Research Preceptor Teams

Students in the Vanderbilt Ph.D. Program in Epidemiology will be matched shortly after acceptance with research preceptor teams. These established multidisciplinary teams include epidemiology faculty, clinical experts and clinical researchers, biostatisticians, and experienced research staff. The research preceptor team commits to involving the student as a co-investigator from the beginning of the student´s graduate studies. The goal is to create a mutually beneficial partnership that produces synergy between education, professional development, and the conduct of research. Top applicants will have an expenses-paid, on-campus interview during which they will meet with research teams. Both students and research teams will then have the opportunity to rank whom they feel would be the best match(es). The student will work with the team 15-20 hours per week during the academic year and full-time during the summer. The student will receive a stipend as compensation for this work. The following are examples of potential research preceptor team opportunities. Check back regularly because we continue to add teams that are accepted to precept students.

Adipose Related Biomarkers of Prostate Cancer 
Adipocytes highly express the GYP19 gene encoding aromatase, an enzyme responsible for metabolizing androgens (e.g., testosterone) to estrogen (E). Estrogens activate estrogen and androgen receptors in prostate cells, and E exposure may permit cancer of the prostate (CaP) to progress to clinical detection and androgen-independence. Recently, a preliminary investigation found the Arg264Cys polymorphism in CYP19 associated with increased CaP risk, suggesting that body adiposity affects prostate carcinogenesis by altering the androgen/estrogen balance. However, epidemiologic studies provide little support for a relationship between obesity and CaP, perhaps because it is difficult to estimate body adiposity in large epidemiologic studies. Our multi-disciplinary team from the fields of nutritional epidemiology, pathology, medical genetics, molecular biology, and urology propose to investigate the role of adipose and estrogen related biomarkers in prostate carcinogenesis. Toward this goal, this pilot molecular epidemiologic study aims to recruit men undergoing confirmatory diagnostic tests for CaP. Fasting blood samples will be collected, and diet, physical activity, body size, and other CaP risk factors will be measured during pretreatment interviews. The recruitment base includes clinical centers serving the African-American population of Nashville, TN. With the endorsement of clinical staff we expect to recruit 70% of all biopsy patients at these centers, providing 250 CaP cases during one year. A control group (n=250), matched to the age and race distribution of the case group, will be selected from men without CaP at biopsy. Bias in the analysis may be reduced by excluding controls with latent CaP, systematic data collection by trained interviewers, and analysis of pre-treatment blood samples. The association between CaP and CYP19 genetic polymorphisms (polymorphisms (Arg264Cys, allele length) will be investigated using multivariable logistic regression controlling for genetic susceptibility to adipocyte differentiation (the Pro12Ala polymorphism of peroxisome proliferator activated receptor- y2), body adiposity (blood leptin), insulin resistance (blood insulin and Cpeptide), diet, physical activity, and other potential CaP risk factors. Greater body adiposity may advance prostate carcinogenesis through an estrogen mechanism, and an analysis of adipose-related biomarkers may suggest that weight reduction would decrease CaP risk or improve prognosis.  Funded by the National Cancer Institute 
Investigators: Jay H. Fowke, M.P.H., Ph.D. (PI)

Association Studies of Regional Genome Scan to Identify Breast Cancer Susceptibility Genes 
This is an ongoing study to perform regional fine mapping to identify candidate genes for breast cancer in two chromosome regions. In this DOD-funded IDEA award, we will use a multiple-phase study design to gain study efficiency, reduce costs, and balance both type I and type II errors. We will 1) perform regional fine mapping to identify candidate genes for breast cancer in two chromosome regions; 2) identify all common sequence variants in the functional regions of identified candidate genes; 3) evaluate the association of sequence variants with breast cancer risk in the Shanghai Breast Cancer Study (SBCS); and 4) confirm the associations in the Nashville Breast Health Study (NBHS). Gene-gene and gene-environment interactions will also be evaluated. Funded by the  Department of Defense (DOD), Congressionally Directed Medical Research Programs  
InvestigatorsQiuyin Cai, M.D., Ph.D. (PI)Xiao-Ou Shu, M.D., Ph.D.Wei Zheng, M.D., Ph.D.Chun Li, Ph.D.; Jirong Long, Ph.D.; Chuanzhong Ye, M.D., Ph.D.

Biomarkers of Methionine Metabolism and Risk for Colorectal Adenoma 
Colorectal cancer is one of the most common malignancies and is the second leading cause of cancer death in the United States.  Most of these cancers arise from adenomatous polyps.  Factors involved in one-carbon metabolism and aberrations in methylation reactions have been implicated in previous studies of colorectal neoplasia.  Methionine metabolism is a key component in one-carbon metabolism and it is integrally involved in most physiologic methylation reactions including DNA and RNA methylation. S-adenosylhomocysteine (SAH), the precursor to homocysteine, has been proposed as a more sensitive marker of alterations in methionine metabolism than homocysteine.  Additionally, the ratio of s-adenosylmethionine (SAM) to SAH is a marker of methylating capacity and a lower SAM:SAH is related to increased DNA hypomethylation.  However, neither of these markers has been evaluated in relation to cancer risk.  We hypothesize that a high SAH level and a low SAM:SAH ratio are associated with an increased risk of colorectal adenoma.  In this proposed study, we will evaluate these hypotheses by conducting a nested case-control study using plasma samples collected as part of the Tennessee Colorectal Polyp Study (TCPS; P50CA95103), an on-going colonoscopy-based case-control study of colorectal adenoma.  We will measure SAH and SAM levels in 265 cases and 265 matched controls.  The hypotheses proposed in the application are novel and based on strong biological plausibility.  Results from this study will not only be useful to understand one-carbon metabolism and colorectal neoplasia but also be informative for developing a sensitive biomarker for identifying high-risk individuals for cost-effective colorectal screening and chemoprevention.  Funded by the  National Cancer Institute 
Investigators: Martha Shrubsole, Ph.D. (PI); Wei Zheng, M.D., Ph.D.

Child Health: Healthy Lifestyle 
Salud con la Familia (Health with the Family) is an on-going family-based community centered intervention trial examining Latino preschool weight trajectories over time. In March 2009, we enrolled a cohort of 100 parent-child dyads to participate in a healthy lifestyle skills building curriculum. The data collected examine the interaction between genetics (saliva samples collected examining common single nucleotide polymorphisms associated with elevated Body Mass Index (BMI)) and behavior (physical activity as measured by accelerometry and nutrition and parental feeding practices as measured by survey measures) on biologic outcomes including BMI, BMI percentiles for children, and fat free mass. Data have been collected at three points in time to examine changes over time in preschool child weight trajectories. To this dataset, we have been funded by the Robert Wood Johnson Foundation to examine, the sustainability of community recreation center use by Latino families with preschoolers. Latino families who participated in Salud con la Familia versus those that didn’t (N= 200) will report on their perception of community recreation centers and use of these for regular physical activity.polymorphisms associated with elevated Body Mass Index (BMI)) and behavior (physical activity as measured by accelerometry and nutrition and parental feeding practices as measured by survey measures) on biologic outcomes including BMI, BMI percentiles for children, and fat free mass. Data have been collected at three points in time to examine changes over time in preschool child weight trajectories. To this dataset, we have been funded by the Robert Wood Johnson Foundation to examine, the sustainability of community recreation center use by Latino families with preschoolers. Latino families who participated in Salud con la Familia versus those that didn’t (N= 200) will report on their perception of community recreation centers and use of these for regular physical activity.

Child Health: Physical Activity 
En Sus Marcas uses data from a randomized controlled trial conducted to examine the feasibility of an innovative obesity intervention for Latino elementary school-aged children at risk for overweight and obese. We are focusing on activity change among 126 parent-child dyads in Forsyth County, NC who were encouraged to increase their physical activity. An area available for student research is to assess the effect of utilizing the pediatric practitioner´s office as an entry point for an active collaboration with a community recreation center

Clinical: Orthopedics and Rehabilitation
There are two Multicenter Orthopaedic Outcome Networks (MOON) at Vanderbilt, one focusing on knee and the other on shoulder. MOON is a consortium of academic centers participating in cohort studies where Vanderbilt serves as the data coordinating center. Knee MOON has an ongoing, prospective cohort study with the goal of identifying the risk factors for reduced quality of life, osteoarthritis and additional surgical procedures in patients treated with anterior cruciate ligament reconstruction (ACLR). Since 2002, MOON has enrolled more than 3,000 individuals with disruption of the anterior cruciate ligament (ACL) and obtained 2-year follow-up data on at least 87% of the cohort. While we have focused on knee, sports-related injuries, and general quality of life, osteoarthritis and recurrent surgery outcomes, the MOON project captures detailed information on patient and injury characteristics and treatment decisions made during initial ACLR surgery. Data cleaning and preliminary analyses have been performed and the database is large enough to investigate additional questions of clinical relevance. Shoulder MOON has an ongoing prospective cohort study enrolling subjects with full-thickness rotator cuff tears into a non-operative rehabilitation program to identify factors associated with success of non-operative treatment , and examine predictors of going on to surgical repair. Areas available for investigation are numerous and include topics such as gender differences in outcomes after ACLR, factors such as fear of re-injury associated with return to activity/sport after ACLR, predictors of ACL graft failure and injury to the contralateral ACL, and whether socioeconomic factors influence outcomes. Psychometric properties of several patient-reported outcome scales could be investigated.

Breast Cancer Survival: Lifestyle & Genetic Determinants
The population of women who survive breast cancer is rising rapidly. Despite the success of initial treatments, many patients constantly battle fears of disease recurrence and early death. However, the effect of non-clinical factors, particularly genetic factors, on breast cancer outcomes is largely unknown. The proposed study will use the existing resources of two well-established cohort studies of 3593 breast cancer patients to comprehensively evaluate the following hypotheses: 1) Breast cancer survival may be associated with genetic polymorphisms in genes encoding angiogenic factors and matrix metalloproteinases, both of which are essential for tumor growth and metastasis. 2) Infiltrating inflammatory cells, particularly tumor- associated macrophages, can produce a large variety of promalignant cytokines and growth factors. Genetic polymorphisms in inflammatory chemokine and cytokine genes may be related to breast cancer survival. 3) Transforming growth factor-B promotes the growth and progression of breast cancer, and genetic polymorphisms in TGF-R pathway genes may be related to breast cancer survival. 4) The cyclooxygenase-2 (COX2) gene is up-regulated in a large proportion of mammary tumors, and this enzyme initiates the biosynthesis of various prostaglandins with diverse, and sometimes opposing, effects on tumorigenesis. Genetic polymorphisms of prostaglandin-pathway genes may be associated with breast cancer survival. A two-phase study design will be applied. In Phase I, all functional variants plus haplotype-tagging single nucleotide polymorphisms (SNP) will be genotyped among 1193 breast cancer patients who have been followed for an average of 7.1 years. All promising associations identified in Phase I will be evaluated in Phase II in an on-going cohort study of 2400 cancer patients (being followed for 5 years). The large sample size and two-phase study design will balance both Type I and Type II errors and provide credible results towards improving the understanding of associations between genetic factors and breast cancer outcomes. Identifying factors that predict risk of relapse and rates of mortality will not only affect the expanding cancer survivor population by providing evidence-based information, but will also positively influence the medical care system and economy at large by making treatment more effective and cost-efficient. The proposed study, built on successfully implemented cohort studies, will be extremely timely and cost-efficient. Funded by the polymorphisms in genes encoding angiogenic factors and matrix metalloproteinases, both of which are essential for tumor growth and metastasis. 2) Infiltrating inflammatory cells, particularly tumor- associated macrophages, can produce a large variety of promalignant cytokines and growth factors. Genetic polymorphisms in inflammatory chemokine and cytokine genes may be related to breast cancer survival. 3) Transforming growth factor-B promotes the growth and progression of breast cancer, and genetic polymorphisms in TGF-R pathway genes may be related to breast cancer survival. 4) The cyclooxygenase-2 (COX2) gene is up-regulated in a large proportion of mammary tumors, and this enzyme initiates the biosynthesis of various prostaglandins with diverse, and sometimes opposing, effects on tumorigenesis. Genetic polymorphisms of prostaglandin-pathway genes may be associated with breast cancer survival. A two-phase study design will be applied. In Phase I, all functional variants plus haplotype-tagging single nucleotide polymorphisms (SNP) will be genotyped among 1193 breast cancer patients who have been followed for an average of 7.1 years. All promising associations identified in Phase I will be evaluated in Phase II in an on-going cohort study of 2400 cancer patients (being followed for 5 years). The large sample size and two-phase study design will balance both Type I and Type II errors and provide credible results towards improving the understanding of associations between genetic factors and breast cancer outcomes. Identifying factors that predict risk of relapse and rates of mortality will not only affect the expanding cancer survivor population by providing evidence-based information, but will also positively influence the medical care system and economy at large by making treatment more effective and cost-efficient. The proposed study, built on successfully implemented cohort studies, will be extremely timely and cost-efficient. Funded by the National Cancer Institute
Investigators Xiao Ou Shu, M.D., Ph.D. (PI)

