| CIHR Institutes |
| III Home |
| About III |
| III Funding |
| III Partnerships |
| III Showcase |
| III Publications & Resources |
| III Calendar of Events |
| Contact III |
Dr. Katherine A. Siminovitch
Professor, Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto
Dr. Eleanor Fish
Professor, Department of Immunology, University of Toronto
CIHR Institute of Infection and Immunity
Dr. Bhagirath Singh
Scientific Director
CIHR Institutes of
Aging, Gender and Health, Nutrition, Metabolism and Diabetes
Part I: Presentations
I: Autoimmune Diseases: Basic Mechanisms and Commonalities
II: Immunological Principles
III: Science and Technology Platform
IV: Autoimmune Diseases: Basic Mechanisms and Commonalities
Part II: Consultation Report
Supports, Opportunities, Challenges and Consideration
Strategic Research Directions
Methods for Early Case Findings
Biometrics
Cohort Methodologies
Tissue Regeneration and Repair
Biomarkers
Immunopathogenesis
Functional Genomics
Microbial Autoimmune Pathogenesis
Conclusion
Appendix I - Organizing Committee
Appendix II - Symposium Agenda
Appendix III - Abstracts and Biographies of Speakers
Appendix IV - Lay Introduction to the Immune System
Appendix V - Background Document
Appendix VI - Determinants of Health
Appendix VII - Symposium Participants
The research symposium Integrating Discovery Platforms in Autoimmune Diseases aimed to develop a framework for a Canadian health research agenda in autoimmune diseases by targeting the following objectives: to explore the current situation in autoimmune diseases, and in particular basic mechanisms leading to, and commonalities among, these diseases; to identify strategic directions and potential research questions to be used, for example, as the basis for CIHR and partner-sponsored Requests for Applications and for integration of autoimmune diseases into the Canadian Lifelong Health Initiative; and to enhance collaboration and partnerships among stakeholders in the autoimmune diseases community.
The presentations informed participants on the current situation in autoimmune research and set a framework for discussions related to strategic planning.
Session I: Autoimmune Diseases: Basic Mechanisms and Commonalities
Dr. Paul Fortin and Dr. Luanne Metz co-chaired Session I. Dr. Amit Bar-Or of the Montreal Neurological Institute presented Multiple Sclerosis: The Neuroimmune Interface. He discussed emerging themes in multiple sclerosis, a chronic inflammatory disease of the central nervous system that specifically targets the brain and spinal cord. Dr. Charles Elson, University of Alabama at Birmingham, spoke about The Inflammatory Bowel Diseases: Disorder of the Host (self) - Microbial (non-self) Interface, noting that the gastrointestinal tract has a number of immune-mediated inflammatory diseases such as autoimmune gastritis and celiac disease for which the target antigens have been defined. Dr. Jayne Danska from the University of Toronto discussed Type 1 Diabetes: Immunogenetic Mechanisms and Prospects for Deconstructing Complex Disease, stressing the need to identify more human markers (both cellular and genetic) of pre-diabetic autoimmunity. Autoimmune Rheumatic Diseases were discussed by Dr. Peter Lipsky, National Institute of Arthritis and Musculoskeletal and Skin Diseases. He described what he called the "Chaos Model of Autoimmunity." This model could be applicable to the pathogenesis of several autoimmune diseases and serve as a common basis for discussion. Dr. Hani El-Gabalawy chaired the final presentation in this session, in which Dr. John McLaughlin from the Samuel Lunenfeld Research Institute and the University of Toronto presented Plans for Cohort Studies of the Canadian Lifelong Health Initiative. He described a cross- cutting, strategic, multi-institute initiative of the CIHR, the Canadian Lifelong Health Initiative, which will include two longitudinal studies-the Canadian National Birth Cohort and the Canadian Longitudinal Study on Aging-to investigate the hypothesis that disease burden is jointly determined by individual genetic endowment and complex environmental factors.
Session II: Immunological Principles
Co-Chairs Dr. Karen Madsen and Dr. Ken Croitoru directed Session II. Dr. Stephen D. Miller, Northwestern University Medical School, presented Immunological Principles Underlying the Pathogenesis and Immunoregulation of T Cell-Mediated Autoimmune Disease, discussing insights from his group's study of animal models of multiple sclerosis, the immunological principles underlying disease pathogenesis, and recent data on disease intervention using antigen-directed immunotherapies. Dr. Andrew Macpherson, University of Zürich, explained the Comparmentalisation of Immune Responses to Commensal Intestinal Bacteria, noting that non-pathogenic environmental organisms shape the immune system. Mr. Steve Kerfoot spoke on behalf of Dr. Paul Kubes from the University of Calgary. He discussed Trafficking of Leukocytes in the Brain: Learning by Watching Leukocyte Behaviour. He explained techniques used in Dr. Kubes's laboratory to image inflammation and leukocyte recruitment in vivo. Mr. Kerfoot concluded that in vivo imaging is a powerful tool to understand leukocyte recruitment by permitting direct observation of the process.
