Thibodeau, Stephen N.

Abstract Over 50 genetic variants have been associated with colorectal cancer (CRC) risk through genome-wide association studies (GWAS), yet these variants represent only a fraction of the total estimated heritability. CRC is a heterogenous disease with diverse tumor etiology. Assessing genetic risk in molecular subtypes may help to identify novel loci and characterize genetic risk among tumor subtypes. We used microsatellite instability (MSI), an established CRC classifier with etiological and therapeutic relevance, to define CRC subtypes for GWAS analyses. We conducted a case-case analysis to estimate odds ratios (OR) and 95% confidence intervals (CI) for association of genome-wide variants with microsatellite stable (MSS) versus unstable (MSI) carcinomas. We ran an inverse-variance weighted fixed-effects meta-analysis across GWAS in a discovery set of 4,163 population-based CRC cases with harmonized microsatellite instability (MSI) marker and imputed genotype data. For each analysis, we used log-additive logistic regression, adjusting for age, sex, and principal components to account for population substructure. We then followed up with replication of 102 SNPs that reached p-values less than 5x10-6 in 1,698 cases. A total of 845 (20.3%) cancer cases were microsatellite unstable in the discovery population and 174 (10.2%) were unstable in the replication population. No variants reached the genome-wide significance level of 5x10-8 in the discovery set. However, we identified two variants that reached a Bonferroni corrected p-value of 4.0x10-4 in the replication set. This included one variant in MLH1 (Replication: OR=1.74, 95% CI=1.53-1.98, p=1.63x10-5; Discovery+Replication: OR=1.45, 95% CI=1.37-1.54, p=9.76x10-11) and one variant in LOC105377645 (Replication: OR=1.70, 95% CI=1.49-1.94, p=5.13x10-5; Discovery+Replication: OR=1.45, 95% CI=1.37-1.54, p=9.76 x 10-11). The MLH1 gene is a DNA mismatch repair gene implicated in Lynch Syndrome, the hallmark of which is microsatellite instability. This is the first genome-wide scan to identify a common variant in MLH1 that is associated with CRC. This variant (minor allele frequency, MAF = 23% in this all European ancestry population) is located in the 5'-untranslated region of MLH1 and is thought to act as a long-range regulator of DCLK3, a potential tumor driver gene. The second variant, located in LOC105377645 with an MAF of 22%, is in an uncharacterized region of the genome and has not previously been implicated in cancer development. These findings suggest that accounting for molecular heterogeneity is important for discovery and characterization of genetic variants associated with CRC risk. We plan to run polytomous regression analyses, increase our sample size, and further investigate CRC subtypes by CIMP, BRAF mutation, KRAS mutation status. Citation Format: Tabitha A. Harrison, Yiwen Lu, Chenjie Zeng, Flora Qu, Kristin Anderson, Hermann Brenner, Daniel D. Buchanan, Peter T. Campbell, Andrew T. Chan, Jenny Chang-Claude, Graham G. Giles, Bethany Van Guelpen, Michael Hoffmeister, Mark A. Jenkins, Noralane M. Lindor, Roger L. Milne, Polly A. Newcomb, Reiko Nishihara, Michael O. Woods, Shuji Ogino, John D. Potter, Martha L. Slattery, Wei Sun, Stephen N. Thibodeau, Li Hsu, Ulrike Peters. Genome-wide association study by colorectal carcinoma subtype [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 229.