Cancer Risk Reduction & Diet: A Cohort Study of Women 
Although nutritional factors are thought to play a role in the etiology of over one -third of all human cancers, information about the preventive potential of specific dietary compounds is scanty. We propose herein the renewal of a large-scale cohort study that offers unique opportunities to fill such knowledge gaps. The Shanghai Women’s Health Study (SWHS) is a population-based cohort study of 75,049 Chinese women who were between 40 to 70 years of age at enrollment during 1997 to 2000 and lived in urban Shanghai, where intake levels of many hypothesized cancer-inhibitory dietary factors are high and diverse. Detailed information on dietary and other lifestyle factors was collected at baseline and again in a follow-up survey. Biological samples were collected from the 87.5 percent of cohort members. The cohort has been followed for cancer occurrence and deaths through linkage with the population-based Shanghai Cancer Registry and the Shanghai Vital Statistics Unit and biannual visits to all living cohort members. We propose in this 5-year renewal to extend the follow-up of this cohort for five more years and to evaluate hypotheses related to the etiology of cancers of the breast, colorectum, lung, and stomach. The primary focus of the study is to determine whether certain diets, specifically those including regular tea drinking and high intakes of folate, soy foods, allium vegetables, and crucifers, are associated with a reduced risk of cancer. We also propose to conduct nested case-control studies of breast, lung, and colorectal cancers to evaluate whether the levels of blood folate and urinary isothiocyanate and phytoestrogens are inversely associated with the risk of these cancers. These biomarkers are aggregate measures of the level of intake, absorption, and metabolism and will provide added insight in elucidating the relations of dietary factors with cancer risk. We will further evaluate in the nested case-control studies whether the effect of dietary crucifer intake and urinary isothiocyanate excretion may be modified by genotypes of the GST-family genes. Finally, we propose to re-survey all living cohort members to obtain updated information on usual dietary intake, physical activity, and other lifestyle factors. This will enable us to refine exposure assessment and characterize and evaluate how temporal changes in exposures may influence cancer risk. Because of its size, setting, and inventory of baseline information and biological specimens, the SWHS provides an exceptional opportunity to address dietary hypotheses for cancer that cannot be adequately investigated in any other existing cohort studies. The results from this study may guide new strategies in the primary prevention of common cancer in both Western and Asian women. Funded by the  National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI)

Cell Cycle/Apoptosis Gene Variants and Breast Cancer Risk 
It is well established that genetic factors play a major role in breast tumorigenesis.  However, the known breast cancer susceptibility genes account for only a minority of breast cancer cases in the general population.  Cumulative evidence from  in vitro  and animal studies suggests that the genes involved in the cell-cycle control and apoptosis pathways may be related to breast cancer.  The cell-cycle control and apoptosis pathways function as an integrated molecular network, and perturbations in one pathway can have profound consequences on the other.  In this study, we will investigate genetic variations of major genes involved in the cell-cycle control and apoptosis pathways in relation to breast cancer risk.  The proposed study will use data and biological samples collected from the Shanghai Breast Cancer Study (SBCS, R01CA64277).  Genetic polymorphisms in 25 candidate genes in the cell-cycle control and apoptosis pathways will initially be evaluated using both the genotype- and haplotype- approaches in 1,250 cases and 1,250 controls recruited from 1996 to 1998.  Promising associations identified in the initial phase will be re-evaluated in a second set of subjects (1,850 cases and 1,850 controls) recruited from 2002 to 2005.  This two-phase study design will effectively balance both Type I and Type 2 statistical errors and provide credible results towards our understanding of the etiology of breast cancer.  We will use the multifactor-dimensionality reduction (MDR) statistical method to investigate any gene-gene interaction in relation to breast cancer risk.  We will 1) investigate interaction between genetic polymorphisms with endogenous estrogen exposure related factors in relation to breast cancer risk; 2) evaluate whether genetic polymorphisms may be associated with the risk of specific subtypes of breast cancer; and 3) conduct polymorphisms in 25 candidate genes in the cell-cycle control and apoptosis pathways will initially be evaluated using both the genotype- and haplotype- approaches in 1,250 cases and 1,250 controls recruited from 1996 to 1998.  Promising associations identified in the initial phase will be re-evaluated in a second set of subjects (1,850 cases and 1,850 controls) recruited from 2002 to 2005.  This two-phase study design will effectively balance both Type I and Type 2 statistical errors and provide credible results towards our understanding of the etiology of breast cancer.  We will use the multifactor-dimensionality reduction (MDR) statistical method to investigate any gene-gene interaction in relation to breast cancer risk.  We will 1) investigate interaction between genetic polymorphisms with endogenous estrogen exposure related factors in relation to breast cancer risk; 2) evaluate whether genetic polymorphisms may be associated with the risk of specific subtypes of breast cancer; and 3) conduct  in vitro  experiments to evaluate the function of genetic variations to further confirm the biological relevance of the association.  With its large size and strong methodology, the SBCS provides a great opportunity to investigate gene-gene and gene-environment interactions in relation to breast cancer risk.  The findings from the study are likely to significantly advance our knowledge of the etiology of breast cancer and will be valuable for identifying high risk women for the primary and secondary prevention of breast cancer. Funded by the  National Cancer Institute 
Investigators:   Qiuyin Cai, M.D., Ph.D. (PI) ; Wei Zheng, M.D., Ph.D.; Xiao-Ou Shu, M.D., Ph.D.;  Jirong Long, Ph.D.; Wanqing Wen, M.D.;  Marylyn Ritchie, Ph.D.;  Yinghao Su, M.D., Ph.D.;  Shimian Qu, Ph.D.

Diet, Genetics, Epigenetics and Colorectal Adenoma Risk 
This research examines lifestyle, genetic and epigenetic susceptibility, and colorectal adenoma risk. Folate, a B vitamin, is essential for regenerating methionine, the methyl donor for DMA methylation. Evidence of a role of low folate and/or methionine in colorectal adenoma/cancer risk is accumulating. Simultaneously, there is increasing evidence that aberrant DNA methylation, such as hypermethylation of tumor suppressor genes, is increased in both colorectal cancers and adenomas. To advance our understanding of the role of diet and DNA methylation in adenoma development, epidemiologic studies examining the role of these factors and genes involved in DNA methylation associated pathways are needed. Thus, the research goals of this proposal are to evaluate in a large colorectal adenoma case-control study (n=3000) risk associated with methyl-group intake (folate, methionine, vitamin B6, vitamin B12), polymorphisms in genes involved in methyl-group transfer (MTHFR, DNMT1, DNMT3A), promoter methylation of tumor suppressor genes in normal rectal mucosa (APC, RASSF1 A, n=4DO), and the combined association of these factors. The methylation phenotype in normal rectal mucosa will be compared with methylation phenotype in adenomas. Additionally, the potential utility of methylation phenotype in normal rectal mucosa as a marker of adenoma risk will be evaluated by studying factors that affect level of methylation. This work will extend our understanding of links between DNA methylation and adenoma risk as well as interactions with genetic susceptibility. It may also serve to help identify high-risk populations for colorectal adenoma. Funded by the polymorphisms in genes involved in methyl-group transfer (MTHFR, DNMT1, DNMT3A), promoter methylation of tumor suppressor genes in normal rectal mucosa (APC, RASSF1 A, n=4DO), and the combined association of these factors. The methylation phenotype in normal rectal mucosa will be compared with methylation phenotype in adenomas. Additionally, the potential utility of methylation phenotype in normal rectal mucosa as a marker of adenoma risk will be evaluated by studying factors that affect level of methylation. This work will extend our understanding of links between DNA methylation and adenoma risk as well as interactions with genetic susceptibility. It may also serve to help identify high-risk populations for colorectal adenoma. Funded by the  National Cancer Institute 
Investigators:  Martha Shrubsole, Ph.D. (PI); Wei Zheng, M.D., Ph.D.;  Robert Coffey,  M.D.; Wael El-Rifai, M.D., M.Sc., Ph.D.;  Jonathan Haines, Ph.D.;  Jirong Long, Ph.D.

Dietary Calcium and Magnesium, Genetics, and Colorectal Adenoma 
Results have been inconsistent on the protective effect of calcium and magensium intake on colorectal cancer and adenoma. We found recently in the Tennessee Colorectal Polyp Study (TCPS; P50CA95103) that the associations between intake of calcium or magnesium and risk of colorectal adenoma and hyperplastic polyps may differ by the common Thr1482Ile polymorphism of the TRPM7 gene, a gene involved in calcium and magnesium (re)absorption and homeostasis. Our finding may partially explain the inconsistency in previous studies on calcium and magnesium. In addition, we found that the ratio of calcium to magnesium intake significantly interacted with the Thr1482Ile polymorphism in relation to both adenomatous and hyperplastic polyps. In response to PAR-07-377, we propose a clinical epidemiologic study to test several novel hypotheses regarding gene-nutrition interactions using data and biological samples collected as part of the TCPS, a large on-going molecular epidemiologic case-control study of colorectal adenoma. Specifically, we will 1) confirm our pilot finding in an independent set; and 2) conduct a two-phase study to evaluate the relationships between other polymorphisms in 14 candidate genes involved in magnesium and calcium (re)absorption, regulation and balance and risk of colorectal adenoma; and investigate whether the associations between intake of calcium and magnesium or the ratio of calcium to magnesium intake and risk of colorectal adenoma differs by the genotypes or haplotypes in the 14 genes. The first phase of the study will include 1200 cases and 2400 controls to comprehensively investigate promising polymorphisms and their interactions with nutrients. All promising variants will be re-evaluated in an independent set of 800 cases and 1600 controls to validate the identified associations or nutrient-gene interactions. The proposed two-phase study design will allow us to effectively address potential false positive findings (Type I error), one of the most serious concerns regarding association studies of low-penetrance genetic factors and will allow us to enhance the statistical power for evaluation of gene-gene and gene-nutrition interactions. The results from our study will help to identify people at a high risk of colorectal adenoma and to develop personalized strategies of dietary changes or nutritional fortification to prevent occurrence of colorectal adenoma, and, thus, colorectal cancer. In the general U.S. population, 1 in 18 individuals will develop colorectal cancer over their lifetime and forty percent will die within five years of diagnosis, mainly due to diagnosis at a late stage. Therefore, development of primary preventive strategies for colorectal cancer is critical. The results from our study will help to identify people at a high risk of colorectal adenoma and to develop personalized strategies to prevent occurrence of colorectal adenoma, and, thus, colorectal cancer through dietary changes or nutritional fortification. Funded by the polymorphisms in 14 candidate genes involved in magnesium and calcium (re)absorption, regulation and balance and risk of colorectal adenoma; and investigate whether the associations between intake of calcium and magnesium or the ratio of calcium to magnesium intake and risk of colorectal adenoma differs by the genotypes or haplotypes in the 14 genes. The first phase of the study will include 1200 cases and 2400 controls to comprehensively investigate promising polymorphisms and their interactions with nutrients. All promising variants will be re-evaluated in an independent set of 800 cases and 1600 controls to validate the identified associations or nutrient-gene interactions. The proposed two-phase study design will allow us to effectively address potential false positive findings (Type I error), one of the most serious concerns regarding association studies of low-penetrance genetic factors and will allow us to enhance the statistical power for evaluation of gene-gene and gene-nutrition interactions. The results from our study will help to identify people at a high risk of colorectal adenoma and to develop personalized strategies of dietary changes or nutritional fortification to prevent occurrence of colorectal adenoma, and, thus, colorectal cancer. In the general U.S. population, 1 in 18 individuals will develop colorectal cancer over their lifetime and forty percent will die within five years of diagnosis, mainly due to diagnosis at a late stage. Therefore, development of primary preventive strategies for colorectal cancer is critical. The results from our study will help to identify people at a high risk of colorectal adenoma and to develop personalized strategies to prevent occurrence of colorectal adenoma, and, thus, colorectal cancer through dietary changes or nutritional fortification. Funded by the  National Center for Complementary and Alternative Medicine 
Investigators:  Qi Dai, M.D., Ph.D.; Jirong Long, Ph.D.; Reid M. Ness, M.D.; Marylyn Ritchie, Ph.D.; Jiajun Shi, Ph.D.; Martha J.  Shrubsole, Ph.D.; Wei Zheng, M.D., Ph.D.

