregates in families. Segregation analyses have confirmed the likelihood of rare, highly penetrant variants affecting lung cancer. Family studies offer a unique opportunity to identify such rare risk variants. We have previously identified a genome-wide significant risk locus for lung cancer using multipoint linkage analysis in highly aggregated lung cancer families. In this study, we performed targeted sequencing on this 6q23-6q27 region on 75 individuals from the 9 most strongly linked families. Parametric two-point linkage analysis using an autosomal dominant model with 10% penetrance for carriers and a 1% phenocopy rate and disease allele frequency of 1% was performed for each family. While we did not find any genome-wide significant results in any of the individual families, we did observe several interesting potential rare risk variants. First, we observed that the significant signal previously observed on 6q is recapitulated in these data but the location of the peak LOD is highly heterogeneous across families and it is likely that each family has a different risk gene. While most of the highest LOD scores were in noncoding variants, we did observe exonic variants with the highest LOD scores in two families (44 and 42). Both variants were rare, with a minor allele frequency (MAF) of approximately 1% and were in LPA and GPR126 respectively. Both these genes have been implicated somatically in lung cancer, however this is the first time that they have been implicated in the germline. While the other families have their highest LOD scores in noncoding variants, some of these genes are also good potential candidate genes including PARK2 (family 30), GRM1 (family 47), and PDE10A (family 102). All three genes have been implicated in lung cancer in some capacity, and this is the first time that GRM1 and PDE10A have been implicated as germline mutations. Again, all these variants are rare (MAF ⇐ 0.01) in the general population, which makes sense for a potential highly penetrant risk variant. In this study, we have elucidated that a previously identified risk locus on 6q25 is heterogeneous in the nine most strongly linked families, with different families appearing to be carrying different risk variants. Many of the top variants in each family are rare variants that are in good potential causal genes that have been identified for the first time here as germline risk variants for lung cancer. Further functional annotation is underway for these variants and some additional linkage analyses using other penetrance matrices may also be performed. Citation Format: Anthony M. Musolf, Claire L. Simpson, Bilal A. Moiz, Mariza de Andrade, Diptasri Mandal, Colette Gaba, Ping Yang, Ming You, Elena Y. Kupert, Marshall W. Anderson, Ann G. Schwartz, Susan M. Pinney, Christopher I. Amos, Joan E. Bailey-Wilson. Highly aggregated lung cancer families reveal a heterogeneous cause for a previous linkage signal on 6q [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 37.

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.

Abstract Background: Lung cancer remains the leading cause of cancer death worldwide, with 15% to 20% occurring in never smokers. To assess genetic determinants for prognosis among never smokers, we conducted a genome-wide investigation in the International Lung Cancer Consortium (ILCCO). Methods: Genomic and clinical data from 1,569 never-smoking patients with lung cancer of European ancestry from 10 ILCCO studies were included. HRs and 95% confidence intervals of overall survival were estimated. We assessed whether the associations were mediated through mRNA expression–based 1,553 normal lung tissues from the lung expression quantitative trait loci (eQTL) dataset and Genotype-Tissue Expression (GTEx). For cross-ethnicity generalization, we assessed the associations in a Japanese study (N = 887). Results: One locus at 13q22.2 was associated with lung adenocarcinoma survival at genome-wide level, with carriers of rs12875562-T allele exhibiting poor prognosis [HR = 1.71 (1.41–2.07), P = 3.60 × 10−8], and altered mRNA expression of LMO7DN in lung tissue (GTEx, P = 9.40 × 10−7; Lung eQTL dataset, P = 0.003). Furthermore, 2 of 11 independent loci that reached the suggestive significance level (P < 10−6) were significant eQTL affecting mRNA expression of nearby genes in lung tissues, including CAPZB at 1p36.13 and UBAC1 at 9q34.3. One locus encoding NWD2/KIAA1239 at 4p14 showed associations in both European [HR = 0.50 (0.38–0.66), P = 6.92 × 10−7] and Japanese populations [HR = 0.79 (0.67–0.94), P = 0.007]. Conclusions: Based on the largest genomic investigation on the lung cancer prognosis of never smokers to date, we observed that lung cancer prognosis is affected by inherited genetic variants. Impact: We identified one locus near LMO7DN at genome-wide level and several potential prognostic genes with cis-effect on mRNA expression. Further functional genomics work is required to understand their role in tumor progression.