AbstractMotivationThe use of single nucleotide polymorphism (SNP) interactions to predict complex diseases is getting more attention during the past decade, but related statistical methods are still immature. We previously proposed the SNP Interaction Pattern Identifier (SIPI) approach to evaluate 45 SNP interaction patterns/patterns. SIPI is statistically powerful but suffers from a large computation burden. For large-scale studies, it is necessary to use a powerful and computation-efficient method. The objective of this study is to develop an evidence-based mini-version of SIPI as the screening tool or solitary use and to evaluate the impact of inheritance mode and model structure on detecting SNP–SNP interactions.ResultsWe tested two candidate approaches: the ‘Five-Full’ and ‘AA9int’ method. The Five-Full approach is composed of the five full interaction models considering three inheritance modes (additive, dominant and recessive). The AA9int approach is composed of nine interaction models by considering non-hierarchical model structure and the additive mode. Our simulation results show that AA9int has similar statistical power compared to SIPI and is superior to the Five-Full approach, and the impact of the non-hierarchical model structure is greater than that of the inheritance mode in detecting SNP–SNP interactions. In summary, it is recommended that AA9int is a powerful tool to be used either alone or as the screening stage of a two-stage approach (AA9int+SIPI) for detecting SNP–SNP interactions in large-scale studies.Availability and implementationThe ‘AA9int’ and ‘parAA9int’ functions (standard and parallel computing version) are added in the SIPI R package, which is freely available at https://linhuiyi.github.io/LinHY_Software/.Supplementary informationSupplementary data are available at Bioinformatics online.

Abstract Background Previous genome-wide association studies (GWAS) have identified 42 loci (P < 5 × 10−8) associated with risk of colorectal cancer (CRC). Expanded consortium efforts facilitating the discovery of additional susceptibility loci may capture unexplained familial risk. Methods We conducted a GWAS in European descent CRC cases and control subjects using a discovery–replication design, followed by examination of novel findings in a multiethnic sample (cumulative n = 163 315). In the discovery stage (36 948 case subjects/30 864 control subjects), we identified genetic variants with a minor allele frequency of 1% or greater associated with risk of CRC using logistic regression followed by a fixed-effects inverse variance weighted meta-analysis. All novel independent variants reaching genome-wide statistical significance (two-sided P < 5 × 10−8) were tested for replication in separate European ancestry samples (12 952 case subjects/48 383 control subjects). Next, we examined the generalizability of discovered variants in East Asians, African Americans, and Hispanics (12 085 case subjects/22 083 control subjects). Finally, we examined the contributions of novel risk variants to familial relative risk and examined the prediction capabilities of a polygenic risk score. All statistical tests were two-sided. Results The discovery GWAS identified 11 variants associated with CRC at P < 5 × 10−8, of which nine (at 4q22.2/5p15.33/5p13.1/6p21.31/6p12.1/10q11.23/12q24.21/16q24.1/20q13.13) independently replicated at a P value of less than .05. Multiethnic follow-up supported the generalizability of discovery findings. These results demonstrated a 14.7% increase in familial relative risk explained by common risk alleles from 10.3% (95% confidence interval [CI] = 7.9% to 13.7%; known variants) to 11.9% (95% CI = 9.2% to 15.5%; known and novel variants). A polygenic risk score identified 4.3% of the population at an odds ratio for developing CRC of at least 2.0. Conclusions This study provides insight into the architecture of common genetic variation contributing to CRC etiology and improves risk prediction for individualized screening.

We examined the Institutional Review Board (IRB) process at 9 academic institutions in the electronic Medical Records and Genomics (eMERGE) Network, for proposed electronic health record-based genomic medicine studies, to identify common questions and concerns. Sequencing of 109 disease related genes and genotyping of 14 actionable variants is being performed in ~28,100 participants from the 9 sites. Pathogenic/likely pathogenic variants in actionable genes are being returned to study participants. We examined each site’s research protocols, informed-consent materials, and interactions with IRB staff. Research staff at each site completed questionnaires regarding their IRB interactions. The time to prepare protocols for IRB submission, number of revisions and time to approval ranged from 10–261 days, 0–11, and 11–90 days, respectively. IRB recommendations related to the readability of informed consent materials, specifying the full range of potential risks, providing options for receiving limited results or withdrawal, sharing of information with family members, and establishing the mechanisms to answer participant questions. IRBs reviewing studies that involve the return of results from genomic sequencing have a diverse array of concerns, and anticipating these concerns can help investigators to more effectively engage IRBs.