Effects of Brassica Intake or Indole-3-Carbinol (I3C) Supplements  on PSA among Men with Biochemical Failure 
There is considerable basic science research to suggest that isothiocyanates (ITC) or several indole analogs (e.g., indole-3carbinol (I3C)) slow cellular proliferation, re-instate apoptotic action, and reduce colon tumor incidence. Humans are exposed to these agents primarily through consuming Brassica vegetables (e.g., broccoli), but it is unknown if greater Brassica consumed could affect colon cancer risk. We propose a short-term pilot dietary intervention to investigate the effects of greater Brassica consumption on markers colon cancer progression. Twenty adenoma patients will participate in a randomized controlled cross-over trial. Recruitment, intervention, and lab protocols have been developed through other on-going studies, improving the cost-efficiency of this pilot project. The intervention-arm of the trial is based on a model developed for the Women’s Health Initiative, and provides social support and information to help participants incorporate these vegetables into their daily diet. Biomarkers measured from rectal biopsies will include Bcl-2 (inhibits apoptosis), Bak and Bax (promote apoptosis), Mib-1 (marker of cellular proliferation), and p21 (marker of cellular differentiation). Multiple 24-hour dietary recalls will measure each participant’s adherence to the intervention. Dietary adherence will be measured further by urinary ITC level (combined with GST enzyme genotype), a unique and specific biomarker of Brassica vegetable intake. Using mixed-model repeated-measures ANOVA, we will compare molecular marker expression when participants consume Brassica vegetables to when these same participants consume an over-the-counter fiber and vitamin supplement. This pilot study will provide the needed estimates of variance in biomarker response to the dietary change for sample size calculations for future applications. Previously, we found that healthy people were able to increase the consumption of these vegetables with minimal social and instructional support, and we used those pilot data to gain funding for a larger randomized trial enrolling breast cancer survivors. Since Brassica vegetables are widely available, inexpensive, and consumed without harm, a change in molecular marker expression consistent with reduced proliferation and enhanced differentiation and apoptosis would suggest further research to evaluate the application of Brassica consumption to reduce colon cancer risk. Funded by the  National Cancer Institute 
Investigators:  Jay H. Fowke, M.P.H., Ph.D. (PI); Sam Chang, M.D.; Saundra Motley, R.N.

Energy Homeostasis and Cancer: Biomarkers and Lifestyles 
Recognition that adipose tissue is both an energy reservoir and a source of proteins with endocrine functions (adipokines), suggests that these proteins, and the genes controlling them, are part of the homeostatic control mechanisms that regulate cell growth. To advance our understanding of the role of energy homeostasis and cancer, epidemiologic studies examining the role of genes involved in energy regulation in humans, as well as the interactions of these exposures with lifestyle factors associated with energy balance are needed. Accordingly, the four aims of this research are as follows. In a case-control study of breast cancer (N=2,500), Aim 1 will examine the effect of polymorphisms in genes involved in energy homeostasis (insulin, IGF-1, leptin, and adiponectin), and Aim 2 will examine gene-environment interactions between these genes and lifestyle factors associated with energy balance (physical activity, energy intake, body size). In a case-control study of adenomatous polyps (N=1,400), Aim 3 will examine the effect of lifestyle factors associated with energy balance (as above), and (Aim 4) will examine the effect of polymorphisms in genes involved in energy homeostasis (as above). Primary research support will come from existing molecular epidemiology studies of the candidate’s mentor. Resources from this award will support genotyping of adipokines not originally examined in these studies. This work will extend our understanding of links between energy homeostasis and cancer, as well as interactions between genetic susceptibility, lifestyle factors, and cancer. This work will also serve as the basis for submission of competitive R01 proposals by the candidate in latter years of this award. Funded by the polymorphisms in genes involved in energy homeostasis (insulin, IGF-1, leptin, and adiponectin), and Aim 2 will examine gene-environment interactions between these genes and lifestyle factors associated with energy balance (physical activity, energy intake, body size). In a case-control study of adenomatous polyps (N=1,400), Aim 3 will examine the effect of lifestyle factors associated with energy balance (as above), and (Aim 4) will examine the effect of polymorphisms in genes involved in energy homeostasis (as above). Primary research support will come from existing molecular epidemiology studies of the candidate’s mentor. Resources from this award will support genotyping of adipokines not originally examined in these studies. This work will extend our understanding of links between energy homeostasis and cancer, as well as interactions between genetic susceptibility, lifestyle factors, and cancer. This work will also serve as the basis for submission of competitive R01 proposals by the candidate in latter years of this award. Funded by the  National Cancer Institute 
Investigators:   Charles E. Matthews, Ph.D. (PI)

Epidemiologic Study of Predictors for Adenoma Recurrence 
Most colorectal cancers arise from adenomatous polyps, and a large proportion of adenoma patients will develop new adenomas after their initial polypectomy. There is considerable controversy regarding an appropriate surveillance interval for adenoma patients following the removal of their initial adenomas. Therefore, studies assessing predictors for recurrent adenoma, particularly among patients with multiple or pathologically advanced adenoma(s), will provide valuable information for designing individualized, cost-effective surveillance and chemoprevention strategies for adenoma patients. Some tumor markers (genetic or epigenetic alternations) involved in the formation of colorectal neoplasms are promising predictors for recurrent adenomas, as they are believed to reflect a field cancerization process or a genetic predisposition to colon adenomas. We hypothesize that patients whose initial adenomas have certain altered genetic or epigenetic profiles may have an elevated risk of adenoma recurrence, and these tumor markers, along with pathologic features of initial adenomas can be used to predict the risk of adenoma recurrence. To evaluate these hypotheses we propose in this application a series of investigations, consisting of both hypothesis-testing and hypothesis-generating components as described below. For the reasons described in section BI, the major focus of this application will be on the study of predictors for recurrent adenomas among patients with multiple or pathologically advanced adenoma(s). Funded by the National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI)

Etiological Studies of Gastric Carcinoma 
The goal of this project has been and continues to be the multi-disciplinary study of the etiology of gastric cancer. This neoplastic disease is second only to lung cancer in incidence and mortality worldwide. In the United States gastric cancer rates have decreased considerably. There are, however, high-risk groups, especially African Americans, Amerindians, and immigrants from Asia, Northern/Eastern Europe and Latin America. It has become increasingly clear that a major etiologic factor is chronic infection with Helicobacter pylori. About one-half of today´s world population is infected, especially groups of lower socioeconomic status. The International Agency for research on cancer has classified Helicobacter infection as a class 1 carcinogen. There are great differences in the outcome of the infection. Most infections are mild and subclinical. Clinical infections may lead to duodenal ulcer accompanied by non-atrophic gastritis, which does not increase gastric cancer risk, or to multifocal atrophic gastritis, which may lead to gastric ulcer and gastric cancer. Our general hypothesis is that the immune and inflammatory responses determine the outcome of the infection. The Program Project has been recently renewed, its 24th year of continued funding by NCI.  The general theme of the research is to explore the mechanisms of causality of gastric carcinoma as a paradigm for carcinogenesis driven by infectious agents. The field work is conducted mainly in Colombia, a "population laboratory" with subgroups of very high- as well as  very low-gastric cancer risk, previously identified by the investigators. Epidemiologic information and biological specimens collected in Colombia are brought to Vanderbilt Univerity to be processed and analyzed by the investigators. The Program Project has  three main projects, intimately inter-connected to address the central theme of the etiology of gastric cancer. The first project involves the follow up of cohorts previously treated for Helicobacter pylori infection. The adults cohort ha been followed up for 17 years and will be followed up to its 20th year. Previous work has suggested a model of the dynamics of cancer prevention as a mirror image of the carcinogenesis model expressed as a sigmoid exponential curve proposed by Doll and Peto for tobacco-induced carcinogenesis. Long term follow up of this cohort should determine if the model is correct and if curing the infection has permanent beneficial effects. The precancerous process in this cohort will be explored with methylation assays of CpG islands to investigate if they are indicators of progression of the precancerous process.  Children´s cohorts will be studied to explore the natural history of the infection, as well as possible genotypic and phenotypic changes in the infecting bacteria over time. The second project, under the leadership of Dr. James Fox from MIT investigates the modulation of the immune response to Helicobacter infection by concomitant helminthic infections in mice and gerbils. The third project, under the leadership of Dr. Keith Wilson studies the oxidative and nitrosative stress in carcinogenesis induced by Helicobacter pylori, in vitro and in vivo utilizing mice and gerbils. Funded by the  National Cancer Institute 
Investigators:  Pelayo Correa, M.D. (PI); Keith Wilson, M.D.

Exercise Intervention for Chemotherapy -related Cognitive Dysfunction 
The major goal of this research is to evaluate the ability of exercise training to enhance cognitive function among cancer survivors. Funded by the  Lance Armstrong Foundation 
Investigators:  Charles E. Matthews, Ph.D. (PI)

Genetic Consequences of Therapies for Cancer 
Over 270,000 survivors of childhood cancer are estimated to be alive today in the United States alone; many are able to have children of their own. Consequently, possible effects of curative cancer treatments (like radiation and chemotherapy) on inherited disorders are increasingly important. However, there is little understanding of genetic consequences of treatment or whether underlying susceptibility can be transmitted to offspring. Further, young adults diagnosed with cancer at ages 20 to 34 years are often overlooked in studies of late effects. While there is little evidence that mutagenic therapies result in transgenerational effects, few studies have looked at risk in terms of treatment dose to testes or ovaries. In this research all persons diagnosed with cancer under age 35 after 1943 in Denmark and after 1952 in Finland are being identified, along with their siblings. Among the 10,000 children with cancer who survived to reproductive ages, 3,000 are estimated to be the parents of 5,600 children. Among the 38,000 patients diagnosed with cancer as young adults, 25,000 survived and had 14,000 children after their cancer diagnosis. Thus, 19,600 offspring of cancer survivors can be studied. Rosters of siblings and their offspring are being developed for comparison purposes. The offspring cohorts in Denmark and Finland will then be linked to outcome registries to identify cancer, birth defects, stillbirths and neonatal and other deaths. Medical records of the cancer survivors will be obtained and radiation records and chemotherapy information abstracted. Radiation doses to gonads (and uterus for females) will be calculated, and the genetic consequences of curative therapies will be assessed. Gonadal exposures to radiation or chemotherapy for many cancer survivors will be high – just below the threshold for infertility. Blood will be collected from a sample of survivors, their spouses, and their offspring to examine mechanistic processes related to cancer predisposition and effect of therapy on health outcomes. Two hundred families will donate lymphocytes and DNA for storage and analyses to include the G2 radiation assay to assess chromosomal radiosensitivity (that might be related to alterations of DNA damage-response/repair genes); to determine whether such a sensitivity can be inherited; to evaluate specific repair genes (e.g. XRCC1) for variant polymorphisms; and to investigate minisatellite inheritance. The study will help answer questions about genetic consequences of mutagenic exposures, explore whether susceptibility states and specific genetic polymorphisms conferring susceptibility can be identified for specific cancers, and evaluate the extent to which-identified susceptibility or genetic damage can be transmitted to future generations. Funded by the polymorphisms; and to investigate minisatellite inheritance. The study will help answer questions about genetic consequences of mutagenic exposures, explore whether susceptibility states and specific genetic polymorphisms conferring susceptibility can be identified for specific cancers, and evaluate the extent to which-identified susceptibility or genetic damage can be transmitted to future generations. Funded by the  National Cancer Institute 
Investigators:  John Dunning Boice, Sc.D. (PI)

