In 2005, 3 multidisciplinary teams each received five-year grants of $2.25 million to address challenging questions in human disease.
2005 CIAP Grantees
Development of the First Test for Common Cancer Risk in the General Population
Andrew P. Feinberg, M.D., M.P.H.,
Johns Hopkins University School of Medicine
Francis M. Giardiello, M.D., M.B.A., Johns Hopkins University School of Medicine
Elizabeth A. Platz, Sc.D., M.P.H., Johns Hopkins Bloomberg School of Public Health
Marcia R. Cruz-Correa, M.D., Ph.D., Cleveland Clinic Foundation & Johns Hopkins University School of Medicine
Christine Iacobuzio-Donahue, M.D., Ph.D., Johns Hopkins University School of Medicine
Holly Taylor, Ph.D., Johns Hopkins Bloomberg School of Public Health
Benjamin Wilfond, M.D., National Human Genome Research Institute
Hengmi Cui, Ph.D., Johns Hopkins University School of Medicine
Molecular Genetics, Gastroenterology, Epidemiology, Community Medicine, Pathology, Health Policy, Bioethics
Although many conventional genetic mutations have been observed in human cancer, currently there are no such molecular abnormalities that occur at high frequency in normal cells in the general population. Thus, despite the inherent promise of molecular biology, no screening tests have been developed that identify patients in the general population at risk of cancer, only small subgroups of known hereditary predisposition, or screening tests that are positive only after the patient develops malignancy. This disappointing truth exists despite the fact that genetic background is thought to contribute significantly to cancer risk. Identifying genetic contributing factors and testing for them would offer a major benefit to prevention of cancer mortality, similar to the role played by lipid screening for cardiovascular disease risk. We have recently discovered an epigenetic marker in normal cells, loss of imprinting (LOI) of the insulin-like growth factor-II gene (IGF2), present commonly in the population, that is associated with personal and family history of colorectal cancer (CRC). We obtained personal and family cancer history, blood samples, and colon biopsy specimens. Imprinting and methylation analysis were performed without knowledge of clinical history. The odds ratio for LOI given a positive family history of CRC was 5.15, and for LOI given a past or present history of CRC in the patient was 21.7. Thus, LOI may serve as the first common molecular marker for cancer risk, which may be assayed with a DNA-based blood test. This work was reported in Science, and another group has also confirmed our result. While these results are extremely exciting and highly statistically significant, the assay is by definition epigenetic, which raises several hurdles that must be overcome before making this test a practical reality. First, we must extend our initial study population in number and to racially and ethnically diverse populations. In addition, since this epigenetic alteration may not be present throughout one’s lifetime, as are conventional DNA mutations, one must design a protocol for testing patients at one time and then determining the outcome at another.
In the past 18 months, we have performed work under a CIAP planning grant with the following accomplishments: (1) we have demonstrated the feasibility of the molecular biology assays to be performed on community-based samples at Johns Hopkins Hospital (JHH) and the Cleveland Clinic Foundation (CCF) in Ft. Lauderdale. Our data examining non-minority patients in that setting support our original conclusions, allowing us to extend this investigation to patients of differing ethnic (Hispanic at CCF) and racial (African American at JHH and CCF) backgrounds. (2) We have demonstrated that samples that were banked over many years are suitable for molecular analysis of IGF2 methylation, allowing us to collaborate with large existing cohorts with archival material. (3) We have developed a highly integrated interdisciplinary approach to developing an epigenetic test for common cancer risk, involving monthly meetings and frequent discussion among a team of molecular biologists, gastroenterologists, pathologists, community physicians, epidemiologists, and bioethicists from Johns Hopkins Schools of Medicine and Public Health, the community-based Cleveland Clinic of Florida, and the Harvard School of Public Health. All of the team members are established experts in their fields. In addition, in work not supported by DDCF, we have created a mouse model that proves that LOI increases intestinal cancer risk, and discovered a differentiation defect in normal colonic epithelia in affected mice, that is also present in normal tissue of patients with LOI. This result, just reported in Science supports the biological basis of our approach, but also suggests a second approach to risk assessment, by immunohistochemical evaluation of normal colonic epithelia.