Abstract Motivation Interpreting genetic variation in noncoding regions of the genome is an important challenge for personal genome analysis. One mechanism by which noncoding single nucleotide variants (SNVs) influence downstream phenotypes is through the regulation of gene expression. Methods to predict whether or not individual SNVs are likely to regulate gene expression would aid interpretation of variants of unknown significance identified in whole-genome sequencing studies. Results We developed FIRE (Functional Inference of Regulators of Expression), a tool to score both noncoding and coding SNVs based on their potential to regulate the expression levels of nearby genes. FIRE consists of 23 random forests trained to recognize SNVs in cis-expression quantitative trait loci (cis-eQTLs) using a set of 92 genomic annotations as predictive features. FIRE scores discriminate cis-eQTL SNVs from non-eQTL SNVs in the training set with a cross-validated area under the receiver operating characteristic curve (AUC) of 0.807, and discriminate cis-eQTL SNVs shared across six populations of different ancestry from non-eQTL SNVs with an AUC of 0.939. FIRE scores are also predictive of cis-eQTL SNVs across a variety of tissue types. Availability and implementation FIRE scores for genome-wide SNVs in hg19/GRCh37 are available for download at https://sites.google.com/site/fireregulatoryvariation/. Supplementary information Supplementary data are available at Bioinformatics online.

Abstract Background: The increased number of T-lymphocytes residing in the tumors of colorectal cancer (CRC) patients is consistently shown to predict better survival of CRC patients independent of stage. In addition to tumor itself, T-lymphocytes also infiltrate normal tissues adjacent to tumors; however, little is known about how the inter-individual T-cell variability in tumor-adjacent tissues affects CRC prognosis. Our goal was to characterize total T-cells in colorectal tumor and tumor-adjacent tissues and study their associations with all-cause and CRC-specific survival in the Iowa Women’s Health Study. Methods: We constructed tissue microarrays and quantified CD3 antibody staining (Clone F.7.238; Dako), an established marker for total T-cells, in paraffin-embedded tissue samples from 463 women diagnosed with incident CRC from 1986-2002 (mean age at diagnosis was 74 years). Up to 3 tumor and 3 tumor-adjacent cores were immunostained for each person. An experienced pathologist quantified CD3+ T cells in each area using 4 categories: non-detected, mild (1-10 cells per 0.28 mm2); moderate (11-29 cells per 0.28 mm2); and strong infiltration (≥30 cells per 0.28 mm2). The obtained scores were averaged over all quantified cores for each person. Cox regression was used to estimate the hazard ratio (HR) and 95% CI for all-cause and CRC mortality in relation to (1) tumor score; (2) tumor-adjacent score; and (3) the ratio of tumor to tumor-adjacent score, hereafter called Score ratio. All scores were categorized in quartiles. The final model adjusted for age at diagnosis, SEER stage, tumor grade, body mass and smoking history. Results: During follow-up for a maximum of 25 years, 67% of participants diagnosed with CRC died (~30% died from CRC). There was a weak but significant correlation between CD3 tumor and tumor-adjacent scores (Spearman coefficient: r=0.15, p=0.02). CD3 tumor score was inversely associated with stage and was higher in those with proximal colon cancer, and for MSI-high, CIMP-high and BRAF-mutated tumors. The HRs (95%CI) for the highest versus lowest category of CD3 tumor score were 0.61 (0.44-0.84) (p-trend=0.003) for all-cause mortality and 0.27 (0.15-0.48) (p-trend &lt;0.0001) for CRC-specific mortality. The tumor-adjacent score was not associated with any characteristics or death. The associations between the Score ratio and all-cause and CRC survival mirrored those for the tumor score but were weaker. The HRs (95%CI) for the highest versus lowest quartile of the Score ratio were 0.70 (0.45-1.09) (p-trend=0.09) for all-cause and 0.45 (0.23-0.87) (p-trend=0.003) for CRC-specific mortality. Conclusions. Based on data from prospectively identified CRC cases in the IWHS cohort, colorectal tumor T-cell infiltration is associated with improved survival. Further investigation is needed to determine the T-cell subtypes that are most predictive of CRC survival outcomes. Citation Format: Anna E. Prizment, Robert A. Vierkant, Thomas C. Smyrk, Lori S. Tillmans, Heather H. Nelson, Charles F. Lynch, Stephen N. Thibodeau, Timothy R. Church, James R. Cerhan, Kristin E. Anderson, Paul J. Limburg. Associations between total T-lymphocytes and colorectal cancer survival in a prospective cohort study of older women [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 2247. doi:10.1158/1538-7445.AM2017-2247