Genetic Factors for Breast Cancer – A Genome Wide Study 
Breast cancer is the most common malignancy among women in many parts of the world.  Genetic factors play an important role in the etiology of breast cancer.  However, to date, only a few breast cancer susceptibility genes have been identified, and they explain only a very small fraction of breast cancer cases in the general population.  A large number of candidate-gene studies have been conducted over the past 10 years.  These studies, however, are clearly inadequate to fully uncover the genetic basis of breast cancer.  With recent significant advances in high-throughput genotyping technologies, it has become feasible to conduct genome-wide association (GWA) studies to systematically evaluate genetic risk factors for breast cancer.  The multi-phase GWA study proposed in this application will be built upon the resources established in two large, on-going studies funded by NCI, the Shanghai Breast Cancer Study (R01 CA64277) a population-based case-control study, and the Shanghai Women´s Health Study (R01 CA70867) a population-based prospective cohort study.  Approximately 8,000 breast cancer cases and controls will be included in this proposed study.  In the first phase of the study, we will conduct a GWA scan in 1,000 cases and 1,000 controls using the Illumina HumanHap550 Beadchip.  We will then select the 10,600 most promising SNPs for a validation study in an independent sample of 1,500 cases and 1,500 controls.  All promising SNPS will be further validated using data from 1,000 cases and 2,000 controls selected from the prospective Shanghai Women´s Health Study.  The parent projects of this newly-proposed study have been exceptionally well-conducted with a strong methodology.  The feasibility and utility of the proposed study have been clearly demonstrated in our pilot study.  The study is unique and has many unique features that facilitate a rigorous evaluation of breast cancer genetic factors.  The results from this study will be valuable in identifying high risk women for primary and secondary prevention of breast cancer.  Funded by the  National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI) ; Jirong Long, Ph.D.; Chun Li, Ph.D.; Qiuyin Cai, Ph.D., M.D.; Xiao Ou Shu, M.D., Ph.D.; Jonathan L. Haines, Ph.D.

Global Health: HIV
Friends in Global Health (FGH) is a Vanderbilt University Medical Center affiliate, under the aegis of the Institute for Global Health. FGH operates programs supporting medical care and preventive services for persons with HIV infection or at-risk of acquiring HIV infection in Sub-Saharan Africa. These programs are funded by the President’s Emergency Plan for AIDS Relief (PEPFAR) through the Centers for Disease Control and Prevention (CDC). FGH employs about 100 staff in Mozambique, Nigeria, and Zambia in the following projects:

Mozambique: FGH provides comprehensive HIV testing, care, and treatment in twelve districts of the nation´s most populous province, Zambézia. Since late 2007, the Vanderbilt FGH team has provided over 5,000 Mozambicans with antiretroviral drugs, and this number will at least triple in 2009. Also, in 2009, we began prevention services (prevention of maternal to child transmission and HIV heterosexual prevention). The evaluation of these programs will be a major challenge. In order to improve monitoring and evaluation under a US Agency for International Development (USAID) grant, our program data will be integrated with several partners (other non-governmental organizations or NGOs) in the province providing other health services, including community health, and development initiatives, including agriculture, water and sanitation, and microcredit. This will provide excellent opportunities to link data and expand evaluation of our health programs using a cross-disciplinary approach. Our treatment cohort will be followed for life. However, we anticipate substantial loss to follow-up, particularly in our most rural areas. The predictors and mediators of such losses are key research questions, as are predictors of non-adherence to daily antiretroviral therapy. All epidemiologic discoveries will in turn inform improvements to the effectiveness of our programs.

Nigeria: In two states in northcentral Nigeria, the Kwara and Niger states, the Vanderbilt FGH team is responsible for two large HIV clinics and three rural satellite clinics with similar challenges and goals to the Mozambique project. However, there are key differences in the two nations. First, our Nigerian projects include a mix of rural and urban venues while the Mozambique venues are all rural. Second, we have a strong Nigerian health team spearheading the nation´s work, while in Mozambique, few indigenous health workers are available, so there is a larger expatriate cadre. Third, even though Mozambique has higher HIV prevalence than Nigeria, the absolute number of persons living with HIV is substantially greater in Nigeria. Our projects began in early 2009, and we expect rapid patient accrual.

Zambia: Dr. Sten Vermund, director of the Vanderbilt Institute for Global Health, founded the Centre for Infectious Disease Research in Zambia (CIDRZ) in 2000, and his former fellow, Dr. Jeffrey Stringer, is responsible for a huge PEPFAR program with over 100,000 persons on antiretroviral therapy. Dr. Stringer is a professor of obstetrics and gynecology at the University of Alabama at Birmingham (UAB) and he has co-mentored two doctoral students with Dr. Vermund. The Vanderbilt activities in Zambia are focused on educational capacity-building in the University of Zambia School of Medicine, as well as collaborative research with the UAB team at CIDRZ.

Our three clinical care databases (one in each nation) consist of about 60 key variables, including sociodemographic, medical, and laboratory data. The Zambian database is the oldest and most substantial. However, the Mozambican and Nigerian programs are growing rapidly and we expect the ability to do substantial cross-national comparisons by late 2010. Our prevention database for mothers and infants includes key process variables in the antenatal prevention cascade (maternal HIV status, whether a test was offered and accepted, whether nevirapine was accepted and taken, and whether the baby nevirapine dose was given). Our team is experienced in HIV prevention and clinical care research in developing countries. We are well poised to nurture a young investigator doing an epidemiology doctoral thesis at Vanderbilt University, particularly one with a passion for global health and health care systems improvement.

Healthcare Epidemiology: Patient Safety and Clinical Quality 
The Center for Perioperative Research in Quality (CPRQ), in collaboration with the Vanderbilt Center for Experiential Learning and Assessment (CELA) and the Middle Tennessee Center for Improving Patient Safety (CIPS), conducts basic and applied research toward improving patient safety and clinical quality. CPRQ conducts studies to understand the cause of unexpected clinical events and how such events can be prevented. Using a range of human factors engineering, health services, and biomedical informatics techniques, CPRQ seeks to elucidate how and why care deviates from optimal. Interventions are then designed and formally evaluated to improve safety and quality. CPRQ supports investigators performing work, for example, in anesthesiology, nursing, surgery, emergency medicine, critical care, and ambulatory medicine. In particular, we are interested in the nature of expertise, clinician-clinician communication, clinical workload, situational awareness, and other intrinsic and extrinsic variables that affect clinical performance during routine and non-routine clinical care. Recent work has focused on team coordination and effectiveness, the design and evaluation of biomedical technologies, and the use of electronically generated clinical data to identify evolving events and support decision-making. On-going research evaluates human-technology interactions as well as individual and group performance-shaping factors such as novel methods of information presentation to generate practical benefits in terms of improved clinical care processes. Existing databases include over 500 comprehensive video recordings of anesthetic cases including hundreds of non-routine events, and nearly 1000 patient care handoffs.

HIV-Cervical Cancer Prevention Program in India
Increasing access to antiretroviral therapy for HIV/AIDS in resource limited settings and the consequent increasing life-spans of HIV-infected persons has necessitated the development of new guidelines for improving their routine preventive health care. Preventing the development of human papillomavirus (HPV)-induced high-grade cervical squamous intraepithelial lesions (HSIL) among HIV-infected women needs accurate, reliable and cost-efficient screening tools and methods. Given the limited usefulness of cytology-based screening programs, alternatives such as visual inspection with acetic acid (VIA) and HPV testing (through self-administered swabs or clinician-collected samples) have been envisioned as cost-effective alternatives for resource-limited settings. Though large scale studies (including community randomized trials) are being conducted to evaluate the effectiveness of these tests, published estimates are devoid of the performance measures, utility, and implementation challenges of these newer tools among HIV-infected women in resource limited settings. Our network of clinical research sites in India aims to develop evidence in this area that has significant policy translational opportunities. We started with support through an initial R21grant from NCI, followed by supplemental funding for research training initiatives through the Fogarty International Center and the Indian Council of Medical Research (ICMR). These efforts are led by Dr. Vikrant Sahasrabuddhe, Program Director, with Dr. Sten Vermund, PI of the parent training grant, the AIDS International Training and Research Program (AITRP). Our primary collaborating institution is the National AIDS Research Institute (NARI), the nodal HIV/AIDS research institute of ICMR located in Pune, India. Our other partner agencies include B.J. Medical College-Sassoon General Hospitals (BJMC-SGH) and the K.E.M.-Vadu Rural Health Program in Pune, the J.N. Medical College (JNMC) in Belgaum, and the National Institute of Epidemiology (NIE) in Chennai. With support of our research training and capacity building efforts, a large multi-site cohort study of cervical cancer screening approaches among HIV-infected women in India is now underway that will provide the backbone for future prevention clinical trials in this population at high-risk of HPV-induced cervical cancer. Our training efforts have focused on clinical training, clinical research, HPV molecular epidemiology research, and social and behavioral research, and have involved clinicians, nurses, counselors, data management personnel, epidemiologists, statisticians, community outreach workers and laboratory technicians. Our collaborating investigators include Dr. Sanjay Mehendale (NARI), Dr. Seema Sahay (NARI), Dr. Arun Risbud (NARI), Dr. Ramesh Bhosale (BJMC-SGH), Dr. Shivaprasad Goudar (JNMC), Dr. Babasaheb Desai (JNMC), Dr. Siddhivinayak Hirve (KEMHRC) and Dr. V. Kumaraswami (NIE). Funded by the  NIH Fogarty International Center  
Investigators:  Sten H. Vermund, M.D., Ph.D. (PI) ; Vikrant Sahasrabuddhe, M.B.B.S., Dr.P.H.

Infectious Disease: HIV
The Epidemiology Outcomes Unit of the Vanderbilt-Meharry Center for AIDS Research (CFAR) has an observational database of demographic, clinical and laboratory data pertaining to HIV-infected persons who receive care at the Comprehensive Care Center (CCC) in Nashville, TN. The CCC opened in 1994, and more than 6,000 patients have been followed longitudinally since then. There are approximately 2,000 patients actively in care, with approximately 300 new patients per year. Most analyses utilize data from 1998-present. There is a specimen repository (plasma, DNA pellets) that allows for translational studies. The Epidemiology Outcomes Unit convenes weekly to discuss ongoing projects. The CFAR also participates in several multicenter collaborations: North American ACCORD (90,000 patients) Antiretroviral Therapy Cohort Collaboration (ART-CC) Vanderbilt-Johns Hopkins-UNC collaboration CFAR Network of Integrated Cohort Studies (CNICS) This substantially increases the sample size available for research, allowing for analyses of rare endpoint. One Vanderbilt CFAR-funded ongoing project is prospectively collecting data on illicit drug use, depression, and adherence to HIV anti-retroviral therapy among 2,500 HIV-infected patients who seek medical care at the Comprehensive Care Center (CCC) in Nashville, TN. These data will supplement the clinical database at the CCC. Study hypotheses and research questions could be proposed in analyzing the data and developing new research grants related to drug use, mental health, and HIV disease progression. The NIH-funded China Integrated Program for Research on AIDS (CIPRA) was completed during 2002 and 2008. This project has large databases from a natural history prospective cohort study among HIV-positive former blood donors including hundreds of key variables about CD4+T cell count, HIV-specific CD8+ T cell responses, HIV-1 neutralizing antibody (Nab) responses, viral Factors, HLA allelic frequencies, and clinical outcomes. It also has a large number of repository blood specimens available for proposing new research questions. China began national community-based ART among AIDS patients in 2003. The national HIV Drug Resistance (HIVDR) surveillance program started in 2004 to provide technical support to the national ATR program. China´s HIVDR surveillance program includes the national HIVDR survey, training, database management, and quality control program, as well as the WHO HIVDR strategy including sentinel surveillance, threshold survey, and early warning system. Around 10% of all patients in treatment are included into the HIVDR surveillance. The HIVDR program has many research projects on improving the HIVDR genotyping and phenotyping technology, dynamics of HIVDR increase and CD4 decline, as well as the impact of HIVDR on antiviral treatment. Another ongoing study is exploring the possibility and methodology of using surveillance data to support clinical treatment and public health HIVDR monitoring for the public health approach of ART.

