Skip to Main Content

An official website of the United States government

Principal Investigator
Name
Meredith Shiels
Degrees
-
Institution
NCI
Position Title
-
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
2010-0201
Initial CDAS Request Approval
Nov 18, 2010
Title
Inflammatory Markers and Smoking-Associated Cancers
Summary
A causal association has been established between tobacco smoking and several cancers, including those arising from the lung, head and neck, esophagus, stomach, liver, pancreas, kidney, urinary bladder, and uterine cervix, as well as myeloid leukemia (1). In addition to having a direct carcinogenic effect, the cancer-causing effects of tobacco may be mediated through inflammation. Additionally, inflammation is likely to arise from non-smoking factors. Inflammatory conditions are known to be associated with increased risk of smoking-associated cancers (2). Inflammation-related genes and anti-inflammatory drugs have also been associated with specific smoking-associated cancers. However, few studies have examined the association between circulating inflammatory markers and smoking-associated cancers. Therefore, the precise inflammatory pathways or biomarkers involved in the development of smoking-associated cancers are as yet unknown. Furthermore, because these cancers share heterogeneous associations with smoking and inflammation, evaluating them together will allow the identification of overlapping and distinct mechanisms of inflammatory pathways involved in carcinogenesis. We propose to comprehensively evaluate the role of chronic inflammation in head and neck, stomach, pancreas, kidney, and urinary bladder cancers by measuring a wide range of systemic inflammatory markers using a case-cohort design. Cancers of the lung, liver and cervix and myeloid leukemia will be excluded from this analysis due to pre-existing studies or a limited number of cases. In 1,327 cases and a sub-cohort of 500 controls from the PLCO trial's screening arm, we will measure circulating levels of 60-65 markers, including pro-inflammatory cytokines, anti-inflammatory cytokines, chemokines, growth factors, soluble receptor markers and angiogenesis factors. We will investigate the association of these inflammation markers with risk of each smoking-associated cancer individually and evaluate heterogeneity in the associations across cancer sites. In addition to providing etiologic insight for each of these cancers, characterizing inflammatory biomarkers associated with increased cancer risk could aid in identifying high risk populations.
Aims

1. To evaluate the association between levels of each of 60-65 inflammatory markers with individual smoking-associated cancers (i.e., cancers of the head and neck, stomach, pancreas, kidney, and urinary bladder). We will measure levels of 60-65 systemic inflammatory markers, including pro-inflammatory cytokines, anti-inflammatory cytokines, chemokines, soluble receptor markers, growth factors and angiogenesis markers in plasma samples. We hypothesize that increased levels of pro-inflammatory cytokines, growth factors and angiogenesis markers and decreased levels of anti-inflammatory cytokines will be associated with increased risk of individual smoking-associated cancers. Through this aim, we will identify specific inflammatory markers that are associated with risk of individual smoking-related cancers. Additionally, we will identify a common set of markers that are predictive of risk for each of the smoking-related cancer types. 2. To evaluate the association between combinations of inflammatory markers with individual smoking-associated cancers (i.e., cancers of the head and neck, stomach, pancreas, kidney and urinary bladder). We will group markers into categories and evaluate associations with risk of each individual smoking-associated cancer. Grouping will help to reduce the 60-65 markers to a more statistically manageable set that also relates to biologic function. Markers will be grouped into categories using two strategies: 1) a priori, based on biologic function (e.g. pro anti-inflammatory cytokines, chemokines, soluble receptor markers, growth factors and angiogenesis markers) and 2) empirically, based on correlation structures using factor analysis among control subjects. We hypothesize that combinations of pro-inflammatory cytokines, growth factors and angiogenesis markers will be associated with increased risk of smoking-associated cancers and combinations of anti-inflammatory cytokines will be associated with decreased risk of smoking-associated cancers. Through this aim, we will identify inflammatory pathways that are associated with risk of smoking-related cancers. 3. To evaluate the impact of smoking on the association of inflammatory markers with risk of each smoking-related cancer. Through this aim, we will identify smoking-mediated vs. smoking-independent components of inflammation that are associated with increased cancer risk. To demonstrate potential mediation, we will: 1) Evaluate associations of inflammatory markers (and combinations of markers) stratified by smoking status (e.g., never, former and current smokers) and 2) Evaluate the association of inflammatory markers with cancer risk unadjusted and adjusted for smoking and calculate a mediation proportion, which quantifies the degree to which smoking explains the association. Evidence of mediation will be present if (1) smoking increases levels of certain inflammatory markers and elevations in these markers are associated with increased cancer risk, as well as (2) showing that adjustment for smoking significantly reduces the association of inflammatory markers with each smoking-related cancer. We hypothesize that both smoking-mediated and smoking-independent mechanisms will exist and be associated with increased cancer risk and that, among smokers, there is a common set of smoking-induced inflammatory markers that increase cancer risk across smoking-associated cancer sites. Secondary Aims: 1. To examine the associations between inflammation-associated genetic variants and inflammatory markers in individuals with available genome wide association study (GWAS) data.GWAS data will be available on a subset of our participants. We will be able to examine the association between inflammation-associated genes and inflammatory markers. 2. To conduct more detailed analyses of the associations between inflammatory markers and each individual smoking-associated cancer site. We will investigate the associations between inflammatory markers and smoking-associated cancers by exploring effect modification of the inflammatory marker and cancer association by known cancer risk factors such as body mass index, alcohol use, diet, NSAID use, and Helictobacter pylori (H. pylori) infection.

Collaborators

Neil Caparaso (DCEG)
Eric Engels (DCEG)
Jonine Figueroa (DCEG)
Neal Freedman (DCEG)
Allan Hildesheim (DCEG)
Hormuzd Katki (DCEG)
Troy Kemp (DCEG)
Maria Teresa Landi (DCEG)
Ruth Pfeiffer (DCEG)
Ligia Pinto (DCEG)
Mark Purdue (DCEG)
Rachael Stolzenberg-Solomon (DCEG)