Session III: Science and Technology Platforms
Dr. Steve Collins and Dr. Pere Santamaria co-chaired Session III. The session opened with a presentation by Dr. Claire Bombardier (in association with Dr. Sheilah Hogg-Johnson), of the Toronto General Research Institute, who used the example of rheumatoid arthritis to illustrate The Challenge of Assembling and Maintaining Clinical Cohorts and Predicting Outcomes. She discussed classic predictors of disease outcome; disease activity at presentation, spread in terms of structural damage; functional ability; and variables such as socio-economic status apart from biological factors that have impact on disease outcome. Dr. John A. Wilkins from the Manitoba Centre for Proteomics presented three specific examples of Proteomic Approaches to the Study of Human Disease and discussed issues related to patient selection and sample acquisition, remarking that our proteomic capabilities are rapidly increasing. Dr. Alexandre Montpetit, of McGill University and the Genome Quebec Innovation Centre, spoke about Emerging Genomic Tools to Study Autoimmune and Other Complex Diseases. He noted that the phenotype associated with a given disease and the underlying genetic defect can be studied by linkage analysis or association; however, due to the large size of the human genome, linkage analysis is more appropriate for mapping on a genomic scale. Dr. Igor Jurisica, University of Toronto and Queen's University, presented Towards an Integrated and Intelligent Molecular Medicine, discussing the computational aspects, challenges and possibilities that high throughput data obtained from microarray and protein array analyses afford in the characterization of complex diseases.
Session IV: US National Institute of Allergy and Infectious Diseases
Dr. Jack P. Antel chaired the final session in which Dr. Daniel Rotrosen, of the National Institutes of Health, USA, presented an overview of NIAID/NIH Funding and Strategic Planning as Related to Autoimmune Disease. In the 2003 fiscal year the NIH awarded an estimated $591 million dollars for autoimmunity research, which represents about 2.2% of the $27 billion dollar NIH budget. Approximately 45% of the autoimmunity expenditures are in projects examining pathogenesis and immune dysfunction, while 3% of the funds are allocated to the development of new animal models. The ultimate goal of the NIH's clinical research programs is to bring new vaccines, immune-based therapies and diagnostics to clinical practice.
Following the presentations, participants worked in small, mixed groups to discuss the following two issues with the goal of selecting strategic research directions. Some groups noted overlap between these two areas. The following is a summary of the discussions.
a) supports (e.g., infrastructure, capacity building) and opportunities for a Canadian health research agenda with a focus on integrated discovery platforms in autoimmune diseases
b) key challenges and considerations when developing a framework for a Canadian health research agenda with a focus on integrated discovery platforms in autoimmune diseases over the next ten years
Conference Recommendations
| New Research Questions |
Supports Required | |
| Methods for Early Case Findings | Benefits and drawbacks of doing direct-to-consumer advertising to generate cases, including analysis of retrieval numbers and false positives Value of billing databases in identifying early diagnosis Benefits of networks with general practitioners to identify early cases, e.g., through continuing medical education |
|
| Biometrics | Development of a clean, standardized, agreed-upon set of common variables for the determinants of health Methods for cohort research, e.g., statistical analysis for innovative data mining of biometrics, patterns of care, patient outcomes Innovative study designs, e.g., crossover designs with strategies for selecting control groups Common, across-disease, early case definitions and innovative statistical methods for grouping clusters of early disease classifications |
|
| Cohort Methodologies | Innovative data collection methods that facilitate participation and retention, e.g., data collection that is integrated into practice, with value added for participating clinicians and applications to community practice Research into methods for addressing practical issues related to cohort retention and follow-up, e.g., what to do when patients change physicians or move to another province, incentives for patients to remain involved in data collection processes Issues related to methods of targeted sampling, e.g., to achieve community representation or to support standard tissue collection across various sites Investigator-initiated research questions, e.g., on courses and prognosis of patients |
Supports are required to ensure that the research questions in this and the two following sections are addressed. A top priority in terms of capacity development involves the creation of a funded standing group or superstructure that enables integrated approaches among autoimmune disease researchers. Clinical researchers in autoimmune disease face many similar problems across Canada, both within individual diseases and across diseases. The purpose of this standing group would be to identify, clarify and address issues such as information technology, bar codes, privacy, data security, standards for tissue collection and handling, innovative methods for data collection and access to billing data. |