Infectious Disease: Tuberculosis
Vanderbilt researchers have several tuberculosis (TB) datasets and ongoing studies. One dataset includes M. tuberculosis isolates from Tennessee TB patients 2002-present that have been tested for phenotypic fluoroquinolone resistance. To date, approximately 900 isolates have been tested. Sequencing of gyrA and gyrB has been performed for all phenotypically resistant isolates, and a sub-set of susceptible isolates. Demographic and clinical data for all patients are also in the database, including data on fluoroquinolone exposure prior to TB diagnosis. Another dataset includes all TB patients reported to the Tennessee Department of Health between 2000-2007 (>2,000 patients). M. tuberculosis isolates are also available for all patients. Risk factors for recurrent TB (including TB relapse) are being assessed. Finally, datasets from the United States (adults) and Brazil (young children) are available for analysis of the immunologic and genetic factors that predispose to TB.

Inflammatory Biomarkers and Colorectal Cancer Risk
Colorectal cancer (CRC) is a major cause of cancer deaths in the United States and many other countries. Chronic inflammation has been suggested to play a major role in the pathogenesis of CRC. We propose in this application to conduct a nested case-control study within the Shanghai Women’s Health Study, a large population-based prospective cohort study, to evaluate the association of CRC risk with measures of several key products of the inflammatory process and with related genetic markers of inflammation. Specifically, we will include 580 incident cases of CRC and their individually matched controls (1 to1 match for biochemical markers and 1 to 3 match for genetic markers). Urine samples collected at baseline will be measured for a major metabolite of prostaglandin E2 (PGE2) using a liquid chromatographic/mass spectrometric assay and F2-isoprostanes using the mass spectrometric method. Baseline blood samples will be measured for soluble tumor necrosis factor-a receptors, interleukin-6, and C-reactive protein. Genomic DNA will be assayed for polymorphisms in genes involved in PGE2 production (PTGS2, PTGES), metabolism (15-PGDH), and signaling (PTGER1-PTGER4). We will perform statistical analyses to evaluate the associations between these biochemical and genetic markers of inflammation and CRC risk and potential interactions of these markers. Given the large sample size, the prospective study design, the availability of comprehensive baseline survey data and biospecimens, as well as the excellent collaborative environment, we believe that this proposed study represents a unique opportunity to evaluate, vigorously and cost-efficiently, the relationship between various markers of inflammation and CRC. Overall, this study will contribute significantly to the understanding of the role of inflammation in the etiology of CRC and to the development of new strategies for the assessment of CRC risk. Funded by the polymorphisms in genes involved in PGE2 production (PTGS2, PTGES), metabolism (15-PGDH), and signaling (PTGER1-PTGER4). We will perform statistical analyses to evaluate the associations between these biochemical and genetic markers of inflammation and CRC risk and potential interactions of these markers. Given the large sample size, the prospective study design, the availability of comprehensive baseline survey data and biospecimens, as well as the excellent collaborative environment, we believe that this proposed study represents a unique opportunity to evaluate, vigorously and cost-efficiently, the relationship between various markers of inflammation and CRC. Overall, this study will contribute significantly to the understanding of the role of inflammation in the etiology of CRC and to the development of new strategies for the assessment of CRC risk. Funded by the  National Cancer Institute

Investigators:  Gong Yang, M.D., M.P.H. (PI); Wei Zheng, M.D., Ph.D.; Xiao Ou Shu, M.D., Ph.D.;  Jason Morrow, M.D.; Ginger Milne, Ph.D.; Qiuyin Cai, Ph.D., M.D.; Xianglan Zhang, M.D., M.P.H.; Wanqin Wen, M.D.;  Jirong Long, Ph.D.;  Leena Choi, Ph.D.

Lung Cancer Risk and Inflammatory Pathways
This is a major project (project 5) of the Vanderbilt Lung Specialized Program of Research Excellence (SPORE). In this study, we will conduct a large nested case-control study within two NCI-funded cohort studies, the Shanghai Women’s Health Study (SWHS) and the Southern Community Cohort Study (SCCS), to investigate biomarkers of inflammation in relation to subsequent lung cancer risk. We will include an estimated 1,130 incident cases of lung cancer that are newly diagnosed after the baseline survey, and an equal number of individually matched controls. Functional/suspected functional polymorphisms and haplotypes in genes involved in PGEpolymorphisms and haplotypes in genes involved in PGE 2 /PGI 2  formation and metabolism will be determined for all the cases and controls. For those with urine and/or blood specimens collected at baseline, we will measure urinary levels of PGE-M, PGI-M, and LTE4, as well as blood levels of C-reactive protein (CRP).  Odds ratios (ORs) adjusted for smoking and other variables will be calculated to measure and test the significance of associations between these biochemical and genetic markers of inflammation and lung cancer risk. Funded by the  National Cancer Institute

Investigators:  Qiuyin Cai, M.D., Ph.D. (PI) ; William Blot, Ph.D.; Wei Zheng, M.D., Ph.D.; Xiao-Ou Shu, M.D., Ph.D.; Jirong Long, Ph.D.; Wanqing Wen, M.D.; Lisa Signorello, Ph.D.

Magnesium, Calcium and Risk for Colorectal Adenoma 
High intake of calcium may protect against both colorectal cancer and adenoma, however, results have been inconsistent. Magnesium, the second most abundant intracelluar cation in the body, plays an essential role in over 300 biological activities.  Ionized magnesium counters the action of ionized calcium in many physiologic activities.  The ratio of calcium to magnesium intake is much higher in the US than in East Asian populations with traditionally low risks of colorectal cancer and other chronic diseases.  However, accurate measurement of body magnesium status has always been difficult in previous epidemiologic studies.  We found very recently in the Tennessee Colorectal Polyp Study (TCPS), a large on-going molecular epidemiologic case-control study of colorectal adenoma, that dietary intake of magnesium may reduce the risk of adenoma, a cancer precursor, particularly when the calcium/magnesium intake ratio is low, whereas under the same conditions, intake of calcium may be associated with a decreased risk. Based on these promising pilot data, we propose a molecular epidemiologic study to confirm our new findings and test some new hypotheses using data and biological samples collected in TCPS.  Primarily, we will evaluate whether erythrocyte magnesium in relation to adenoma risk. The results from our study will help to identify people at a high risk of colorectal adenoma and to develop personalized strategies to prevent occurrence of colorectal adenoma, and, thus, colorectal cancer through dietary changes or nutritional fortification.  Funded by the  American Institute for Cancer Research 
Investigators:  Qi Dai, M.D., Ph.D. (PI); Jirong Long, Ph.D.; Reid M. Ness, M.D.; Martha J. Shrubsole, Ph.D. ; Wei Zheng, M.D., PhD.

Mechanisms and Management of Familial Colorectal Cancer 
Advances in genomics offer the potential for novel clinical applications designed to identify individuals at greater susceptibility to cancer and develop personalized risk-reduction strategies. As such, the family history assessment is becoming increasingly important for cancer control and prevention. While significant advances have been made in our understanding of the etiology and clinical management in inherited colorectal cancer syndromes, such as familial adenomatous polyposis or hereditary non-polyposis colorectal cancer, most individuals with a family history of colorectal cancer do not have one of these well-described, Mendelian conditions. Our understanding of the mechanisms underlying the familial aggregation of colorectal cancers and the most effective screening strategies for individuals at increased risk for colorectal cancer remains limited. The long-term goals of Dr. Murff are to better delineate the complex genetic and environmental factors associated with familial colorectal cancer and translate these findings into improved management of high-risk individuals. The research aims of this proposal include: 1) To determine the effect of a family history of colon cancer on the initiation of colorectal cancer screening interventions; 2) To determine if an interaction exists between having affected relatives with colorectal cancer and patient lifestyle factors on adenoma risk; 3) To evaluate the relationship of functional genetic polymorphisms within the transforming growth factor beta signaling pathway to colonic adenoma risk and family colon cancer history; 4) To determine the cost-effectiveness of colorectal cancer screening in individuals at increased risk based on their family history. The early phases of this career development award will build on the candidate’s prior work and determine current practices for cancer screening in high-risk individuals. Through the candidate’s didactic coursework and mentorship plan, Dr. Murff will develop additional skills in genetics, cancer biology, and molecular epidemiology. The latter phases of the award will build on this new knowledge to investigate the genetic underpinnings of familial colon cancer. The experience gained from this career development award will ensure Dr. Murff’s transition into an independent researcher in cancer epidemiology. Funded by the polymorphisms within the transforming growth factor beta signaling pathway to colonic adenoma risk and family colon cancer history; 4) To determine the cost-effectiveness of colorectal cancer screening in individuals at increased risk based on their family history. The early phases of this career development award will build on the candidate’s prior work and determine current practices for cancer screening in high-risk individuals. Through the candidate’s didactic coursework and mentorship plan, Dr. Murff will develop additional skills in genetics, cancer biology, and molecular epidemiology. The latter phases of the award will build on this new knowledge to investigate the genetic underpinnings of familial colon cancer. The experience gained from this career development award will ensure Dr. Murff’s transition into an independent researcher in cancer epidemiology. Funded by the  National Cancer Institute 
Investigators:   Harvey Murff, M.D., M.P.H. (PI)

Molecular Epidemiologic Study of Breast Cancer 
Most epidemiological studies of breast cancer have focused on investigations of genetic polymorphisms in the genes involved in estrogen and carcinogen metabolism. A number of promising associations have been identified, many of which have yet to be replicated. In this competitive renewal application, we propose to confirm some of the promising associations identified in the initial funding cycle and extend our work to the investigation of other novel genes in estrogen metabolism and several major genes in the pathways of angiogenesis, extracellular matrix remodeling, inflammatory response, and prostaglandin synthesis. These genes are involved in regulating the tumor microenvironment and are likely related to, not only the prognosis, but also the risk of cancers large enough to be detected clinically. The proposed study will use data and biological samples collected from the Shanghai Breast Cancer Study (RO1CA64277). Genetic polymorphisms in approximately 25 candidate genes will initially be evaluated using both genotype- and haplotype-approaches in approximately 1200 cases and 1200 controls recruited from 1996 to 1998. Promising associations identified in the initial phase will be re-evaluated in the second set of subjects (1300 cases and 1300 controls) recruited since 2002 in the parent study. This 2-phase study design will effectively balance both Type I and Type 2 errors and provide credible results towards our understanding of the etiology of breast cancer. Results from this study will be valuable in identifying high risk women for primary and secondary prevention of breast cancer. Funded by the polymorphisms in the genes involved in estrogen and carcinogen metabolism. A number of promising associations have been identified, many of which have yet to be replicated. In this competitive renewal application, we propose to confirm some of the promising associations identified in the initial funding cycle and extend our work to the investigation of other novel genes in estrogen metabolism and several major genes in the pathways of angiogenesis, extracellular matrix remodeling, inflammatory response, and prostaglandin synthesis. These genes are involved in regulating the tumor microenvironment and are likely related to, not only the prognosis, but also the risk of cancers large enough to be detected clinically. The proposed study will use data and biological samples collected from the Shanghai Breast Cancer Study (RO1CA64277). Genetic polymorphisms in approximately 25 candidate genes will initially be evaluated using both genotype- and haplotype-approaches in approximately 1200 cases and 1200 controls recruited from 1996 to 1998. Promising associations identified in the initial phase will be re-evaluated in the second set of subjects (1300 cases and 1300 controls) recruited since 2002 in the parent study. This 2-phase study design will effectively balance both Type I and Type 2 errors and provide credible results towards our understanding of the etiology of breast cancer. Results from this study will be valuable in identifying high risk women for primary and secondary prevention of breast cancer. Funded by the  National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI) ; Xiao Ou Shu, M.D., Ph.D.; Qiuyin Cai, Ph.D., M.D.; Jirong Long, Ph.D.; Jay H. Fowke,  M.P.H., Ph.D.;  Shimian Qu, Ph.D.; Chun Li, Ph.D.