Abstract Background: Lung cancer kills more people than any other cancer in the United States. In addition to environmental factors, lung cancer has genetic risk factors as well, though the genetic etiology is still not well understood. We have performed whole exome sequencing on 262 individuals from 28 extended families with a family history of lung cancer. Methods: Parametric genetic linkage analysis was performed on these samples using two distinct analyses—the lung cancer only (LCO) analysis, where only patients with lung cancer were coded as affected, and the all aggregated cancers (AAC) analysis, where other cancers seen in the pedigree were coded as affected. Results: The AAC analysis yielded a genome-wide significant result at rs61943670 in POLR3B at 12q23.3. POLR3B has been implicated somatically in lung cancer, but this germline finding is novel and is a significant expression quantitative trait locus in lung tissue. Interesting genome-wide suggestive haplotypes were also found within individual families, particularly near SSPO at 7p36.1 in one family and a large linked haplotype spanning 4q21.3-28.3 in a different family. The 4q haplotype contains potential causal rare variants in DSPP at 4q22.1 and PTPN13 at 4q21.3. Conclusions: Regions on 12q, 7p, and 4q are linked to increased cancer risk in highly aggregated lung cancer families, 12q across families and 7p and 4q within a single family. POLR3B, SSPO, DSPP, and PTPN13 are currently the best candidate genes. Impact: Functional work on these genes is planned for future studies and if confirmed would lead to potential biomarkers for risk in cancer.

Abstract Lung cancer (LC) is the primary cause of cancer-related deaths in the United States. Whereas smoking and other environmental factors strongly increase LC risk, multiple genetic variants also contribute to risk in smokers. Furthermore, among smokers, some families have been identified with an abnormally high prevalence of LC, suggesting that unknown genetic factors can greatly increase LC risk in smokers. The typically short survival after LC diagnosis impedes collection of detailed genotypic information on any single large family pedigree, impairing the identification of putative high-risk factors. Therefore, the Genetic Epidemiology of LC Consortium has collected epidemiological and genetic data from a number of families with high numbers of LC cases from eight different sites across the US. In this study, we have obtained whole exome sequences (WES) from 290 members of 28 families, including 66 LC cases. We used a gene-based approach to allow for the possibility that different families may contain different variants of the same gene. Variants were filtered for i) allele frequency, ii) functional effect using combined annotation-dependent depletion (CADD), and, iii) affecting a gene with either a known or suspected role in cancer. We further selected variants based on their segregation in family pedigrees in a pattern consistent with a large effect on LC risk. Candidate LC risk genes were then identified as those represented in at least two families by the same or different variants. We further culled the list of genes by requiring the presence of at least one rare, functional variant enriched in the WES of 1060 cases relative to 899 controls from the Transdisciplinary Research on Cancer of the Lung consortium. This analysis narrowed our results to two genes, one being E2A, a member of the E family of bHLH transcription factors. Whereas loss-of-function mutations in E2A drive lymphoid cancers, the E2A protein also participates in an oncogenic heterodimer with TWIST1 that promotes the epithelial-mesenchymal transition and is implicated in multiple cancer types. Furthermore, the E2A/TWIST1 heterodimer is the primary TWIST1-containing complex implicated in oncogenesis, and silencing of E2A in KRAS-mutant non-small cell lung cancer (NSCLC) cell lines results in oncogene-stimulated senescence and apoptosis. Our data identified three distinct E2A variants present in all 10 sequenced LC cases in the 5 families in which those variants are found. Finally, two of these E2A variants are located only 57 nucleotides from each other, immediately adjacent to sequences encoding a transcription activation domain, suggesting that both variants alter the same specific protein function. These data identify E2A variants as likely high risk factors for LC in smokers and validate our general approach for identifying genetic factors with a large impact on LC risk. Citation Format: Claudio W. Pikielny, Anthony M. Musolf, Mariza de Andrade, Diptasri Mandal, Colette Gaba, Ping Yang, yafang Li, Ming You, Richard Wilson, Elena Y. Kupert, Marshall W. Anderson, Ann G. Schwartz, Susan M. Pinney, Ambrose I. Granizo-Mackenzie, Yanhong Liu, Ramaswamy Govindan, James McKay, Rayjean Hung, John K. Field, David C. Christiani, Joan E. Bailey-Wilson, Christopher I. Amos. Familial studies identify variants in the E2A transcription factor as putative risk factors for lung 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 LB-053.