| Tissue Regeneration and Repair | Mobilization of progenitor cells to enhance repair Manipulation of pre-existent cells to promote down regulation of cytopathic receptors, or upregulation of growth receptors Molecular response during damage, recovery, repair and remodeling The process of tissue response to damage (fibrosis, gliosis), when it ceases to be beneficial and becomes detrimental to repair |
Effective collaboration and cohesion among the agendas of hospitals, research institutes, universities, as well as public and private funding agencies Coordinated access to relevant tissues to optimize research Rational integration of the multiple federal programs in existence to facilitate and optimize hiring practices and initiation of research |
| Biomarkers | Biomarker identification and validation (risk, activity, progression, response to drugs, and disease) Development and implementation of biomarkers and bioassays in well- designed clinical trials Hyper-accelerated progression of biomarkers in clinical trials Tissue- and species-specific biomarkers Population-based studies involving the building of new, early cohorts to identify/develop biomarkers Interface with chemical genomics - array data (gene) and screen library Imaging biomarkers including molecular biomarkers for in vivo imaging of target organs Development of methodology for dealing with large data sets that are unique to Autoimmune Diseases (AID) |
There is a need for infrastructure and core facilities and the development of national consortia to enable integrated approaches to AID that can minimize duplication among the agencies involved. Integrated technology platforms in core facilities are required to develop partnerships and collaborations for accessing, developing and standardizing specialized assays. Hyper-accelerated programs are needed to support the development and implementation of bioassays into clinical trials of AID. Academic and clinical consortia/teams are required to build on Canadian traditions of collaborative projects, e.g., to - enable support for clinician buy-in - facilitate standardization of bioassays - help with the mechanics of getting patient samples, i.e., blood draws, information technology, ethics |
| Immunopathogenesis | Exploring regulatory pathways for treatment from the genetics to expression and function Immunotherapies that alter and cure disease Correlate lessons learned from oncology |
Capacity building through training and recruiting Exploration of linkages to established immune tolerance networks |
| Functional Genomics | How to validate the functional impact of genetic polymorphisms identified in human or animal models How to harness outputs of large-scale genomic/proteomic data sets to translate to molecular phenotype and pathophysiology of disease Genetic regulation of pre-clinical phenotypes (based on biomarkers) |
Organizational infrastructure for a data coordinating centre, including financial administration, communication between research centres, information and ready access to core facilities for genomics, proteomics, imaging Sustained funding adequate to allow new teams to organize, establish platforms, produce data and take the risks required for innovation |
| Microbial Autoimmune Pathogenesis | Construction of new animal models of autoimmune disease (including non-GI) using proscribed infections with well-defined microbial constituents Looking at responses to infection and products of replicated fetal/ immediate newborn in a gnotobiotic controlled environment Determinants of immunoreactivity throughout life Translation to human neonatal physiology/biology imprinting |
Gnotobiotic facilities for experimental animals Data bases, e.g., mining and construction of appropriate questions Cryopreservation of animal models for interprovincial transfer Free availability of new animal models Viral/microbial bank |
Drs. Siminovitch, Finegood and Singh each addressed the participants with closing remarks, emphasizing the importance of the meeting as a foundation for developing new research initiatives on integrated discovery platforms. Dr. Singh confirmed the cross-cutting nature of autoimmune diseases and the need to have Voluntary Health Organizations (VHO) involved in the development and implementation of research frameworks. He also acknowledged the benefits of having VHO representatives at the symposium and referred to their remarks on the second afternoon, when VHO participants emphasized the importance of inclusive, collaborative approaches to research that would result in clear health outcomes for both patients and caregivers. He further noted the presence of researchers across the four CIHR themes and the importance of following through on new relationships developed at this session. He will be sharing the results of the workshop with all parties in attendance, and will also be holding further discussions with the NIH to follow through on suggestions made regarding possible long-term infrastructure partnerships.
The purpose of the research symposium "Integrating Discovery Platforms in Autoimmune Diseases" was to develop a framework for a Canadian health research agenda in autoimmune diseases. The following objectives were targeted:
The following presentations informed participants on the current situation in autoimmune research and set a framework for discussions related to strategic planning.
