Visiting Scholars Seminar Series - 2010-2011 *
Spotlighting the Research of EGRP-funded grantees
The EGRP Visiting Scholars Seminar Series brings outstanding scientists in the extramural population sciences community to NCI to share their latest research and facilitates an exchange of ideas on ways to continue moving the science forward. The seminars, which are held in Rockville, Maryland are open to the public. View directions to the Executive Boulevard campus.
The speakers during academic year 2010-2011 are:
Seminar Series Contact: Rao Divi, Ph.D. phone: 301-443-5539 e-mail: divir@mail.nih.gov |
||
October 18, 2010 |
 |
Association of Prostate-specific Antigen and Kallikrein-related Peptidase 2 with Long-term Risk and Outcome of Prostate Cancer (summary) |
November 8, 2010 |
 |
Exploring the Uptake and Effectiveness of KRAS Testing for Metastatic Colorectal Cancer: Can We Improve Health Outcomes and Influence Decision-Making in Personalized Medicine? (summary) |
December 13, 2010 |
 |
International Epidemiologic Research - A Unique Population Laboratory for Studying the Cause of Cancer (summary) |
January 10, 2011 |
 |
Why Study Familial Chronic Lymphocytic Leukemia? (summary) |
February 7, 2011 |
 |
The DNA Methylome Dynamics in Aging and Cancerc (summary) |
March 14, 2011 |
 |
SNPs in IGF Signaling Genes and Breast Cancer Risk (summary) |
April 11, 2011 |
 |
Integrated Genomic and Environmental Epidemiology Research in Cancer (summary) |
May 9, 2011 |
 |
The NYU Women’s Health Study: Results on Risks of Breast and Endometrial Cancers (summary) |
* Some session titles have not yet been finalized and will be announced.
October 2010
Hans Lilja,
M.D., Ph.D.
Hans G. Lilja, Attending Research Clinical Chemist in the Departments of Clinical Laboratories, Surgery (Urology Service), and Medicine (Genitourinary Oncology Service) at Memorial Sloan-Kettering Cancer Center, delivered the October Visiting Scholars Seminar Series presentation titled “Association of Prostate-Specific Antigen (PSA/KLK3) and Kallikrein-Related Peptidase 2 (hk2/KLK2) With Long-Term Risk and Outcome of Prostate Cancer.”
Dr. Lilja explained that although PSA-based prostate cancer screening can reduce cancer mortality, the test has a high percentage of false positives and has a poor ability to distinguish aggressive cancers from indolent ones. Thus, it has led to overdetection of cancers that pose little threat to a man’s life.
New ways of using the PSA test may increase its value in the detection of life-threatening prostate cancers. Dr. Lilja described analyses performed on archived blood samples from a cohort of men who participated in a cardiovascular study in Sweden at a time when PSA testing rates were low. In these men, a single measure of PSA at or before age 50 predicted diagnosis of advanced prostate cancer up to 25 years later. These findings indicate that a single baseline PSA can be used to identify men at high and low risk for prostate cancer and to stratify them for prostate cancer screening and chemoprevention. Data from the same study also indicated that PSA levels at age 60 strongly predict lifetime risk of metastasis and death from prostate cancer. Many men can be exempted from further screening and chemoprevention on the basis of a low PSA level at this age.
Whether to perform a biopsy on a man with an elevated PSA level is an important clinical question. Attempts to set decision criteria based on higher PSA thresholds or on the velocity of the increase in PSA levels have been disappointing. Dr. Lilja described another approach, involving a panel of four kallikrein markers. In a wide variety of cohorts, the use of this panel has consistently been shown to accurately predict the outcome of prostate biopsy. Its use could improve clinical decision-making; a higher ratio of cancers would be found compared to the number of biopsies conducted.
Future directions in research in this field will include investigations of biomarker interaction with genetic variation in various cell lines and genetically engineered mouse models. There is a strong interaction between the rs198977 genotype and the kallikrein-related peptidase hk2, and an interaction between the genotype rs10993994 and the biomarker MSP (PSP94) also has been observed. Better accuracy in predicting prostate cancer risk may be achievable with a model based on biomarkers, genotype, and genotype-biomarker interaction rather than one based on biomarkers alone.
November 2010
Katrina Goddard,
Ph.D.
Katrina Goddard, Ph.D., Genetic Epidemiologist at the Kaiser Permanente Center for Health Research and Director, Biorepository, Oregon Clinical Translational Research Institute, delivered the November Visiting Scholars Seminar Series presentation titled “Exploring the Uptake and Effectiveness of KRAS Testing for Metastatic Colorectal Cancer: Can We Improve Health Outcome and Influence Decision-Making in Personalized Medicine.”