Abstract Lung cancer (LC) kills more people in the United States each year than any other cancer. While it is well known that a variety of environmental factors (particularly tobacco smoke) strongly increase the risk of LC, there are multiple associated genetic variants with small contributions to risk. High aggregation of LC within rare individual families suggest that there are high-risk genetic variants as well. However, these genetic risk factors for LC are under studied due to the rapid fatality of LC. We studied 28 highly aggregated extended high-risk familial lung cancer (HRFLC) families collected from eight different sites across the US. Whole exome sequencing was performed on 290 individuals from these families to identify potential risk variants for HRFLC using genetic linkage analysis. Quality control was performed on the sequence data, filtering on parameters such as read depth, genotype quality, missingness, and Mendelian inconsistencies. Identity-by-descent (IBD) values were also calculated to verify correct familial relationships. Quality control procedures left approximately 400,000 SNVs and indels for analysis.Parametric two-point linkage analysis was performed assuming an autosomal dominant mode of inheritance. Disease allele frequency was set to 1% with a penetrance of 80% for carriers and 1% phenocopy rate. While we did not identify any genome-wide significant variants across the 28 families, multiple suggestive variants were identified. The largest cluster of suggestive variants was located at 14q32 in the CATSPERB gene. Given the likely locus heterogeneity in LC (combined with the lack of power in some families), it is not surprising that none of the variants were significant across the families; looking at the individual family results proved more informative. Long haplotypes linked to LC risk were identified in multiple families. These long runs of positive linkages, which have little to no negative linkage evidence across them, are characteristic of true linkage signals in these types of analysis. Two of the most interesting linked regions were at 7p36.1 and 4q21.23-28.23. The 7p signal (observed in a single family) was genome-wide suggestive and located within the SSPO gene. SSPO has been implicated in breast and skin cancer (melanoma); it is a novel lung cancer signal. The 4q linkage (again observed in a single family) covers a large chunk of 4q and contains multiple potential candidate genes, however, the best candidate gene is PTPN13, a gene implicated in lung cancer but never in familial lung cancer. We are currently evaluating the individual family results of several other pedigrees and plan to perform additional analyses to confirm these linkages. Citation Format: Anthony Musolf, Haiming Sun, Bilal A. Moiz, Claudio W. Pikielny, Mariza de Andrade, Diptasri Mandal, Colette Gaba, Ping Yang, Yafang Li, Ming You, Richard K. Wilson, Elena Y. Kupert, Marshall W. Anderson, Ann G. Schwartz, Susan M. Pinney, Christopher I. Amos, Ramaswamy Govindan, Joan E. Bailey-Wilson. Familial lung cancer exhibits multiple novel linked haplotypes within pedigrees [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 4176.