| Topic | Presenters |
| Multiple Sclerosis: The Neuroimmune Interface | Dr. Amit Bar-Or |
| The Inflammatory Bowel Diseases: Disorders of the Host (self) - Microbial (non-self) Interface | Dr. Charles Elson |
| Type 1 Diabetes: Immunogenetic Mechanisms and Prospects for Deconstructing Complex Disease | Dr. Jayne Danska |
| Immunopathogenesis of Rheumatoid Arthritis and Lupus | Dr. Peter Lipsky |
Chair: Dr. Hani El-Gabalawy
| Topic | Presenters |
| Plans for the Cohort Studies of the Canadian Lifelong Health Initiative | Dr. John McLaughlin |
Multiple Sclerosis: The Neuroimmune Interface
Dr. Amit Bar-Or, Montreal Neurological Institute
Dr. Amit Bar-Or discussed emerging themes in multiple sclerosis (MS). MS is regarded as a chronic inflammatory disease of the central nervous system (CNS) that specifically targets the brain and spinal cord. Damage to the CNS is associated with injury to both myelin and myelin-producing cells (oligodendrocytes) as well as axonal loss. MS is the major cause of neurological disability in young adults, with diagnosis in the 20's or 30's being quite common. Most patients stop working within 10 to15 yrs of diagnosis, which causes a great burden in terms of their life, their families and society in general. Interestingly, the prevalence of disease is non-randomly distributed with the incidence increasing as one moves either north or south of the equator. Ethnicity is also an important consideration, with Caucasians suffering from MS most commonly. Moreover, within the Caucasian population, the incidence is approximately 1.5 to 2.5 times higher in females.
One of the clear features of MS is its highly variable and unpredictable course. Magnetic resonance imaging (MRI) has aided diagnosis and has allowed clinicians and researchers to monitor some aspects of the disease. The incorporation of gadolineum (Gd) enhancement into MRI scans has provided a window into understanding disease activity, whereby seepage of Gd reveals areas where the blood brain barrier (BBB) is compromised. Gd enhancement has also been used as a surrogate marker to study clinical trial outcomes in phase II trials. Dr. Bar-Or noted that recent studies in pathology and particularly imaging have suggested a possible degenerative component of this disease (tissue compromise with little immune cell infiltration) which may start early in the illness and might occur relatively independently of inflammation. These findings have resulted in a paradigm shift in treatment such that patients now begin treatment much earlier. Ideally, such treatments will target neuroprotection/repair in addition to immune modulation. Another method for studying the disease course employs magnetic resonance spectroscopy which allows the assessment of metabolites in the brain. In the white matter of a normal brain one would expect to see a characteristic signature of the chemical composition which is abnormal in the MS brain, indicating axonal compromise. To add to the complexity of MS, it has recently been proposed that there are four distinct patterns of tissue injury associated with demyelination. This implies that the disease could present as one of four different pathophysiological entities or perhaps the patterns represent snapshots of predominant processes occurring in patients that change over time. This is a very important point to delineate.
Dr. Bar-Or remarked that MS fits the model of the delicate balance between genes, pathogens and failed immune regulation. For example, three genome screens have revealed that multiple genes contribute to the risk of developing the disease, and with respect to the environment, researchers have considered a role for toxins, nutrients and infectious agents such as viruses. MS is considered to be a disease in which there is peripheral activation of CNS reactive cells, which are associated with waves of inflammation in the brain. The most commonly used animal model of MS is the experimental autoimmune encephalomyelitis (EAE) model. This model has shown that CD4+ T cells that react to the immunizing CNS antigen can transfer disease. With respect to the Th1/Th2 paradigm it appears that Th1 cells can induce disease while Th2 cells with the same specificity may be protective in EAE. The T cell autoimmune model of MS can be broken down into the following steps: 1) peripheral immune cell activation; 2) upregulation of adhesion molecules; 3) attraction through chemokine/chemokine receptor interaction; 4) active invasion of immune cells into the brain by elaboration of lytic enzymes; and 5) reactivation of T cells by local or invading antigen presenting cells (APCs) that then participate in the injury process. Molecules that can potentially act as therapeutic targets at each of these steps have been identified in MS patients; however, to date no single biomarker of the disease exists. Dr. Bar-Or concluded by highlighting a theme common to all autoimmune diseases which involves the need to develop and integrate bioassays into clinical trials. Such assays would provide insight into the therapeutic mode of action as well as open a window into the disease pathophysiology. Lastly, Dr. Bar-Or noted that clinical trials have revealed that some treatments should be used with caution. For example, anti-TNF-a therapies have been used to successfully treat several autoimmune diseases. However, in some patients there is an emergence of demyelinating disease and in early trials of these agents in patients with MS there have been reports of disease exacerbation.
