Robust Tests for Additive Gene-Environment Interaction in Case-Control Studies Using Gene-Environment Independence

There have been recent proposals advocating the use of additive gene-environment interaction instead of the widely used multiplicative scale, as a more relevant public health measure. Using gene-environment independence enhances statistical power for testing multiplicative interaction in case-control studies. However, under departure from this assumption, substantial bias in the estimates and inflated type I error in the corresponding tests can occur. In this paper, we extend the empirical Bayes (EB) approach previously developed for multiplicative interaction, which trades off between bias and efficiency in a data-adaptive way, to the additive scale. An EB estimator of the relative excess risk due to interaction is derived, and the corresponding Wald test is proposed with a general regression setting under a retrospective likelihood framework. We study the impact of gene-environment association on the resultant test with case-control data. Our simulation studies suggest that the EB approach uses the gene-environment independence assumption in a data-adaptive way and provides a gain in power compared with the standard logistic regression analysis and better control of type I error when compared with the analysis assuming gene-environment independence. We illustrate the methods with data from the Ovarian Cancer Association Consortium.

[1]  Nilanjan Chatterjee,et al.  An exposure‐weighted score test for genetic associations integrating environmental risk factors , 2015, Biometrics.

[2]  D. Cramer,et al.  Androgen receptor cytosine, adenine, guanine repeats, and haplotypes in relation to ovarian cancer risk. , 2005, Cancer research.

[3]  Nilanjan Chatterjee,et al.  Semiparametric maximum likelihood estimation exploiting gene-environment independence in case-control studies , 2005 .

[4]  P. Hartge,et al.  Multiple births and risk of epithelial ovarian cancer. , 2000, Journal of the National Cancer Institute.

[5]  Carolyn Hutter,et al.  Powerful Cocktail Methods for Detecting Genome‐Wide Gene‐Environment Interaction , 2012, Genetic epidemiology.

[6]  Nilanjan Chatterjee,et al.  Likelihood ratio test for detecting gene (G)-environment (E) interactions under an additive risk model exploiting G-E independence for case-control data. , 2012, American journal of epidemiology.

[7]  S. Olson,et al.  Phytoestrogen consumption from foods and supplements and epithelial ovarian cancer risk: a population-based case control study , 2011, BMC women's health.

[8]  Kconfab Investigators,et al.  Identification of six new susceptibility loci for invasive epithelial ovarian cancer , 2015 .

[9]  Herbert Yu,et al.  Detectable Symptomatology Preceding the Diagnosis of Pancreatic Cancer and Absolute Risk of Pancreatic Cancer Diagnosis. , 2015, American journal of epidemiology.

[10]  Eric J Tchetgen Tchetgen,et al.  A general approach to detect gene (G)-environment (E) additive interaction leveraging G-E independence in case-control studies. , 2019, Statistics in medicine.

[11]  Raymond J Carroll,et al.  Shrinkage Estimators for Robust and Efficient Inference in Haplotype-Based Case-Control Studies , 2009, Journal of the American Statistical Association.

[12]  C. Carlson,et al.  No Evidence of Gene–Calcium Interactions from Genome-Wide Analysis of Colorectal Cancer Risk , 2014, Cancer Epidemiology, Biomarkers & Prevention.

[13]  Sebastian M. Armasu,et al.  Inherited Determinants of Ovarian Cancer Survival , 2010, Clinical Cancer Research.

[14]  M. Pike,et al.  Association between endometriosis and risk of histological subtypes of ovarian cancer: a pooled analysis of case–control studies , 2012, The Lancet. Oncology.

[15]  D. Thomas,et al.  Methods for investigating gene-environment interactions in candidate pathway and genome-wide association studies. , 2010, Annual review of public health.

[16]  M. Pike,et al.  Hormonal factors and the risk of invasive ovarian cancer: a population-based case-control study. , 2004, Fertility and sterility.

[17]  Yurii B. Shvetsov,et al.  Identification of six new susceptibility loci for invasive epithelial ovarian cancer , 2015, Nature Genetics.