The Specific Aims of our full grant are: (1) to develop and validate a practical test for colorectal cancer risk, suitable for large-scale application to patient populations, based on loss of imprinting of IGF2; (2) to determine whether LOI is stable over time within an individual, and whether LOI is present in family members of patients with LOI; (3) to evaluate the predictive value of IGF2 LOI for future risk of colorectal cancer and adenomatous polyps in a prospective cohort without a prior diagnosis of colorectal neoplasia; and (4) to assess the uptake of LOI testing in a primary care setting and address three important ethical questions regarding informed consent, decision-making about LOI testing, and the receipt of LOI results. This study embodies a high degree of novelty, including the novelty of a common cancer risk test itself, its highly interdisciplinary nature, its strong bioethics component, and its emphasis on community-based populations. These attributes should maximize the potential of this work to have a major impact on cancer mortality in the general population.
Clinical Application of Molecular Imaging to Oncology
Michael V. Seiden, M.D., Ph.D.,
Massachusetts General Hospital
Arlan Fuller, M.D., Massachusetts General Hospital
Jeffrey G. Supko, Ph.D., Massachusetts General Hospital
Debra Bell, M.D., Massachusetts General Hospital
Mukesh G. Harisinghani, M.D., Massachusetts General Hospital
Medical Oncology, Gynecologic Oncology, Pathology, Pharmacology, Radiology
Molecular imaging in oncology offers the opportunity to not only improve the detection of disease but also to provide unique molecular characterization of biologic processes within the malignancy or the surrounding supportive tissue. This information, obtained in real time and through non-destructive technology, offers the ability to better understand the biology of cancer and to more intelligently select and monitor the effects of therapeutics on a specific tumor. Ovarian cancer and pancreatic are both highly lethal and common malignancies that both lack effective screening techniques. Both malignancies tend to spread early in their natural history to the peritoneal cavity and such spread greatly impacts prognosis and in the case of pancreatic cancer, surgical management. Current imaging techniques such as magnetic resonance imaging (MRI), computed axial tomography (CAT) scan, and positron emission tomography (PET) scan are unable to discern sub-centimeter tumors due partly to tumor size. Effective imaging of small tumors using these technologies is also hampered by the mobility of intraperitoneal surfaces such as loops of intestine that impedes optimal signal acquisition. Laparoscopy or laparotomy is the current standard approach for evaluating the peritoneal cavity in ovarian and pancreatic cancer patients although both techniques have limited capability of detecting sub-millimeter disease. The clinical management of both malignancies would be vastly improved by the development of imaging technologies that allow detection and characterization of very small volume intraperitoneal tumor. This proposal to the Doris Duke Foundation describes the development, testing, and use of both novel imaging equipment and novel imaging agents to better evaluate the peritoneal cavity in individuals with either ovarian cancer or clinically localized pancreatic cancer. Specifically this will include the construction of laparoscopic imaging equipment and associated software that will allow simultaneous inspection of the peritoneal cavity in both white light and in the near infrared spectrum. The inspection in the near infrared will allow the detection of near infrared imaging agents designed to specifically identify small tumors and to provide biologic characterization of the tumor. The testing the imaging system and the imaging agents will occur through a series of clinical trials enrolling women with ovarian cancer and individuals with 'localized' pancreatic cancer at the Massachusetts General Hospital. Execution will involve a team of medical oncologists, gynecologic oncologists, and general surgeons along with a pharmacologist, pathologist, and molecular biologist. Incorporated into these trials will be pharmacology of novel near infrared imaging agents that are designed to specifically detect tumor-associated proteases. In addition, the program will include a thorough pathologic evaluation of tumors identified by the various imaging agents to correlate optical signal with biologic characteristics of the tumor. Three industrial collaborators will aid the project with expertise in laparoscope design and construction, optical engineering, and probe construction. Although the specific clinical scenarios in which this technology will be applied vary, the technology is positioned to more intelligently determine the best surgical and systemic approaches to the treatment of both ovarian and pancreatic cancer. In addition, use of molecularly targeted near infrared probes potentially allows evaluation of therapeutic efficacy of various anti-angiogenic agents in real time in situ. It is likely that this new imaging technology will lead to changes in the clinical management of ovarian and pancreatic cancer by the conclusion of the project in 2010.