Dr. Goddard described genomics translation research currently being conducted by the Health Maintenance Organization (HMO) Cancer Research Network (CRN) as part of the CERGEN (Comparative Effectiveness Research in Genomics and Personalized Medicine for Colorectal Cancer) project, which is funded through a Grand Opportunities (GO) Grant from the NCI. The CERGEN project researchers are working with seven large, well-defined populations with well-organized electronic medical record (EMR) data and existing specimen biorepositories, which make it possible to examine a variety of questions about the translation of scientific findings into medical practice. The research setting does pose some limitations, however, because study groups from different timeframes may not be comparable and because genomic information may not always be accessible in EMRs.
A current focus of research is the use of the KRAS test, which distinguishes between two forms of the KRAS gene, the wild type and a mutated form present in 40 percent of patients with metastatic colorectal cancer. The epidermal growth factor (EGFR) inhibitors, a class of drugs used in the treatment of metastatic colorectal cancer, are ineffective in patients with the mutated KRAS gene. The American Society for Clinical Oncology issued a provisional opinion in February 2009 that all patients with metastatic colorectal cancer who are candidates for the use of these drugs should have their tumor tested for KRAS mutations, and the Food and Drug Administration changed the labeling for EGFR inhibitors to refer to KRAS in July 2009. The CERGEN researchers are assessing the extent and timing of changes in the utilization of KRAS testing in metastatic colorectal cancer patients at their HMOs.
Another test being studied is that for Lynch Syndrome, a hereditary syndrome that increases the risk of several types of cancer, especially colorectal cancer. The Centers for Disease Control and Prevention’s (CDC) Evaluation of Genomic Applications in Practice and Prevention (EGAPP) concluded in January 2009 that there was sufficient evidence to recommend genetic testing for Lynch Syndrome in all newly diagnosed colorectal cancer patients so that efforts could be made to reduce cancer incidence and death in their relatives. Dr. Goddard and her colleagues currently are examining the uptake of this test at their HMOs.
In both instances, the CERGEN researchers are examining the patient, provider, and system characteristics that determine which patients receive tests. Future research will expand on the current analyses by evaluating the relationship of test utilization to both patient management and health outcomes.
December 2010
Wei Zheng,
M.D., Ph.D.
Wei Zheng, M.D., Ph.D., Professor and Director, Vanderbilt Epidemiology Center, Vanderbilt University School of Medicine, delivered the December Visiting Scholars Seminar Series presentation titled “International Epidemiologic Research: A Unique Population Laboratory for Studying the Cause of Cancer.”
Dr. Zheng explained that international studies can provide unique opportunities that do not exist in the United States to evaluate important issues related to cancer prevention, diagnosis, and treatment. Studies in China are especially informative because China represents more than 20percent of the world’s population, has unique patterns of cancer occurrence and of lifestyle and environmental exposures, and provides population groups that are readily accessible for research, with very high rates of participation and compliance. The U.S.-China cancer research program was launched in 1979, and the scope of the binational collaboration has been expanding ever since, most recently in molecular and genetic epidemiology.
Dr. Zheng, who is directing the NCI-funded Shanghai Women’s Health Study and a breast cancer case-control study in Shanghai, described findings from the Shanghai studies that illustrate the value of international epidemiologic research. An analysis of data on Shanghai breast cancer survivors showed that both total mortality and breast cancer recurrence were significantly lower in women with higher soy food intakes, thus dispelling concerns about findings in experimental animals that had suggested that the isoflavones in soy might be associated with an increase in cancer risk. An analysis of this type would not have been possible in the United States because soy food intakes in the United States are lower and are correlated with other factors that may influence breast cancer outcomes. Data from Shanghai also have allowed for rapid evaluation of new hypotheses about prostaglandin E2 biomarkers in relation to colorectal cancer risk in a population with low use of non-steroidal anti-inflammatory drugs. High levels of a urinary metabolite of this prostaglandin were strongly associated with colorectal cancer risk. Research is ongoing to evaluate if this biomarker may predicts the occurrence of other types of cancer as well.
Populations from Shanghai and other parts of Asia also have proved to be of great value in genome-wide association studies of various types of cancer. They have enabled the identification of novel variants and loci that would have been difficult to identify in populations of European origin and have allowed fine-scale mapping to identify causal variants. The Shanghai studies also have made important contributions to international consortia that are evaluating hypotheses that cannot be adequately tested in any single study. Dr. Zheng concluded that international consortia are powerful approaches to addressing significant issues related to the etiology and prevention of cancer.