Molecular and Genetic: Genetic Factors for Breast Cancer 
Breast cancer is the most common malignancy among women in many parts of the world. Genetic factors play an important role in the etiology of breast cancer. However, to date, only a few breast cancer susceptibility genes have been identified, and they explain only a very small fraction of breast cancer cases in the general population. A large number of candidate-gene studies have been conducted over the past 10 years. These studies, however, are clearly inadequate to fully uncover the genetic basis of breast cancer. With recent significant advances in high-throughput genotyping technologies, it has become feasible to conduct genome-wide association (GWA) studies to systematically evaluate genetic risk factors for breast cancer. This multi-phase GWA study is built upon the resources established in two large, ongoing studies funded by NCI, the Shanghai Breast Cancer Study, a population-based case-control study, and the Shanghai Women´s Health Study, a population-based prospective cohort study. Approximately 8,000 breast cancer cases and controls are included in this study. In the first phase of the study, we are conducting a GWA scan in 1,000 cases and 1,000 controls using the Illumina HumanHap550 Beadchip. We will then select the 10,600 most promising SNPs for a validation study in an independent sample of 1,500 cases and 1,500 controls. All promising SNPS will be further validated using data from 1,000 cases and 2,000 controls selected from the prospective Shanghai Women´s Health Study. This study has many unique features that facilitate a rigorous evaluation of breast cancer genetic factors. The results from this study will be valuable in identifying high risk women for primary and secondary prevention of breast cancer.

Nashville Breast Health Study 
This proposal is for a population-based case-control study of breast cancer in Nashville, Tennessee. The primary hypotheses are 1) regular use of nonsteroidal anti-inflammatory drugs (NSAID) may reduce the risk of breast cancer, and this association may be modified by the genotypes of NSAID metabolizing enzymes; 2) well-done (charred) meat intake, and thus exposures to the mammary carcinogens heterocyclic amines and polycyclic aromatic hydrocarbons, may be related to an increased risk of breast cancer, particularly among women with certain genotypes of the carcinogen-metabolizing enzymes; 3) the positive association between well-done meat intake and breast cancer risk may be modified by regular NSAID use; 4) certain polymorphic genes involved in estrogen metabolism may interact with each other in the etiology of breast cancer. We propose to recruit 1,500 incident cases and 1,500 controls for this case-control study. Breast cancer cases will be identified through a rapid case-ascertainment system established for the study. Controls will be selected randomly from the general population and frequency-matched to cases by age and race. Telephone interviews will be conducted to obtain relevant exposure information. Exfoliated buccal cell samples will be collected to extract DNA for analyzing 38 polymorphisms in 14 candidate genes that are involved in the metabolism of NSAIDs, mammary carcinogens, and estrogens. DNA samples will also be stored for future studies of additional genetic factors and their interactions with lifestyle factors in the risk of breast cancer. NSAIDs are among the most commonly used medications, and high-temperature cooking has been widely used for meat preparation. Information regarding their associations with breast cancer risk could have important public health implications in the primary prevention of breast cancer. Studies investigating gene-gene and gene-environment interaction could provide valuable information in identifying high-risk individuals for designing cost-effective preventive strategies for breast cancer. Funded by the National Cancer Institute 
Investigators:  Wei Zheng, M.D., Ph.D. (PI); Mark C. Kelley, M.D.;  Sandy Deming, Ph.D.; Xiao Ou Shu, M.D., Ph.D.;  Martha J. Shrubsole, Ph.D.; Qiuyin Cai, Ph.D., M.D.;  Jirong Long, Ph.D.;  Fritz F. Parl, Ph.D., M.D.;  Bill Dupont, Ph.D.;  David Page,  M.D.

Nashville Men’s Health Study: Genetic and Endocrine Pathways Linking Obesity with Prostate Cancer 
Prostate cells respond to estrogens, insulin, and other factors largely regulated in men by adipose mass. Several recent studies report obesity associated with high-grade prostate cancer, progression, and mortality, however the association with low-grade cancer common in the PSA era remains unclear. Challenges include measuring fat deposition patterns, excluding latent cancer from control groups, and controlling for several potential biases associated the effects of obesity on prostate cancer detection. Our study aims to address these challenges and determine the relationship between total adiposity (e.g., BMI, estrogens) and visceral adiposity (e.g., waist circumference, waist-to-hip ratio (WHR), insulin) across high-grade cancer, low-grade cancer, and prostatic intraepithelial neoplasia (PIN). Preliminary analyses (R21 CA98348, n=304 cancer, 120 PIN, 424 controls) found WHR significantly associated with PIN (WHR>1.03: OR = 4.75 95% Cl (1.71, 13.2), ptrend<0.01, adjusted for PSA, BMI, prostate volume, age race, ORE result, # cores). Also, BMI>35 was associated with high-grade (Gleason score=7) cancer (ORadj=3.49 (0.84, 14.4), ptrend = 0.05). Thus, visceral adiposity and the related metabolic syndrome may impact early prostate carcinogenesis, while an estrogen- rich environment associated with greater BMI may accelerate progression to high-grade/clinically relevant disease. Using our established multi-centered rapid-recruitment protocol, we will recruit an additional 1,106 prostate cancer cases (42% Gleason score=7), 435 PIN cases, and 1,544 controls without cancer or PIN at prostate biopsy. Data and specimens (questionnaires for diet, physical activity, and other risk factors; body measures for BMI, WHR, sitting height, and percent body fat (BIA); blood for DNA and hormone levels) are collected before diagnosis. Genes representing pathways linking total adiposity (e.g., Lep, LepR, CYP19, ER,AR, SHBG) or visceral adiposity (Res, Adip, AdipR1/2, INS, IRS1/2, IGF1, IGFBP3, PPARy2) to PIN or cancer will be investigated using multivariable logistic regression. Also, we will investigate blood markers of adiposity and PIN in an individually matched analysis (total adiposity: leptin, E2/T ratio, SHBG; visceral adiposity: HbA1c, adiponectin, resistin). Obesity is epidemic in the U.S., and prostate cancer is a leading cause of cancer-related death. Ongoing chemoprevention studies target PIN, and our results may identify new obesity-based prevention approaches or improve the prognosis of prostate cancer patients. Funded by the National Cancer Institute 
Investigators: Jay H. Fowke, M.P.H., Ph.D. (PI); Joseph Smith, M.D.; Saundra Motley, R.N. ; Raoul Concepcion, M.D.

Nutritional: Soy Food and Coronary Heart Disease in Women
This is an ancillary study of the Shanghai Women´s Health Study (SWHS) aiming to test the hypothesis that soy food consumption decreases the risk of coronary heart disease (CHD) and CHD death, particularly among women who have a high risk of developing CHD. This study includes two components: 1) a longitudinal analysis of dietary soy intake and CHD risk and 2) a nested case-control study evaluating the association of urinary isoflavonoid levels, biomarkers of soy intake with CHD risk. The modification effect of blood lipid profile and C-reactive protein, known risk factors for CHD, on the soy-CHD association will also be evaluated.

Oxidative Stress, Antioxidants, and Breast Cancer Risk
Substantial evidence from laboratory studies has indicated an important role of reactive oxygen species (ROS), produced by estrogen metabolism, in mammary carcinogenesis. This application is to evaluate whether the biomarker of systemic oxidative stress may be associated with breast cancer risk, whether antioxidant polyphenols, consumed in large quantities among Chinese women in Shanghai, may reduce the risk of breast cancer, particularly among those with a high exposure to ROS. To achieve these research goals, we propose to conduct a nested case-control study of 430 breast cancer cases and 860 individually matched controls within the cohort of the Shanghai Women’s Health Study (RO1CA 70867), a population-based cohort study of approximately 75,000 Chinese women. In addition to in-person interviews, blood (or buccal cell) and urine samples were collected from 87.5% cohort members. For this proposed study, urinary level of total phenols will be assayed with Folin-Ciocalteu Phenol Reagent, urinary levels of polyphenol flavonols and flavanols will be determined using liquid chromatography photo-diode array electrospray mass spectrometry, urinary level of 2,3-dinor-5,6-dihydro-15-F2t-lsoP will be quantified by gas chromatography/negative ion chemical ionization mass spectrometry, blood levels of tocopherols and carotenoids will be analyzed using high performance liquid chromatography (HPLC) methods, and genotypes of the enzymes involved in ROS detoxification will be assayed using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) methods. The associations between these biomarkers and breast cancer risk will be evaluated. We also will perform statistical analyses to evaluate the joint effects or potential interactions of total phenols, flavonols and flavanols with soy isoflavonoids, antioxidant nutrients, genotypes of ROS detoxification enzymes and oxidative stress level in relation to breast cancer risk. The study population will be unique given the high intake levels of total phenols, flavonols, flavanols, isoflavonoids. Because questionnaire data and specimen collection, follow-up survey as well as assays for isoflavonoids are supported by the existing study, this project will be very cost-efficient. In addition, we will also be able to expand our study to a larger scale if the results are promising. Findings from this study will have important public health implications in the primary prevention of breast cancer. Funded by the National Cancer Institute
Investigators: Qi Dai, M.D., Ph.D. (PI); Qiuyin Cai, M.D., Ph.D.; Ginger Milne, Ph.D.; Xiao-Ou Shu, M.D., Ph.D.; Wei Zheng, M.D., Ph.D.

Parent Child Communication about Cancer 
Over 12,400 children age 19 and under are diagnosed with cancer in the United States each year, and 2,300 die annually from the disease (Ries et al., 2004). One of the most difficult questions facing parents of children with cancer is, “How do I talk with my child about cancer?” Further, parents of children with a poor prognosis struggle with the question, “How do I talk with my child about the possibility of death?” Parents often turn to health care providers for guidance, but ultimately parents must determine the type and amount of information they share with their child about the disease, treatment, and prognosis. Further, parents must decide how to balance this information in the context of managing their own and their child’s fear and worries. Although the National Cancer Institute (NCI) has recommended that parents communicate openly and honestly about cancer, little research has addressed this issue in pediatric oncology. Moreover, no studies have used direct observation of parents and children as they discuss the diagnosis or examined how communication is associated with coping and adjustment over time. This application will use questionnaires and direct observation to assess parent-child communication about cancer and determine how the content and process of communication are associated with family outcomes. More than 560 families of children newly diagnosed with cancer or a recurrence will be recruited from two sites to complete questionnaires about coping, communication and adjustment 1 month after the child’s diagnosis. Approximately 224 families are expected to complete an observation of family communication 3 months after the child’s diagnosis, and follow-up questionnaires will be completed 12 months after diagnosis. We will pursue 3 aims: 1) To use a standardized observation protocol to describe the content and process of parent-child communication about cancer and examine differences in communication as a function of medical (for example, diagnosis, prognosis), child (for instance, age, gender), and family (for example, history of loss, family environment) factors; 2) To examine associations between communication and distress in parents and children; 3) To examine associations between communication and children’s subsequent coping and emotional distress. The study has potentially significant public health implications, as findings from this study will enable health care professionals to better guide parents on how to communicate with their child in order to optimize their child’s adjustment and care by providing guidelines about the content and quantity of information to share. Thus, the long-term goal of the proposed research is to provide health care professionals with evidence based recommendations for facilitating parent-child communication regarding childhood cancer and to decrease children’s and parents’ psychological distress. Funded by the  National Institute of Mental Health 
Investigators:  Bruce E. Compas, Ph.D.(PI)