A Mitochondrial Basis for Metabolic Syndrome
Douglas C. Wallace, Ph.D., University of California, Irvine, Colleges of Medicine and Biological Sciences
J. Jay Gargus, M.D., Ph.D., University of California, Irvine
F. Sherwood Rowland, Ph.D., University of California, Irvine
Donald R. Blake, Ph.D., University of California, Irvine
Bruce J. Tromberg, Ph.D., Beckman Laser Institute, University of California, Irvine
Biological Chemistry, Ecology & Evolutional Biology, Physiology & Biophysics, Chemistry, Biomedical Engineering
The "Metabolic Syndrome" encompasses diabetes and hyperglycemia, obesity and hypertrigiyceridemia, and hypertension. Considerable biochemical and genetic evidence has now accumulated implicating mitochondrial defects in the etiology of these symptoms. We now propose a series of biochemical, physiological and genetic studies to test this mitochondrial hypothesis for the Metabolic Syndrome and to develop more effective diagnostic tools for this class of diseases.
The Metabolic Syndrome patients to be studied include individuals of Chinese descent living in the southwestern United States (US) seen at the University of California, Irvine (UCI) Health Science Center and in Taipei, Taiwan seen at the National Taiwan University Hospital. Metabolic Syndrome patients and matched controls for the southwestern US will be subjected to detailed physiological and biochemical analysis. First, these subjects' mitochondrial function will be evaluated using our advanced diagnostic procedures including pulmonary exercise stress test, 31P MR spectroscopy, and muscle biopsy mitochondria respiration and enzymological studies. These data should delineate any mitochondrial dysfunction in these patients and will permit the evaluation of two new non-invasive approaches for monitoring mitochondrial dysfunction: micro-organic breathe analysis and trans-muscular Near Infrared (NIR) spectroscopy. The micro-organic breathe test evaluates exhaled organic molecules using an ultra-sensitive chemical detection system capable of detecting 20 parts per quadrillion. This system has revealed a wide variety of new organic chemicals in human breathe which appear to track with the individual's physiological state. For example, exhaled ethanol permits monitoring of serum glucose levels during glucose tolerance tests. The trans-muscular NIR spectroscopy system employs an advanced multiple laser diode array to interrogate the absorption wavelengths for oxygenation and deoxygenated hemoglobin and oxidized versus reduced cytochonre c oxidase. When combined with an exercise regime, this new technology promises to permit the rapid non-invasive assessment of mitochondrial dysfunction.
These new physiological and biochemical tools will be coupled with a series of new molecular genetic approaches to search for clinically relevant mitochondrial DNA (mtDNA) sequence variants in the Metabolic Syndrome. Two classes of genetic studies will be conducted: family studies involving multiple large Taipei, Taiwan pedigrees and case control studies involving a cohort of 500 patients and controls from Taipei. The pedigrees will first be screened for mtDNA mutations known to result in Metabolic Syndrome. Those pedigrees that are negative will be further evaluated for heteroplasmic mtDNA mutations by Surveyor Nuclease digestion of heteroduplexes and for homoplasmic mtDNA mutations by complete mtDNA sequencing coupled with haplogorup analysis. Putative mtDNA mutations will be confirmed by linking specific mitochondrial defects detected by MITOCHIP expression profiles and/or biochemical aberrations to mtDNA variants via cybrid transfer experiments. The role of ancient mtDNA adaptive polymorphisms in Metabolic Syndrome will also be tested using the 500 subject Taiwan Metabolic Syndrome cohort. A series of mtDNA haplogroup-specific single nucleotide polymorphisms (SNPs) will be examined for all 500 subjects. These will then be correlated with their Metabolic Syndrome phenotypes. Preliminary studies already suggest that certain mtDNA haplogorups might be associated with hypertension.
From our preliminary results we anticipate that we will be able to confirm that the Metabolic Syndrome is a mitochondrial disease, identify a number of mtDNA mutations associated with these diseases, and perfect some effective new approaches to assist with the diagnosis of these common problems. More importantly, these studies should suggest new avenues for the treatment of these epidemic diseases.