January 2011
Susan Slager,
Ph.D.
Susan Slager, Ph.D., Associate Professor of Biostatistics, Section of Computational Genomics, Mayo Clinic, delivered the January 2011 Visiting Scholars Seminar Series presentation titled “Why Study Familial Chronic Lymphocytic Leukemia (CLL)?”
CLL, an incurable malignancy of B-cells, follows a variable course, is identified by four cell surface markers and B-cell counts greater than 5 x 109/L, and is most often diagnosed without symptoms. No consistent modifiable risk factors are known; however, consistent risk factors include advanced age, Caucasian ancestry, male gender, and family history. Among lymphoproliferative diseases, CLL has one of the highest familial risks–8.5 fold greater in first degree relatives. Familial CLL (F-CLL) is defined as a patient with at least one relative with CLL. Dr. Slager demonstrated no differences in diagnostic and prognostic markers between F-CLL and Sporadic CLL. Thus, clinical misclassification of CLL (sporadic versus familial) is insignificant. However, continued study of F-CLL will shed valuable insight into genetic susceptibility. In addition to studies in twins, evidence of genetic susceptibility is also seen in Asian migration studies–the rate of CLL in U.S.-born Asians is the same as in native born; in contrast, breast cancer rates are increased in U.S.-born Asians compared to native born.
The Genetic Epidemiology CLL (GEC) Consortium is aimed at identifying families with multiple cases of CLL (rare event), evaluating potential pre-malignancy states, mapping and characterizing CLL susceptibility genes, and identifying genetic and environmental differences and similarities in F-CLL and sporadic CLL. The eligibility criteria for study patients included 18 years of age or older and having at least one living relative with CLL. Study participants were obtained from nine recruitment sites nationwide and were asked to donate 4-5 tablespoons of blood, provide cheek cells using a self-collection kit, and complete a detailed exposure questionnaire.
The GEC Consortium used the Affymetrix 6.0 Array to conduct a Genome Wide Association Study (GWAS) and thereby validated 9 of the 10 previously reported GWAS findings and also identified 2 novel loci. The resulting novel findings were within the IRF8 gene and the HLA region. In particular, the HLA region was only detected in the F-CLL cases.
Monoclonal B-cell lymphocytosis (MBL) is an asymptomatic, pre-malignant, pre-CLL condition. Compared to the general population, incidence is more than double in F-CLL families. Dr. Slager assessed the association between CLL GWAS SNPs in the MBL samples and controls. The results indicate a connection between CLL SNPs and risk of MBL; CLL SNPs could potentially be used to predict MBL progression to CLL.
Dr. Slager hypothesizes that the rate of progression from MBL to CLL is much higher in familial MBL samples than in sporadic MBL samples, which will be investigated in the future. Other upcoming research includes combining all three GWAS CLL studies, which will likely lead to the discovery of a large number of low penetrant variants.
February 2011
Jean-Pierre Issa
Pd.D.
Jean-Pierre Issa, Ph.D., Professor in the Department of Leukemia and Co-Director of the Center for Cancer Epigenetics at the University of Texas, M.D. Anderson Cancer Center, delivered the February Visiting Scholars Seminar Series presentation titled “The DNA Methylome Dynamics in Aging and Cancer.”
Dr. Issa described epigenetic CpG DNA methylation, cancer as an epigenetic disease, and the effects of lifestyle and currently used drugs on hypermethylation. Epigenetics can be defined as mitotically stable and irreversible changes in gene expression that manifest differences in phenotype. As a discipline, extensive genetic research has missed the impact of epigenetic variation. He pointed out that not all DNA methylation is epigenetic (e.g., histone modifications that lack stability are not considered epigenetic). Recent data have shown that the most important defect in cancer is epigenetic in nature. In his presentation, Dr. Issa focused on the “deep” silencing (low gene expression) caused by DNA methylation of promoter CpG islands.
Recent advances in experimental technology have dramatically increased the accuracy of epigenetic variation data results. Dr. Issa described digital restriction enzyme analysis of methylation, or DREAM, and its ability to quantitatively measure the degree of methylation at more than 50,000 CpG sites. DREAM studies have revealed hypermethylated CpG islands in cancer tissues and increased methylation in aging cells; about 80 percent of hypermethylation in cancer also can be detected in low levels in aging cells. Thus, cancer cell hypermethylation is actually accelerated aging cell hypermethylation.