The Inflammatory Bowel Diseases: Disorder of the Host (self) - Microbial (non-self) Interface
Dr. Charles Elson, University of Alabama at Birmingham
Dr. Elson noted that the gastrointestinal (GI) tract has a number of immune-mediated inflammatory diseases such as autoimmune gastritis and celiac disease for which the target antigens have been defined. Ulcerative colitis and Crohn's disease are chronic inflammatory diseases that affect the colon or most of the intestine respectively. There is some evidence for autoreactivity in these diseases, as autoantibodies against self-antigens have been identified in ulcerative colitis, such as pANCA, while antibodies against bacterial antigens, such as ASCA, have been characterized in Crohn's disease. Approximately 10 years ago, mouse models of inflammatory bowel disease (IBD) began to emerge, which emphasized two common themes: CD4+ T cells are the effector cells that mediate disease, and bacterial flora drives the CD4+ T cell response such that when the animals are rendered germ-free they no longer develop the disease. Dr. Elson discussed a current hypothesis for IBD pathogenesis, which suggests that disease results from an abnormal mucosal immune (CD4+ T cell) response to enteric bacterial antigens in a genetically susceptible host. Interestingly, the dichotomy between Th1 and Th2 cells is not apparent in IBD as either subset can induce disease. Recent studies have shown that regulatory T cells may play a protective role suggesting this subset could be harnessed as a therapeutic to treat autoimmune disease. In general, if one refers back to the mouse models, disease can be said to result from either impaired T cell regulation or excessive T cell effector function.
A second hypothesis, that Dr. Elson termed the epithelial hypothesis, suggests that abnormal function of the epithelial layer could result in chronic intestinal inflammation even in the presence of a normal immune system. This concept is supported by findings such as the abnormal barrier function for small molecules in Crohn's disease. In fact, it is now recognized that there are dynamic interactions between enteric bacteria, the epithelium and lymphocytes in the gut, although these interactions have yet to be clearly defined. However, it has been shown that enteric bacteria can induce altered gene expression in the epithelium. Epithelial cells then transduce signals not only to lymphocytes but also back to the bacteria themselves. Interestingly, CARD15/NOD2 was the first susceptibility gene identified in Crohn's disease. CARD15/NOD2 is a pattern recognition receptor (PRR) that binds bacterial peptidoglycan thereby activating NF-?B, which upregulates downstream genes, such as TNF-a. PRR and Toll-like receptors (TLR), recognize pathogen associated molecular patterns (PAMP) expressed by microorganisms. It is important to note that even commensal bacteria express PAMPs, for example flagellin that binds TLR5. Dr. Elson's group identified a locus on chromosome 3 that regulates the response to bacterial antigens in IL-10 deficient mice. This suggests that the susceptibility to colitis may be governed by how the immune system responds to commensal bacteria.
Dr. Elson made the interesting point that the immune system evolved in filth. Over time, microbes have acquired host genes and in order to mount a defense the host must be able to respond to self, which is why we have autoimmunity. This is supported by the observation that in areas where infections are endemic there is decreased incidence of autoimmunity. In his concluding remarks Dr. Elson noted that there are tremendous opportunities to develop new technologies to dissect microbial-epithelial-lymphocyte interactions to gain improved understanding of IBD.
Type 1 Diabetes: Immunogenetic Mechanisms and Prospects for Deconstructing Complex Disease
Dr. Jayne Danska, University of Toronto
Dr. Danska discussed type 1 diabetes (T1D), a T cell-mediated autoimmune disease. An incredible benefit to the study of T1D is the availability of spontaneous inbred animal models that recapitulate several aspects of human disease, namely the NOD mouse and BB rat. These models have been critical in understanding T1D pathogenesis and have permitted genetic and immunologic studies. T1D is probably the best studied autoimmune disease at the genetic level. It is clearly multigenic, with polymorphism at the major histocompatibility (MHC) locus playing an important role both in humans and animal models. Researchers have also dissected several cellular mechanisms underlying T1D susceptibility and progression including dysregulated lymphocyte homeostasis, requirement for immunoregulatory T cells, and defective macrophage and dendritic cell (DC) differentiation and function.
Dr. Danska also mentioned that the animal models have permitted the disease to be examined as a progression of steps. The disease course is predictable in NOD mice, taking place in a time frame of months. It is well established that a number of critical events must take place before ß cell death is accomplished, which actually occurs late in the disease process. This suggests that identifying the molecular basis of these steps will provide multiple opportunities for therapeutic intervention. Specifically, Dr. Danska's group has studied the recruitment of T cells to the early lesions. Using a series of genetic and genomic approaches, they identified several regions of the murine genome that control this single step, which underscores the complex genetics of this disease. Through recombinant congenic and genomic strategies it is possible to narrow these regions, but there is still a large number of genes to be vetted as individual candidates. Ultimately, this would require platforms to evaluate the functionality of the genetic variants. Using these types of approaches it should be possible to dissect the function of specific regions of the genome that control specific steps in T1D. Moreover, a collaboration funded by Genome Canada was established with the goal of identifying shared pathways in T1D pathogenesis between NOD mice, BB rats and T1D families using gene expression microarrays.