Personalized Prevention of Colorectal Cancer
High calcium intake and magnesium may protect against colorectal cancer and adenoma, however, results have been inconsistent. Although the mean magnesium intake in the US population is similar to East Asian populations with traditionally low risks of colorectal cancer, the ratio of calcium to magnesium is much higher in the US. We propose to conduct an intervention trial of 240 participants to investigate the efficacy of modulating the dietary ratio of calcium to magnesium to change markers directly related to tumorigenesis, including apoptosis biomarkers (e.g. TUNEL and Bax), COX-2 (inflammation), Ki-67 (proliferation index), and TRPM7/TRPM6 in colorectal mucosa as well as total erythrocyte magnesium and urinary excretion of prostaglandin E2 metabolite (PGE-M) as primary endpoints. The progressive resistance to apoptosis is one hallmark for almost all cancer types. The apoptosis index is a strong predictor of future adenoma occurrence. The resistance to apoptosis is accompanied by an elevation in COX-2 expression during tumorigenesis. We found in a population-based cohort study that urinary levels of prostaglandin E2 metabolite (PGE-M) were associated with a substantially increased risk of colon and rectal cancers. Urinary level of PGE- M was also elevated among participants with large adenomas compared to those who had either no or small polyps. The primary aims of this study are to conduct a randomized placebo-controlled intervention trial to test whether reducing the calcium to magnesium intake ratio through supplementation of magnesium has effects on the above-mentioned biomarkers. Furthermore, we will examine whether the effect of modulating dietary intake ratio of calcium to magnesium may be more pronounced among those who carry the 1482Ile allele (GA or AA) compared those who do not carry the 1482Ile (GG). If findings from the study are promising, we will propose to conduct a large-scale clinical trial using recurrence of flat, depressed, and polypoid colorectal adenomas or colorectal cancer as clinical endpoints. The results from our study may ultimately help to develop personalized strategies to prevent the occurrence of colorectal adenoma, and, thus, colorectal cancer. In the general US population, 1 in 18 individuals will develop colorectal cancer over their lifetime and forty percent will die within five years of diagnosis, mainly due to diagnosis at a late stage. Therefore, development of primary preventive strategies for colorectal cancer is very critical. The results from our study will help to identify people at a high risk of colorectal adenoma and to develop personalized strategies to prevent occurrence of colorectal adenoma, and, thus, colorectal cancer through dietary changes or nutritional fortification. Funded by the  National Cancer Institute
Investigators:   Qi Dai, M.D., Ph.D.   (PI)  

Pharmacoepidemiology
The Division of Pharmacoepidemiology in the Vanderbilt Department of Preventive Medicine is one of the first pharmacoepidemiology units in the country. Divisional research is mainly performed using automated databases that contain information on drug exposures and clinical outcomes in large populations. Although the computerized databases are extremely valuable for epidemiologic studies, many studies require review of medical records for case confirmation, establishment of time of disease onset, and identification of patient risk factors. The Division has arrangements in place with the more than 100 acute care hospitals in Tennessee to obtain and review medical records for research studies. The Vanderbilt University Center for Education and Research on Therapeutics (CERT) is one of nine independent centers, selected through a highly competitive peer-review process and sponsored by the Agency for Healthcare Research and Quality (AHRQ). The CERT´s mission is to conduct research and provide education that will advance the optimal use of drugs, medical devices and biological products. Vanderbilt is also one of the thirteen Developing Evidence to Inform Decisions about Effectiveness (DEcIDE) research centers in the country. The main purpose of the DEcIDE network is the timely development of valid scientific evidence and methodologies about the outcomes, comparative clinical effectiveness, safety, and appropriateness of health care items and services for improving the quality, effectiveness, and efficiency of healthcare delivered through the Medicare, Medicaid, and SCHIP programs. Pharmacoepidemiology research is also conducted by the Vanderbilt Center for Health Services Research, which currently has more than 30 core faculty and more than 50 affiliated faculty with over $70 million in extramural funding. A current pharmacoepidemiology study, which has potential opportunities for a doctoral student, is assessing the safety of biologic disease modifying anti-rheumatic drugs (DMARDs) in patients with rheumatoid arthritis, including the risks and benefits. Conditions under investigation include serious infections, lymphoproliferative neoplasias, coronary heart disease, and diabetes. The cohort of patients will include veterans with rheumatoid arthritis, and data will encompass 13 years and include nearly 40,000 participants.

Pharmacogenomics in Lung Cancer Therapy: A Pilot Study 
The major aim of this pilot study is to evaluate the feasibility of conducting a full-scale pharmacogenomic study in lung cancer patients diagnosed and/or treated at the Vanderbilt University Hospital and Nashville VA Medical Center. Funded as a pilot project in the Vanderbilt SPORE in lung cancer
Investigators:  Qi Dai, M.D., Ph.D. (PI); Wei Zheng, M.D., Ph.D.; David H. Johnson, M.D.; David Carbone, M.D., Ph.D.

Racial Differences in Environmental and Genetic Determinants of Obesity: A Major Risk Factor for Breast Cancer
The main goal of this research is to describe genetic and environmental determinants of obesity and adult weight gain that relate to breast cancer among African American and Caucasian women in the Southern Community Cohort Study. Funded by the  Susan G. Komen Foundation
Investigators:   Charles E. Matthews, Ph.D. (PI) ,  William Blot, Ph.D. (PI)

Reproductive Epidemiology: Pregnancy Health and Outcomes
Right From the Start (RFTS) is an ongoing, community-based prospective cohort study with the goal of advancing knowledge about maternal and fetal health from conception to birth. Since 1999, RFTS has enrolled more than 5,500 women, 20% prior to conception, in order to study determinants of fecundability, miscarriage, spontaneous preterm birth, and other adverse pregnancy outcomes. While we have focused on varied primary exposures of interest including water disinfection by-products, uterine fibroids, caffeine consumption, and physical activity, the core dataset includes a broad range of carefully measured covariates and putative risk factors for adverse outcomes that remain to be examined. We have more than 4,800 completed pregnancies in the core dataset. This includes information collected at enrollment, results of a research ultrasound conducted around 6 weeks gestation, a detailed computer-assisted-telephone interview in the first trimester, follow-up during later pregnancy, medical records abstractions, and for some women, an interview after pregnancy. Data quality control is strong and the data set has reached ample size to have good power to investigate additional important etiologic and clinically relevant questions. Areas available for investigation are numerous and include topics like racial/ethnic disparities in miscarriage risk; the relationship of alcohol consumption to a range of outcomes including miscarriage, growth restriction, and preterm birth; and methodological aspects attributing exposure time relative to ultrasound documentation of arrest of embryonic development among losses. We are eager to welcome a graduate student interested in reproductive epidemiology to our research team.

Shanghai Breast Cancer Study (renewal) 
Breast cancer is the most common malignancy among women in the United States and many other parts of the world.  It is believed that most breast cancers are caused by genetic factors and environment interactions.  Over the past 10 years many genetic polymorphlsms have been investigated in relation to breast cancer risk, yet few of the associations found have been confirmed.  As part of this renewal application for the Shanghai Breast Cancer Study (SBCS R01CA64277), a large population-based case-control study of breast cancer funded by NCI since 1996, we propose several novel approaches to circumvent the limitations of the methodologies currently used for studying low-penetrance genetic variants for the risk of complex, multifactorial diseases.  First, in addition to studying common genetic variants using haplotype tagging SNPs (htSNP), we propose to test the hypothesis that a large part of breast cancer susceptibility may be due to the summation of the effect from multiple low-frequency genetic variants.  We also propose to use re-sequencing data to enhance htSNP selection and reduce misclassification errors and use quantitative functional data to define risk groups.  With these new approaches, we will comprehensively evaluate genetic polymorphisms in the TGF2 signaling pathway in a two-phase study including approximately 6,200 cases and controls.  Genetic variants in major TGF2 pathway genes will be screened in Phase I, and promising associations will be validated in Phase II along with a series of in vitro functional assays.  Breast cancer patients are being followed for cancer recurrence, relapse, and death, and the association of TGF2 pathway gene variants with breast cancer survival will be evaluated in a two-phase study.  Numerous in vitro and animal studies have clearly demonstrated that the TGF2 signaling pathway plays a pivotal role in the development and progression of breast cancer.  It remains unclear, however, whether these laboratory findings can be translated into cancer prevention strategies and clinical practice.  The proposed study, with its strong methodology and novel approaches, has outstanding potential for discovering genetic markers that will be valuable in identifying high-risk women for cost-efficient breast cancer prevention and personalized treatment and follow-up care after cancer diagnosis.  The proposed novel approaches will not only facilitate the evaluation of the study hypotheses, but also provide significant data to expand and guide future studies of low-penetrance genetic factors for breast cancer and other complex, multifactorial diseases.  Funded by the  National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI) ; Xiao Ou Shu, M.D., Ph.D.; Qi Dai, M.D., Ph.D.; Qiuyin Cai, Ph.D., M.D.; Jirong Long, Ph.D.;  Chun Li, Ph.D.

Shanghai Breast Cancer Study 
This is an ongoing study funded since 1996 to investigate genetic and lifestyle factors as well as other biomarkers for breast cancer risk and survival. Almost all epidemiologic studies of breast cancer have been conducted in Western populations with homogeneous and high exposure to certain hypothetical risk factors; the narrow range of exposure has substantially reduced efficiency in these studies. A population based case-control study is proposed among Chinese women in Shanghai where the incidence rate of breast cancer has increased dramatically in the past two decades. The investigators state that the unique lifestyle pattern and diverse exposures in this population will facilitate a rigorous examination of some important etiologic hypotheses that cannot be adequately addressed among women in the United States. The primary aims of this study are to investigate the associations of breast cancer with dietary fat intake across the range 14% to 36% of calories (median of lowest to highest quintile), nutritional status during adolescence (including adolescent diet and certain anthropometries), estrogen metabolic pattern, body mass and fat distribution, oral contraceptive use, induced abortions, and breast feeding. The study will include a total of 1200 incident breast cancer cases aged 25-64 and an equal number of controls (frequency matched to cases on age) randomly selected from the general population in Shanghai. Cases will be identified through two well tried rapid case finding systems and the population based Shanghai Cancer Registry. In person interview data, anthropometric measurements and fasting blood and urine will be collected. For 500 pairs of cases and controls, urine samples will be assayed for levels of 2 and 16-alpha hydroxyestrones. Tumor tissue blocks will be collected for assays of estrogen and progesterone receptors to assess whether receptor defined subgroups of breast cancer are etiologically distinct. The remaining biospecimens will be stored at -70C for future molecular epidemiologic studies of serum organochlorine pesticides, somatic mutations in tumor tissue, genotypes of HRAS and certain metabolizing enzymes using DNA from white blood cells and other biomarkers. The investigators state that this study will provide valuable information on the interplay of hormonal, genetic, dietary, environmental, and lifestyle factors in the development of breast cancer. They further state that it will also create a valuable resource for future studies of genetic factors and gene-environmental interaction in the etiology of breast cancer. Funded by the  National Cancer Institute 
Investigators:  Wei Zheng, M.D., Ph.D.(PI) ; Xiao Ou Shu, M.D., Ph.D.; Qi Dai, M.D., Ph.D.; Qiuyin Cai, Ph.D., M.D.; Jirong Long, Ph.D.; Chun Li, Ph.D.