In humans, epigenetic variation co-occurs with age, chronic inflammation, gene-specific single nucleotide polymorphisms, and folate consumption. Evidence indicates that an aging individual on a high dose of folate exhibits increased DNA methylation in the colon relative to an aging individual on a low folate dose. It is not understood whether folate directly methylates DNA or indirectly causes hypermethylation (e.g., by increasing inflammation in the colon); however, the dominant factor in epigenetic variation is aging—CpG methylation increases 1 percent every three years in humans. Certain lifestyle habits (e.g., diet and exercise) may potentially reduce the rate of hypermethylation; however, effectively conducting an intervention study on a human subject would require 20 years of study. Hence, these studies can be performed in mice but not in humans.
Epigenetic DNA methylation can only be reversed by drugs. Functional screening studies have revealed that a large number of FDA-approved drugs affect epigenetic modulation. Future studies will address whether these epigenetic changes result from a drug action mechanism or from drug toxicity, which then leads to the follow-on question of whether anti-cancer epigenetic drugs may exert activity on other diseases.
March 2011
Susan Neuhausen
Pd.D.
Susan L. Neuhausen, Ph.D., Morris & Horowitz Families Professor of Cancer Etiology and Outcomes Research in the Department of Population Sciences at the Beckman Research Institute of the City of Hope Cancer Center, CA, delivered the March Visiting Scholars Seminar Series presentation titled “SNPs in IGF Signaling Genes and Breast Cancer Risk.”
Dr. Neuhausen is presently studying genetic variation and breast cancer risk, particularly as it relates to insulin-like growth factor (IGF) signaling. Using genetics to understand the etiology of this disease and develop more accurate models for risk estimation is essential to ultimately tailor prevention strategies and therapies. Based on extensive in vivo and in vitro studies, one important pathway for breast cancer pathogenesis may be the IGF signaling pathway, which regulates both cellular proliferation and apoptosis. Dr. Neuhausen’s research focuses on the association of variants in genes involved in insulin-like growth factor signaling and risk of breast cancer in a cohort of women carrying pathogenic BRCA1 or BRCA2 mutations and in women not carrying known mutations.
BRCA mutation testing is widespread, and a positive result carries a lifetime risk ranging from 40 to 80 percent. Dr. Neuhausen and others have investigated putative risk factors and the majority of modifiers of BRCA1 cancer risk identified to date have been genetic variants in proteins that interact with BRCA1 in DNA repair. She presented their results which suggest that IGF signaling modifies breast and ovarian cancer penetrance in BRCA1 and BRCA2 mutation carriers. They identified significant associations of genetic variants involved in IGF signaling. Given the known interaction of BRCA1 and IGF signaling, and their current results, Dr. Neuhausen suggested that signaling through AKT may modify breast cancer risk in BRCA1 carriers.
Dr. Neuhausen also presented results of her studies of IGF signaling in African-American breast cancer cases, in which they identified significant associations at the IGFBP2_IGFBP5 locus and breast cancer in both African Americans and Nigerians. In her ongoing NCI funded study investigating a more extensive set of IGF signaling genes using the resources of the NCI Breast Cancer Family Registry, they have identified and replicated additional variants associated with risk to develop breast cancer.
Dr. Neuhausen discussed that there are now many loci that have been associated with breast cancer risk, but that all have small effects. She believes that personalized genetic tests for estimating risk are premature given that the known common alleles account for little of the risk and are not causal. Until the underlying causes of the current SNP associations are known, modeling will be imprecise. Upon identifying the actual causal variation, prediction models will require development, testing, and validation. Future studies will test a new paradigm of risk, epigenetics. The identification of epigenetic changes that alter risk to develop breast cancer may be used in the future to improve breast cancer risk prediction models and to develop targeted prevention based on the methylation changes.
April 2011
Habibul Ahsan
M.D.
Habibul Ahsan, M.D., Professor of Epidemiology, Medicine, and Human Genetics, and Director of the Center for Cancer Epidemiology and Prevention (University of Chicago, IL); and Professor of Clinical Epidemiology (Columbia University, NY), delivered the April Visiting Scholars Seminar Series presentation titled “Integrating Genomics and Environment in Epidemiology and Prevention Research.”
Both genomic and epidemiologic data require examination when conducting genomic investigations of human health. Dr. Ahsan presented integrated genomics and environment population research findings using examples from: (a) recent studies from a developing country population cohort, and (b) ongoing breast cancer studies in a consortium of developed countries.
