BA6 - Inflammation Genes, Diet, Medication, and Cancer
This page provides study documentation for BA6. For description of the specimen results, see Specimen Results Description (open to public). Data sets of the specimen results are included in the existing WHI datasets located on the WHI Data on this site (sign in and a completed Data Distribution Agreement are required; see details on the Data site).
Investigator Names and Contact Information
Jianfeng Xu, M.D., Dr. P.H., Wake Forest University School of Medicine, Winston-Salem, NC
Introduction/Intent
The study seeks to identify genetic variants in genes involved in inflammation and immunity which are associated with the risk of cancer (breast, colon and rectum, and lung) in European Americans and African Americans. The research team will test associations between dietary, supplement and non steroidal anti-inflammatory use (NSAID) with inflammatory markers and risk of overall cancer.
They will then study interaction effects of genetic variants with dietary, supplement, and NSAID exposure on cancer risk.
An association between inflammation and cancer has been appreciated for over a century. Two hypotheses have been proposed concerning the role of the immune system in tumorigenesis. In the tumor surveillance hypothesis, tumor-infiltrating leukocytes control tumor growth and eradicate newly forming tumors, and failure of this surveillance permits tumor establishment. In contrast to the immune surveillance hypothesis, there is abundant evidence emerging suggesting that immune-mediated inflammation promotes tumorigenesis. Immune-mediated inflammation is thought to contribute to tumorigenesis by a variety of mechanisms: 1) induction of hyperplasia by either release of growth factors or disruption of tissue homeostasis by destruction of a large portion of the tissue; 2) release of reactive oxygen species (ROS) resulting in DNA damage-mutation; 3) promoting vascularization as part of “normal” tissue repair processes; 4) releasing of anti-inflammatory mediators by innate immune cells to suppress the tumor surveillance system.
Because immune cells and soluble mediators can infiltrate a wide variety of tissues, inflammation has the potential to contribute to the initiation and progression of a wide array of cancers through above mentioned mechanisms. We have therefore hypothesized that chronic inflammation is a common contributor to the development of a wide array of cancers and that genetic polymorphisms and environmental exposures that promote chronic inflammation are risk factors for the development of cancer in general.
Hypothesis: Genetic variants in genes involved in inflammation and their interactions with dietary, supplement, and medication exposures modulate inflammatory responses and affect overall cancer risk
Overall objectives: Taking advantage of DNA samples from a large number of patients with various types of cancer patients in the WHI study, we will first identify genetic variants in genes involved in inflammation that are associated with overall cancer risk using a novel study design and innovative high-throughput technology. By further taking advantage of the unique collection of data on dietary, supplement, and medication exposures as well as serum samples collected prior to cancer diagnosis among cancer patients in the WHI study, we will then identify interaction effects on overall cancer risk between identified genetic variants and environmental exposures. In addition, our study design makes it possible to assess associations of genetic variants and their interactions with environmental exposures on overall cancer risk in African Americans. Our study not only will contribute to our understanding on the etiology of cancer, but also will provide important guidance in the prevention of cancer.
Specific Aims:
1. To test for main effects of genetic variants in ~1,000 genes involved in innate immunity and inflammatory responses on overall cancer risk.
a) To identify associations of SNPs in the inflammatory genes and overall cancer risk in European Americans. About 9,000 SNPs in ~1,000 genes involved in immunity and inflammation will be genotyped using a ParAllele Immunity and Inflammation array among 1,500 European American cancer cases (500 cases each from colorectal, lung, and breast cancer patients) and 1,500 frequency matched controls. Association tests will be performed in each type of cancer. SNPs that are significant (P = 0.1) and in the same direction of association among colorectal, lung, and breast cancer will be identified as overall cancer risk variants. With this unique study design, only ~2 out of ~9,000 tested SNPs are expected to be significant by chance. On the other hand, we have = 88% power to detect risk SNPs (MAF = 15%) that confer modest overall cancer risk (OR = 1.5). For the genes that are implicated in overall cancer risk, fine mapping association tests will be performed to identify the strongest associated SNPs (or haplotype) in each of these genes using tagging and functional SNPs approach.
b) To test for associations between the implicated genes and overall cancer risk in African Americans. The inflammatory genes that confer overall cancer risk identified in European Americans will be tested among African Americans, with a total of 500 cases from breast, colorectal, lung cancer patients and 500 frequency matched controls. Race-specific tagging SNPs and functional SNPs within each of these genes will be tested.
2. To test for interaction effects of genetic variants with dietary, supplement, and NSAID exposures on overall cancer risk.
a) To identify dietary and supplement patterns that are related to inflammatory biomarkers using a reduced rank regression (RRR) method. Various food and supplement intakes estimated from the WHI FFQ and two inflammatory biomarkers (CRP and TNFa) among all 2,000 cancer cases and 2,000 cancer-free controls will be used to identify these patterns.
b) To test for main effects of each environmental exposure of interest, including dietary patterns, supplement patterns, NSAIDs, and inflammatory biomarkers on overall cancer risk. Association tests will be performed for each of the exposure using unconditional logistic regression and adjusted for total energy and other covariates.
c) To test for interaction effects on overall cancer risk among genetic variants and environmental exposures that have demonstrated main effects. Model building in a logistic regression framework will be used to test for their joint effects. The analyses will be performed among all 2,000 cases and 2,000 controls and stratified by race.
d) To identify higher orders of gene-gene and gene-environmental interaction effects on overall cancer risk. This analysis will be performed among 1,500 European American cases and 1,500 frequency matched controls where information about ~9,000 SNPs as well as dietary and supplement patterns and NSAIDs are available. The analysis will be performed using a classification and regression tree method (CART) and Logic Regression (LG).