In terms of moving forward, Dr. Danska mentioned the need to identify more human markers (both cellular and genetic) of pre-diabetic autoimmunity. With respect to Canada's position, Dr. Danska emphasized the need to build on existing infrastructure to support national and international collaboration in discovery research and clinical trials. Dr. Danska concluded by encouraging collaborations that merge disciplines and platforms to examine: 1) rodent modeling and population based human studies; 2) genomic analysis of heritable susceptibility with epidemiology of potential environmental triggers; 3) molecular biology of host-pathogen interactions and response to autoantigens; and 4) sex bias and developmental determinants of autoimmune disease.
Immunopathogenesis of Rheumatoid Arthritis
Dr. Peter Lipsky, National Institute of Arthritis and Musculoskeletal and Skin Diseases
Dr. Lipsky began his talk with the remark that "autoimmunity is everywhere" and noted that in a Newsweek article of the top 10 most important medical stories of the past year, autoimmunity was listed as number 6. Dr. Lipsky described a model which he felt would be applicable to the pathogenesis of several autoimmune diseases and serve as a common basis for discussion. This model could conceivably be called the "Chaos Model of Autoimmunity" whereby in complex diseases, genetic polymorphisms create a subtle but profound difference in the host. These changes must be subtle as disease can take years to develop, which is in stark contrast to animal models where a single gene can be manipulated to reproduce disease in a defined time frame. In humans, the entire array of polymorphisms that are characteristic of certain diseases creates a responsive unit in each individual that deals with the environment in a unique manner. In this model it can be envisioned that non-specific inflammation could lead to tissue alterations caused by 1) altered homing of inflammatory cells; 2) stromal cell maturation such that the cells function in a way that fosters immune reactivity (i.e., they function like follicular DCs); 3) neoantigen expression in the inflammatory sites; and 4) DC maturation (i.e., protolerogenic DCs become proinflammatory DCs). Ultimately, this leads to activation of autoreactive T cells and formation of germinal centre-like structures leading to autoantibody production. Dr. Lipsky explained that ectopic germinal centres are lymphoid aggregates that form in places where they do not belong. For example, in patients with rheumatoid arthritis (RA) germinal centres can develop in the synovium. In these structures, B cells encounter autoantigens and are positively selected. Therefore, unlike germinal centres in secondary lymphoid organs, ectopic germinal centres do not delete autoreactive B cells. The net effect of these activities is augmented inflammation resulting in tissue damage. The implications of such a model are that 1) there is no specific causative agent for individual autoimmune/rheumatic diseases; 2) organ involvement relates to genetics, the site of non-specific inflammation and the nature of the immune response; 3) similar but distinct processes drive all autoimmune/rheumatic diseases (a radical idea that needs to be considered); and 4) effective therapies target pathologic processes and not causation, which suggests that many of the same therapies will work in these diseases. Actually, the preceding statement is contradicted by our current knowledge of the response of various immune-mediated inflammatory diseases to therapeutic interventions. One example of this would be methotrexate, which is effective in patients with RA but does not work in those with psoriasis or IBD.
Finally, Dr. Lipsky cited a study from Holland which taught an important lesson about RA. In this study the researchers wanted to address whether an abnormal immune response could be identified in persons with RA before they presented with symptoms. Critical to the success of this study was the fact that Holland has a very organized medical system with thorough records of blood donors. The group wrote letters to RA patients asking them to give blood and of the 80 people who responded, 79 had been blood donors before RA onset. Next the researchers looked for the presence of autoantibodies in the serum samples and they found that about half of the patients had detectable autoantibody titers (anti-rheumatoid factor, anti-citrolynated peptide or a combination of both) as early as 8 to10 years before the onset of symptoms. This suggests that the disease process starts very early as B cells are activated to produce specific antibody years before symptoms occur. More importantly this information could be used to identify a population at risk and later study what causes a person to go from autoantibody production to disease development.