Abstract BACKGROUND: Approximately 5% of gliomas occur in individuals with a family history of glioma, and first-degree relatives of brain tumor cases have a two-fold increase in risk of brain tumor. Recent somatic characterization has shown that tumors from familial cases are indistinguishable from sporadic cases, suggesting that familial cases may arise through similar mechanisms of gliomagenesis, and therefore may be associated with common variants as well as rare mutations. In this analysis, we assessed whether previously identified common risk variants are associated with familial glioma. METHODS: Data were obtained from the Glioma International Case Control (GICC) and Gliogene studies for 448 cases with reported family history, 4,405 cases without reported family history, and 3,288 controls. We assessed 25 risk loci previously identified by glioma GWAS, and odds ratios (OR) and 95% confidence intervals (95%CI) were calculated using an additive genetic logistic regression model adjusted for age, sex, and the first two principal components for familial cases versus unaffected controls, and non-familial cases versus controls. Results were considered significant at p<0.002 (Bonferroni correction for 25 tests). RESULTS: Significant associations were detected at 5/25 loci, including TERT, EGFR, CCDC26, CDKN2B, and RTEL1. The strongest association was at rs55705857 (CCDC26, OR=2.7, p=7.49x10-17). For GBM (222 familial cases), significant associations were detected at 6/26 loci (TERT, EGFR, CDKN2B, TP53 and RTEL1), while in non-GBM (205 familial cases) significant associations were detected at 3/25 loci (LRIG1, CCDC26, PHLDB1). These SNPs were further examined using a case-only approach comparing familial to non-familial cases, and there was no significant difference in allele frequencies by family history status. There was a strong correlation between log(OR) for familial cases only versus non-familial cases (adjusted R2=0.88). CONCLUSIONS: In this analysis we identified a significant association between familial glioma and five common risk loci previously identified by glioma GWAS. This provides further evidence of shared pathways of genetic risk and gliomagenesis between familial and non-familial glioma. Further exploration is necessary to determine the overall contribution of common genetic variation to risk of familial glioma. Citation Format: Quinn T. Ostrom, Georgina Armstrong, Christopher I. Amos, Jonine L. Bernstein, Elizabeth B. Claus, Jeanette E. Eckel-Passow, Dora Il'yasova, Christoffer Johansen, Daniel H. Lachance, Rose K. Lai, Ryan T. Merrell, Sara H. Olson, Joellen H. Schildkraut, Sanjay S. Shete, Richard S. Houlston, Robert B. Jenkins, Margaret R. Wrensch, Beatrice Melin, Jill S. Barnholtz-Sloan, Melissa L. Bondy. Previously identified common glioma risk SNPs are associated with familial glioma [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 4173.

Glioma incidence is highest in non‐Hispanic Whites, and to date, glioma genome‐wide association studies (GWAS) to date have only included European ancestry (EA) populations. African Americans and Hispanics in the US have varying proportions of EA, African (AA) and Native American ancestries (NAA). It is unknown if identified GWAS loci or increased EA is associated with increased glioma risk. We assessed whether EA was associated with glioma in African Americans and Hispanics. Data were obtained for 832 cases and 675 controls from the Glioma International Case–Control Study and GliomaSE Case–Control Study previously estimated to have <80% EA, or self‐identify as non‐White. We estimated global and local ancestry using fastStructure and RFMix, respectively, using 1,000 genomes project reference populations. Within groups with ≥40% AA (AFR ≥0.4 ), and ≥15% NAA (AMR ≥0.15 ), genome‐wide association between local EA and glioma was evaluated using logistic regression conditioned on global EA for all gliomas. We identified two regions (7q21.11, p = 6.36 × 10 −4 ; 11p11.12, p = 7.0 × 10 −4 ) associated with increased EA, and one associated with decreased EA (20p12.13, p = 0.0026) in AFR ≥0.4 . In addition, we identified a peak at rs1620291 ( p = 4.36 × 10 −6 ) in 7q21.3. Among AMR ≥0.15 , we found an association between increased EA in one region (12q24.21, p = 8.38 × 10 −4 ), and decreased EA in two regions (8q24.21, p = 0. 0010; 20q13.33, p = 6.36 × 10 −4 ). No other significant associations were identified. This analysis identified an association between glioma and two regions previously identified in EA populations (8q24.21, 20q13.33) and four novel regions (7q21.11, 11p11.12, 12q24.21 and 20p12.13). The identifications of novel association with EA suggest regions to target for future genetic association studies.