During the past decade, Dr. Ahsan’s team used NIH funding to establish a population-based research laboratory in Bangladesh to examine the environmental impact of drinking water contaminated by naturally occurring arsenic; nearly one in five deaths in the country are attributable to arsenic. Subsequently, the laboratory has become a unique resource in a developing country setting for high-quality, population-based biomedical research; the lab has generated nearly 70 publications.
Individual genomes are known to influence biological processes and disease; genomic RNA and DNA variation is studied to understand risk profiles in humans. Through this genomic research, selenium emerged as a multifactorial intervention to reduce cancer in arsenic-exposed populations, triggering highly specific and novel differential gene expression. Dr. Ahsan’s team is currently conducting genome-wide association studies (GWAS) of body mass index (BMI), plasma cytokines, and several other biological markers.
To conduct a GWAS investigation of breast cancer, Dr. Ahsan’s team is studying several NCI consortium data sets: breast cancer risk in young women (14,000 early-onset and controls from 12 countries), breast cancer risk in all women (60,000 cases and controls from 15 countries), and prognosis of early-onset breast cancer (4,000 early-onset breast cancer patients from 4 countries). Analysis of glucocorticoid receptor-binding gene involvement in breast cancer currently is under way with more than 6,000 cases/controls, 140 SNPs, and 53 genes.
Dr. Ahsan noted the many challenges inherent in genomic research, including gene × gene (G×G) and gene × environment (G×E) interactions, computational capacity, integration of multiple genomic metrics, and statistical methodology. Future genomic variation and human health research endeavors include GWAS based on sequence data, re-use of GWAS databases, post-GWAS functional studies, integration of multiple genomic data sets, risk estimation and modeling, statistical genomics, bioinformatics, computational research, clinical genomics, public health genomics, and environmental/meta genomics.
May 2011
Anne Zeleniuch-
Jacquotte M.D.
Anne Zeleniuch-Jacquotte, M.D., Associate Professor of Environmental Medicine at Langone Medical Center and Principal Investigator of the New York University’s Women’s Health Study (NYUWHS), New York University, NY, delivered the May Visiting Scholars Seminar Series presentation entitled "The NYU Women's Health Study: Results on Risk of Breast and Endometrial Cancers." Dr. Zeleniuch-Jacquotte researches the impact of hormones on women’s health. She presented data on estrogen and androgen levels from a large prospective cohort of more than 14,000 women. Her team assessed various biomarkers in serum samples collected yearly. Participants were either followed up actively by questionnaires every 2–4 years or passively by tumor registries and the National Death Index. Nested case-control studies were performed for biomarker analysis, and all study samples were matched for age, storage time, menopause status, and race.
The NYUWHS found that increasing estrogen levels were associated with an increasing breast cancer risk, which was confirmed by subsequent prospective studies; this finding triggered her interest in the role of estrogen metabolites in breast cancer risk. Her team examined the hypothesis that 2-hydroxyesterone decreases cancer risk because it is rapidly metabolized, whereas 16α-hydroxyesterone increases cancer risk because it is highly mitogenic. However, initial studies have not demonstrated an association between these metabolites or their ratio and breast cancer risk. Analyses on other estrogen metabolites would be of interest.
Androgens produce estrogen upon aromatization in adipose tissue, and this is the main source of estrogens in post-menopausal women. Androgens have both stimulatory and inhibitory effects on cell proliferation; their role in breast cancer development is unclear due to conflicting data from experimental studies. Prospective studies in pre-menopausal women consistently have demonstrated an increasing risk of breast cancer with increasing androgen levels, but most studies to date have been limited to older pre-menopausal women. An earlier pooled analysis in post-menopausal women also described an increasing risk of breast cancer with increasing androgen levels. These results indicate that caution should be exercised when prescribing androgens to peri- and post-menopausal women. Further research is required to determine if androgen analysis could improve breast cancer risk models, screening recommendations, and chemoprevention decision making.
Higher levels of estrogens and androgens have also been associated with an increased risk for endometrial cancer. Similar to breast cancer, no association was observed between the 2:16α-hydroxyestrone ratio and risk of endometrial cancer. Dr. Zeleniuch-Jacquotte and her team are currently conducting analyses on the impact of androgens on endometrial cancer risk in pre-menopausal women. Initial findings suggest that androgens are not associated with an increased risk of endometrial cancer before menopause.
Dr. Zeleniuch-Jacquotte concluded that maintenance of a healthy weight, particularly at menopause, is one of the few modifiable factors to prevent breast and endometrial cancers. In future work, she is interested in learning the usefulness of sex hormones in risk prediction models.