Abstract The Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO) in collaboration with the Colorectal Cancer Family Registry (CCFR) aims to identify genetic variants and environmental risk factors that impact colorectal cancer (CRC). Over 30 studies from North America, Europe, and Australia participate in GECCO. These studies have collected clinical, epidemiological, and survival data, as well as blood and tumor biospecimens, for over 80,000 CRC cases and controls. The current study aims to conduct targeted deep sequencing of tumors and matching normal DNA to identify recurrent and novel somatic and germline variants in 4,200 CRC cases. To achieve this goal, an AmpliSeq targeted sequencing panel of 1.12 Mbp was constructed to sequence the coding regions of 190 significantly mutated genes identified from whole exome sequencing datasets generated by the Nurses’ Health Study and Health Professional’s Follow-up Study, and The Cancer Genome Atlas. The panel also covers coding regions of 15 genes with germline high penetrance mutations in CRC, 54 regions to detect CRC-related copy number alterations (CNAs), and microsatellite and homopolymer repeat regions to identify defective DNA mismatch repair. Primers were also included to detect Fusobacterium nucleatum DNA in tumor biopsies, as F. nucleatum is thought to promote CRC carcinogenesis. Sequencing of the DNA libraries on Illumina HiSeq 2500 produced a mean coverage of greater than 500X for tumor DNA and 100X for normal DNA, with &gt;85% of the bases covered at the target at 50x. So far, targeted sequencing of &gt;1,500 DNA samples from CRC tumors and normal tissues has identified recurrent and novel somatic mutations, germline genetic variants, and hypermutation status of the tumors due to defective DNA mismatch repair or pathogenic mutations in the POLE gene. Targeted sequencing has also allowed quantification of the F. nucleatum DNA in tumor biopsies; the results were validated by a multiplex QPCR assay. At the AACR annual meeting, we will present targeted sequencing results generated from the first two GECCO-participating studies (n=1,300 cases). These data will be valuable for future association testing of somatic mutations, CNAs, hypermutation status, and F. nucleatum with germline genetic variants, lifestyle, and environmental risk factors and survival. This large study will allow development of better strategies for diagnosis, treatment, and prevention of CRC. Citation Format: Syed H. Zaidi, Wei Sun, Jeroen Huyghe, Catherine S. Grasso, Quang Trinh, Charles Connolly, Amy French, Jasmine Mu, Marios Giannakis, Eve Shinbrot, Ivan Borozan, Michael J. Quist, Hermann Brenner, Daniel Buchanan, Peter Campbell, Andrew Chan, Jenny Chang-Claude, Vincent Ferretti, Charles Fuchs, Andrea Gsur, Marc Gunter, Tabitha Harrison, Michael Hoffmeister, Wen-Yi Huang, Paul Krzyzanowski, Stephen Lee, Mathieu Lemire, Jessica Miller, Danielle Pasternack, Cherie Teney, Elaine Mardis, Polly Newcomb, Lincoln Stein, Lee Timms, David Wheeler, Christina Yung, Niha Zubair, Levi Garraway, Shuji Ogino, Li Hsu, Steven Gallinger, Stephen Thibodeau, Thomas Hudson, Ulrike Peters. Targeted deep sequencing of colorectal tumor tissues to study associations of tumor subtypes with germline genetic, lifestyle, and environmental risk factors [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 1286. doi:10.1158/1538-7445.AM2017-1286