Plans for Cohort Studies of the Canadian Lifelong Health Initiative
Dr. John McLaughlin, Samuel Lunenfeld Research Institute, and the University of Toronto
Dr. McLaughlin described a cross-cutting, strategic, multi-institute initiative of the CIHR, the Canadian Lifelong Health Initiative. Included within this program are two longitudinal studies, the Canadian National Birth Cohort (CNBC) and the Canadian Longitudinal Study on Aging (CLSA), which are currently being considered and designed. Dr. McLaughlin discussed the status of these studies with respect to the concept, design and progress towards creating the cohorts. Identified as one of the main underlying themes is the study of gene/environment interactions; this concept relates to the theory that most disease burden is jointly determined by individual genetic endowment and complex environmental factors. These gene/environment interactions require decades to fully manifest over the life course. Moreover, diseases and conditions of later life occur in some individuals and not others because of the relationship between particular genetic constitutions and exposure to certain social and physical environments. Little is known, however, about the underlying causes of several conditions and why they are increasing in frequency (e.g., asthma). To understand the causal pathways and develop disease prevention and control strategies, sequential events must be studied in large numbers of people on whom baseline genetic and repeated environmental exposures are taken over time. The determinants of disease represent the genetic component, environment, diet and lifestyle, and from the population and public health perspective, the social structure. Ultimately, this web of causation can be approached by studying population genetics and by genetic epidemiology. Considering the life course perspective, from pregnancy through infancy to adulthood there are several factors which influence the development and occurrence of subclinical and clinical conditions that ultimately affect health outcomes.
The CLSA concept was initially led by the CIHR Institute of Aging; several other Institutes have agreed to assist in the development of this cohort. The CLSA design and procedures are being developed by a research team consisting of 3 principal investigators, 20 co-investigators and 200 collaborators representing 26 universities in 10 provinces. The rationale behind the CLSA includes the reasoning that longer life expectancies seen in the Canadian population create serious burdens for the health care system and social programs. This presents the need to characterize aging beyond the presence of disease, disability and frailty. In fact, little is known about the aging process. The preliminary aims of the CLSA are to examine aging as a dynamic process; investigate the interrelationship among intrinsic and extrinsic factors from midlife to old age; capture the transitions, trajectories and profiles of aging; and provide infrastructure and build the capacity for high-quality research on aging in Canada. More specifically the goal would be to determine how changes over time in things such as genetic and biochemical factors and exercise, nutrition or other health behaviors, are interrelated and influence disease states and how they might contribute to "healthy aging." If funded, the study is likely to involve a longitudinal design of Canadian men and women aged 40 and over, with a large sample size (e.g., up to 50,000) and requiring a long period of follow-up, possibly 20 years. It should also be noted that embedded within this large infrastructure would be the opportunity for more detailed substudies. The fundamental goal would be to generate a publicly accessible national database.
In contrast, the CNBC is at an earlier stage of development and it will probably be several years before the study is implemented. The CNBC presents an exciting opportunity to be recognized internationally as unique by designing a multigenerational birth cohort. The objective of designing this cohort would be to study common genetically complex/multifactorial outcomes up to age 15. In conclusion, Dr. McLaughlin remarked that ethical, legal and social issues are clearly a concern for both cohorts. In order for these studies to benefit Canadians and the scientific community they must have very strong foundations. Public trust is required for public participation and is therefore vital to the success of such endeavors. Ways to deal with informed consent and disclosure issues related to the use of biological samples must be addressed. Ultimately, it would be of utmost importance to create a useful link between the findings of both the CLSA and CNBC studies.
| Topic | Presenters |
| Immunological Principles Underlying the Pathogenesis and Regulation of T Cell-Mediated Autoimmune Disease | Dr. Steve Miller |
| Compartmentalisation of Immune Responses to Commensal Intestinal Bacteria | Dr. Andrew Macpherson |
| Trafficking of Leukocytes in the Brain: Learning by Watching Leukocyte Behaviour | Mr. Steven Kerfoot (for Dr. Paul Kubes) |
Immunological Principles Underlying the Pathogenesis and Immunoregulation of T Cell-Mediated Autoimmune Disease
Dr. Stephen D. Miller, Northwestern University Medical School
Dr. Miller discussed insights from his group's study of animal models of multiple sclerosis (MS), the immunological principles underlying disease pathogenesis, and recent data on disease intervention using antigen-directed immunotherapies. As a CD4 T cell-mediated autoimmune disease, MS attacks myelin in the CNS. There are thought to be two possible triggering events in MS: one is the loss of immune regulation leading to the activation of autoimmune responses against neuroantigens, while the other is an infectious agent trigger for at least some forms of disease. The latter hypothesis is suggested by epidemiological evidence. Dr. Miller's laboratory studies two mouse models of MS, experimental autoimmune encephalomyelitis (EAE) and Theiler's virus-induced demylinating disease. In SJL mice, EAE has a relapsing-remitting clinical course whereas Theiler's virus-induced demylinating disease has a chronic-progressive course. It is interesting that in the same genetic background there are two autoimmune diseases which present themselves clinically in completely different ways; however, if you examine the lesions in the CNS it is difficult to distinguish between them. Moreover, both diseases are characterized by epitope spreading. In the EAE model, where disease is induced by the defined proteolipid protein (PLP) 139-151 peptide, CD4 T cells specific for the initiating antigen are responsible for the acute phase. The primary relapsing episode is caused by T cells specific for non-cross reactive epitopes of PLP, and is termed intramolecular spreading. As disease progresses to the secondary relapse, the response switches to an epitope of myelin basic protein (MBP); this is termed intermolecular spreading. The same phenomenon drives the induction of autoimmunity in the virus-induced model, where disease onset is induced by T cells recognizing viral antigens and by the late chronic phase the response is against myelin and viral epitopes. Epitope spreading is initiated in the CNS and is associated with the appearance of what are believed to be CD11c+ DCs. The implications of epitope spreading to the pathogenesis and immunotherapy of MS include the intimation that these autoimmune responses are dynamic and evolve over the course of this chronic disease and that it would be problematic to use peptide-induced tolerance as an antigen-specific therapy because determining which epitope would be next in the pathologic sequence would be practically impossible. This suggests that treatments targeting co-stimulatory molecules which do not require prior knowledge of autoreactive epitopes, but which can result in antigen-specific tolerance, may be effective in treating MS. The current clinical treatments employed in MS include the use of corticosteroids, interferon-ß, copolymer 1 and bone marrow transplantation in severe cases. Unfortunately, these approaches are largely ineffective and are non-antigen-specific.