Shanghai Endometrial Cancer Study (SECS) 
Estrogen plays a central role in the etiology of endometrial cancer. The association of exogenous estrogen use and high endogenous hormone exposure with endometrial cancer has been well characterized. The effect of soyfoods, rich source of phytoestrogens that have both weak estrogenic and anti-estrogenic effects, has only been studied in two studies and results were inconsistent. Given the dual effect of phyotoestrogens, we hypothesize that the effect of phytoestrogens on endometrimum depends on the levels of endogenous estrogens. The level and biological effect of estrogens are determined by multiple genes and thus the risk of endometrial cancer may be associated with genetic polymorphisms of genes involved in estrogen biosynthesis, metabolism, binding and signaling, and their joint effect with soyfood intake and other lifestyle factors. To evaluate the above hypotheses, we propose to conduct a population-based case-control study including 1150 incident cases and 1150 age-matched controls in urban Shanghai. In-person interviews will be conducted to collect dietary and other exposure information. A 10-ml peripheral blood sample (or a buccal cell sample if blood sample could not be obtained) will be collected from all cases and controls. Genomic DNA will be analyzed for the genotypes of the genes involved in estrogen biosynthesis (CYPI7, polymorphisms of genes involved in estrogen biosynthesis, metabolism, binding and signaling, and their joint effect with soyfood intake and other lifestyle factors. To evaluate the above hypotheses, we propose to conduct a population-based case-control study including 1150 incident cases and 1150 age-matched controls in urban Shanghai. In-person interviews will be conducted to collect dietary and other exposure information. A 10-ml peripheral blood sample (or a buccal cell sample if blood sample could not be obtained) will be collected from all cases and controls. Genomic DNA will be analyzed for the genotypes of the genes involved in estrogen biosynthesis (CYPI7, CYP19, and HSD17B1), inactivation (SUTL1AI, UGT1, and COMT), binding (SHBG), and signal transduction (ER-a and ER-b). Associations of endometrial cancer with soyfood intake and polymorphisms of above-mentioned genes will be evaluated separately, jointly and in conjunction with conditions related to estrogen levels (e.g., obesity, physical activity, dietary fat intake, and menopausal status). Incidence rate and prevalence rates of traditional risk factors (e.g., estrogen replacement therapy, obesity, nulliparity) of endometrial cancer are considerable low among women in Shanghai than their counterparts in the US, and this will minimize potential confounding effects in testing new hypotheses. Consumption level of soyfood is high and hysterectomy rates are extremely low among Chinese women, providing a unique opportunity to test the hypotheses posed in the application that are difficult to be evaluated in the U.S. population. Funded by the polymorphisms of above-mentioned genes will be evaluated separately, jointly and in conjunction with conditions related to estrogen levels (e.g., obesity, physical activity, dietary fat intake, and menopausal status). Incidence rate and prevalence rates of traditional risk factors (e.g., estrogen replacement therapy, obesity, nulliparity) of endometrial cancer are considerable low among women in Shanghai than their counterparts in the US, and this will minimize potential confounding effects in testing new hypotheses. Consumption level of soyfood is high and hysterectomy rates are extremely low among Chinese women, providing a unique opportunity to test the hypotheses posed in the application that are difficult to be evaluated in the U.S. population. Funded by the  National Cancer Institute 
Investigators:   Xiao Ou Shu, M.D., Ph.D. (PI) ; Wei Zheng, M.D., Ph.D.; Qi Dai, M.D., Ph.D. ;  Wanqing Wen, M.D.; Qiuyin Cai, M.D., Ph.D.;  Jirong Long, Ph.D.

Shanghai Men’s Health Study (SMHS) 
While it has long been recognized that dietary factors play an important role in the etiology of cancer, information regarding the preventive potential of specific dietary compounds is scarce. The Shanghai Men’s Health Study (SMHS) is a population-based cohort study of 61,582 Chinese men between 40 and 74 who lived in urban Shanghai at enrollment. Because intake levels of many suspected cancer-inhibitory dietary factors are high and diverse in Shanghai, the SMHS offers unique opportunities to fill such knowledge gaps. Detailed information on dietary and other lifestyle factors was collected at baseline and is being updated in a follow-up survey. Biological samples were collected from 89% of cohort members. The cohort has been followed for cancer occurrence and deaths via linkage with the population-based Shanghai Cancer Registry and the Shanghai Vital Statistics Unit, as well as through visits to all living cohort members every 2 years. We propose in this renewal application to extend the follow-up of this cohort for 5 more years and to evaluate dietary hypotheses for the lung, stomach, and colorectal (CRC). The primary focus of the study is to determine whether regular tea consumption and high intake of folate, soy foods, allium vegetables, and crucifers are associated with a reduced risk of cancer. We also propose to conduct a nested case-control study to evaluate whether the levels of blood folate, urinary phytoestrogens (CRC only) and isothiocyanate are inversely associated with the risk of cancers of colorectum, lung, and stomach. These biomarkers are aggregate measures of level of intake, absorption, and metabolism and will provide added insight in elucidating the relationship of dietary factors with cancer risk. Blood level C-reactive protein and H.pylori antibodies will also be evaluated in the nested case-control study for CRC and stomach cancer. Finally, we propose to re-survey all living cohort-members to update information on usual dietary intake and other lifestyle factors to refine exposure assessments and to characterize and evaluate how temporal changes in exposures may influence cancer risk. Because of its size, setting, and unique exposure patterns and biological specimens, the SMHS provides an exceptional opportunity to address dietary hypotheses for cancer that cannot be adequately investigated in any other existing cohort study. The results from this study may guide new strategies in the primary prevention of common cancers in both Western and Asian men. Funded by the National Cancer Institute 
Investigators:   Xiao Ou Shu, M.D., Ph.D. (PI) ; Wei Zheng, M.D., Ph.D.; Gong  Yang, M.D., M.P.H.; Wanqing Wen, M.D. ; Qiuyin Cai, Ph.D., M.D.; Yu Shyr, Ph.D.

The Shanghai Women’s Health Study (SWHS)
This is a large population-based prospective cohort study initiated in 1996. From 1996 to 2000, approximately 75,000 Chinese women who lived in Shanghai were recruited into the study (AJE, 162:1123, 2005). In addition to survey data, most study participants donated blood and urine samples at baseline. This cohort of women are being followed for the occurrence of cancer and several common diseases through biennial home visit and record linkage with files routinely collected by the Shanghai Cancer Registry and the Vital Statistical Unit. In the current funding cycle, the study focuses on evaluating dietary factors that may reduce the risk of cancers. The resources from this study have supported multiple projects that address etiologic hypotheses for cancers, and other chronic diseases, such as coronary heart diseases, asthma, diabetes, stroke, hypertension, and bone fracture. Funded by the  National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI) ; Xiao-Ou Shu, M.D., Ph.D.; Gong Yang, M.D., M.P.H.; Wanqing Wen, M.D., M.P.H.; Qiuyin Cai, M.D., Ph.D.; Jay H. Fowke, M.P.H., Ph.D.

Southern Community Cohort Study (SCCS) 
It has long been known that cancer incidence and mortality are elevated among African Americans. The Southern Community Cohort Study (SCCS) is a landmark prospective investigation into the determinants of these disparities. Over the past five years, by partnering with Community Health Centers (CHCs), facilities providing basic health care mainly to the uninsured across 12 southern states, we have overcome barriers that have traditionally restricted participation of African Americans in health studies. The current phase of the SCCS, proposed herein, will enable completion of enrollment so that the cohort will include approximately 90,000 men and women, nearly 70% African American, age 40-79. The new enrollees will be recruited from CHCs and complete an in-person interview about medical, lifestyle and other characteristics, with > 90% expected to provide biologic specimens (blood, buccal cells, and/or urine). Follow up of the entire cohort to identify deaths and incident cancers and update exposure profiles will be carried out. We will implement enhanced approaches to active follow up. Longitudinal analyses and nested case-cohort studies using the interview data and biologic specimens will be initiated during this funding period to evaluate specific hypotheses about cancer among African Americans which can uniquely or with special advantage be assessed within the SCCS. The hypotheses are related to energy balance (weight gain, obesity and physical inactivity), vitamin D, inflammation, selenium and other nutrient intakes, tobacco metabolism, and cancer screening practices and their impact on incidence and/or mortality of the major cancers (lung, prostate, breast, and colon/rectum). Initial SCCS data show sometimes marked racial differences in these variables. The cohort possesses attributes, such as a 44% prevalence of obesity (reaching 57% among Black women) and a 44% prevalence of current smoking which place it at exceptionally high risk of cancer. The SCCS is thus unique among all cohorts and comprises a population of urban and rural blacks and whites, often of low income, seldom if ever included in previous studies. The SCCS is a national resource, with comprehensive biologic and questionnaire data available for assessing, both within the SCCS and in combination with other molecular epidemiology consortia, the etiology of cancer and reasons for the largely unexplained higher rates among blacks. The ultimate public health benefit will be progress towards the development of measures aimed at cancer prevention, the elimination of cancer inequalities, and reduction of the cancer burden among all groups. Funded by the  National Cancer Institute 
Investigators:  William Blot, Ph.D. (PI) ; Lisa Signorello, Ph.D. (PI); Maciej Buchowski, Ph.D.; Raymond Burk, M.D.; Qiuyin Cai, M.D., Ph.D.; Robert Dittus, M.D., MPH; Jay H. Fowke, M.P.H., Ph.D.; Jirong Long, Ph.D.; Chun Li, Ph.D.; Charles E. Matthews, Ph.D.; Pierre Massion, M.D.; Joseph McLaughlin, Ph.D.; Neeraja Peterson, M.D.; David Schlundt, Ph.D.; Martha Shrubsole, Ph.D.; Xiao Ou Shu, M.D., Ph.D.; Robert Tarone, Ph.D.; Scott Williams, Ph.D.; Wei Zheng, M.D., Ph.D.

Tennessee Colorectal Polyp Study (TCPS) 
Colorectal cancer is one of the most common malignancies in the United States and many other countries. Most colorectal cancers arise from adenomatous polyps. As part of the Vanderbilt SPORE in GI cancer, we are conducting a colonoscopy-based study to evaluate biomarkers and lifestyle factors for colorectal polyps. As of November 2007, nearly 6,000 patients were recruited into the study. Most of the study participants completed a detailed survey of lifestyle factors and provided biologic samples to the study. The resources established in this study have supported multiple research projects addressing biomarkers and risk factors for colorectal polyps. Funded by the  National Cancer Institute 
Investigators:   Wei Zheng, M.D., Ph.D. (PI) ; Reid M. Ness, M.D.; Martha J. Shrubsole, Ph.D;  Walter E. Smalley, M.D., M.P.H.; Shimian Qu, Ph.D.

Tumor Markers and Recurrent Adenomas: A Follow-up Study 
Most colorectal cancers arise from adenomatous polyps, and a large proportion of patients with adenomas will develop recurrent adenomas. There is considerable controversy regarding the appropriate surveillance interval following initial colonoscopy. Studies assessing predictors for recurrent adenomas will provide valuable information for designing individualized surveillance strategies, particularly for patients with either multiple adenomas or pathologically advanced adenoma. We propose in this application to recruit and follow 2000 patients diagnosed in 1996 to 2001 with incident multiple or advanced adenomas to evaluate the utility of a panel of promising tumor markers in predicting the risk of adenoma recurrence. The tumor markers proposed for this study reflect major events that occur during the formation and progression of adenomas. Specifically, we will evaluate the following four groups of tumor markers in relation to the risk of adenoma recurrence: 1) proliferation and apoptosis, including the apoptosis index (TUNEL assay) and the expression of Ki-67 (Mibl), epidermal growth factor receptor (EGFR), and transforming growth factor B receptor type II (TGF-J3 RI]); 2) genomic instability - loss of heterozygosity (LOH) events on chromosomes 5q, l7p, 15q, ip, and 18q; 3) Wingless/Writ signaling pathway - expression of the CTTNB1 (Beta-catenin gene), Cyclin D1 CMYC, and COX2 gene products; and 4) DNA methylation - methylation status of the promoters of the MLH1, MGMT, CDKN2A/P16, and APC. Study patients will be followed through a combination of telephone interviews and medical chart reviews. Paraffin-embedded blocks of initial adenomas will be retrieved for bioassays of tumor markers. The diagnosis of initial and recurrent adenomas will be reviewed and confirmed by study pathologists. This study is likely to provide valuable information for identifying high-risk adenoma patients for close surveillance and chemoprevention. Funded by the  National Cancer Institute 
Investigators:  Wei Zheng, M.D., Ph.D. (PI) ; Reid M. Ness, M.D.; Martha J. Shrubsole, Ph.D; Walter E. Smalley, M.D., M.P.H.; Kay Washington, Ph.D, M.D.; Bill Grady, M.D.; Ayumi Shintani, Ph.D., M.P.H.; Robert Dittus, M.D., M.P.H.