[18]  Jaeil Ahn,et al.  Tests for gene‐environment interaction from case‐control data: a novel study of type I error, power and designs , 2008, Genetic epidemiology.

[19]  J. Vandenbroucke,et al.  Practice of Epidemiology What Do Case-Control Studies Estimate? Survey of Methods and Assumptions in Published Case-Control Research , 2008 .

[20]  C. Carlson,et al.  Genome-Wide Diet-Gene Interaction Analyses for Risk of Colorectal Cancer , 2014, PLoS genetics.

[21]  S Greenland,et al.  Concepts of interaction. , 1980, American journal of epidemiology.

[22]  Jeffrey R. Marks,et al.  Association between DNA Damage Response and Repair Genes and Risk of Invasive Serous Ovarian Cancer , 2010, PloS one.

[23]  Brooke L. Fridley,et al.  GWAS meta-analysis and replication identifies three new susceptibility loci for ovarian cancer , 2013, Nature Genetics.

[24]  Stephen Gruber,et al.  Accounting for error due to misclassification of exposures in case–control studies of gene–environment interaction , 2008, Statistics in medicine.

[25]  Matthew Burnell,et al.  Recruitment of newly diagnosed ovarian cancer patients proved challenging in a multicentre biobanking study. , 2011, Journal of clinical epidemiology.

[26]  Peter Kraft,et al.  Additive interactions between susceptibility single-nucleotide polymorphisms identified in genome-wide association studies and breast cancer risk factors in the Breast and Prostate Cancer Cohort Consortium. , 2014, American journal of epidemiology.

[27]  Wei Lu,et al.  Multiple independent variants at the TERT locus are associated with telomere length and risks of breast and ovarian cancer , 2013, Nature Genetics.

[28]  W. Chung,et al.  Genome-Wide Association Study in BRCA1 Mutation Carriers Identifies Novel Loci Associated with Breast and Ovarian Cancer Risk , 2013, PLoS genetics.

[29]  Mathieu Lemire,et al.  Material Supplementary Cited Articles E-mail Alerts Characterization of Gene–environment Interactions for Colorectal Cancer Susceptibility Loci American Association for Cancer Research , 2022 .

[30]  Sebastian M. Armasu,et al.  Identification and molecular characterization of a new ovarian cancer susceptibility locus at 17 q 21 . 31 , 2013 .

[31]  R. Sankila,et al.  [Cancer incidence in the first-degree relatives of ovarian cancer patients]. , 1997, Duodecim; laaketieteellinen aikakauskirja.

[32]  Jaeil Ahn,et al.  Tests for Gene-Environment Interactions and Joint Effects With Exposure Misclassification. , 2016, American journal of epidemiology.

[33]  Nilanjan Chatterjee,et al.  Common genetic polymorphisms modify the effect of smoking on absolute risk of bladder cancer. , 2013, Cancer research.

[34]  H. Risch,et al.  PGR +331 A/G and Increased Risk of Epithelial Ovarian Cancer , 2006, Cancer Epidemiology Biomarkers & Prevention.

[35]  D. Conti,et al.  Efficient Two‐Step Testing of Gene‐Gene Interactions in Genome‐Wide Association Studies , 2013, Genetic epidemiology.

[36]  H. Leeb,et al.  Sparse Estimators and the Oracle Property, or the Return of Hodges' Estimator , 2007, 0704.1466.

[37]  Jack A. Taylor,et al.  Non-hierarchical logistic models and case-only designs for assessing susceptibility in population-based case-control studies. , 1994, Statistics in medicine.

[38]  B. Thomsen,et al.  Hormone therapy and the impact of estrogen intake on the risk of ovarian cancer. , 2004, Archives of internal medicine.

[39]  A. Whittemore,et al.  Identification and molecular characterization of a new ovarian cancer susceptibility locus at 17q21.31 , 2013, Nature Communications.

[40]  N. Weiss,et al.  Menopausal Hormone Therapy and Risk of Epithelial Ovarian Cancer , 2007, Cancer Epidemiology Biomarkers & Prevention.

[41]  M. Widschwendter,et al.  Tubal ligation and the risk of ovarian cancer: review and meta-analysis. , 2011, Human reproduction update.

[42]  M. Goodman,et al.  Association of two common single-nucleotide polymorphisms in the CYP19A1 locus and ovarian cancer risk. , 2008, Endocrine-related cancer.

[43]  Guang Yong Zou,et al.  On the estimation of additive interaction by use of the four-by-two table and beyond. , 2008, American journal of epidemiology.

[44]  A. Whittemore,et al.  A genome-wide association study identifies a new ovarian cancer susceptibility locus on 9p22.2 , 2009, Nature Genetics.

[45]  D. Seminara,et al.  Cancer risk estimates for family members of a population-based family registry for breast and ovarian cancer. , 2000, Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology.

[46]  A. Whittemore,et al.  A genome-wide association study identifies susceptibility loci for ovarian cancer at 2q31 and 8q24 , 2010, Nature Genetics.

[47]  Bhramar Mukherjee,et al.  Exploiting Gene‐Environment Independence for Analysis of Case–Control Studies: An Empirical Bayes‐Type Shrinkage Estimator to Trade‐Off between Bias and Efficiency , 2008, Biometrics.

[48]  David N. Rider,et al.  Genetic variation in the one-carbon transfer pathway and ovarian cancer risk. , 2008, Cancer research.

[49]  Harvey Risch,et al.  Combined and interactive effects of environmental and GWAS-identified risk factors in ovarian cancer. , 2013, Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology.

[50]  Susan Halabi,et al.  Cyclin E Overexpression in Epithelial Ovarian Cancer Characterizes an Etiologic Subgroup , 2008, Cancer Epidemiology Biomarkers & Prevention.

[51]  T. VanderWeele.  A word and that to which it once referred: assessing "biologic" interaction. , 2011, Epidemiology.

[52]  R. Luben,et al.  Cumulative effects and predictive value of common obesity-susceptibility variants identified by genome-wide association studies. , 2010, The American journal of clinical nutrition.

[53]  A. Green,et al.  Talcum powder, chronic pelvic inflammation and NSAIDs in relation to risk of epithelial ovarian cancer , 2008, International journal of cancer.

[54]  R. Pyke,et al.  Logistic disease incidence models and case-control studies , 1979 .

[55]  L. Sobin,et al.  Reproductive factors and epithelial ovarian cancer risk by histologic type: a multiethnic case-control study. , 2003, American journal of epidemiology.

[56]  M. Pike,et al.  Markers of inflammation and risk of ovarian cancer in Los Angeles County , 2009, International journal of cancer.

[57]  W. Gauderman,et al.  Gene-environment interaction in genome-wide association studies. , 2008, American journal of epidemiology.

[58]  T J VanderWeele,et al.  Sample Size and Power Calculations for Additive Interactions , 2012, Epidemiologic methods.

[59]  A. Whittemore,et al.  Common variants at 19p13 are associated with susceptibility to ovarian cancer , 2010, Nature Genetics.

[60]  C R Weinberg,et al.  Designing and analysing case-control studies to exploit independence of genotype and exposure. , 1997, Statistics in medicine.

[61]  S Lemeshow,et al.  Confidence interval estimation of interaction. , 1992, Epidemiology.

[62]  Tyler J. VanderWeele,et al.  Explanation in Causal Inference: Methods for Mediation and Interaction , 2015 .

[63]  J. Chang-Claude,et al.  Low‐dose oral contraceptives: Protective effect on ovarian cancer risk , 2001, International journal of cancer.

[64]  Jaeil Ahn,et al.  Testing gene-environment interaction in large-scale case-control association studies: possible choices and comparisons. , 2012, American journal of epidemiology.

[65]  Collaborative Group on Hormonal Factors in Breast Cancer,et al.  Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23 257 women with ovarian cancer and 87 303 controls , 2008, The Lancet.

[66]  R. Ness,et al.  Contraception methods, beyond oral contraceptives and tubal ligation, and risk of ovarian cancer. , 2011, Annals of epidemiology.