Dr. Miller's laboratory is interested in designing specific therapies to intervene in the epitope spreading cascade which leads to chronic disease. One of their approaches involves using antibodies to block co-stimulatory molecule interactions or the CD3 signalling complex. Another strategy would be to use antigen-specific tolerance to prevent the activation of initiating T cells or in animals with ongoing disease to inhibit epitope spreading. Studies in SJL mice have shown that intervention using the F(ab') fragment of the anti-B7.1 molecule, which blocks the B7/CD28 interaction, effectively decreases the number of relapses if mice are treated during disease remission. This also correlates with a period of unresponsiveness in the T cells specific for the region of the PLP molecule that is involved in epitope spreading. One of the complications that can arise from antibody therapy is exemplified by anti-B7 molecules. Disease can be exacerbated if the intact anti-B7.1 molecule is used, suggesting that the intact antibody may signal, whereas the F(ab') fragment blocks signalling. This strategy has also been successful using anti-CD40L. If anti-CD40L is applied at the time of disease priming, it can very efficiently prevent the initiation of EAE; but more importantly, if given at the peak of the acute phase or during a relapse anti-CD40L, it can prevent further relapses. It is believed that therapies that block costimulation inhibit the differentiation of proinflammatory Th1 cells. In collaboration with other laboratories, Dr. Miller's group has shown that non-mitogenic anti-CD3 F(ab')2, if given at the time of disease onset or at the peak of the acute phase, can efficaciously inhibit disease onset or relapses respectively. Interestingly, if anti-CD3 is given at the time of disease priming it has no effect, suggesting that it targets previously activated T cells, which would be ideal for treating autoimmune disease. Lastly, Dr. Miller discussed the possibility of inducing antigen-specific tolerance. This can be achieved using APCs pulsed with peptide and treated with a chemical cross linker to prevent the delivery of the co-stimulatory signal. If antigen-specific tolerance is applied at the time of disease remission following the acute phase, it can very effectively prevent the progression of EAE.
Compartmentalisation of Immune Responses to Commensal Intestinal Bacteria
Dr. Andrew Macpherson, University of Zürich
Dr. Macpherson addressed how non-pathogenic environmental organisms shape the immune system. For example, germ-free animals kept in isolator cages and fed sterile food and water have no intestinal bacteria. However, post-colonisation these animals have hardly any IgA in the small intestine, the Peyer's patches are hypoplastic with relatively few germinal centres, and there are differences in the T cell content of the intestine. Despite a strong local immune response, clean mice are systemically ignorant of their commensals. These features denote the profound differences in the mucosal immune system, which is not to say that microorganisms comprising the flora are ignored; actually, they have a very important effect on the mucosal and systemic immune systems. This raises two very interesting questions: 1) How can the mucosal immune response to the commensals be separated from the systemic response? and 2) Does breaking systemic ignorance prime autoimmune pathology? Dr. Macpherson's group has addressed the first issue by demonstrating that small numbers of commensal bacteria are carried to the mesenteric lymph nodes (MLN) in DCs. The evidence for this is provided by a simple, reproducible experiment whereby mice are inoculated with Enterobacter cloacae by gavage or intravenous injection in the tail vein and various tissues are examined for live bacteria at different time points. In mice that are gavage fed, a peak in bacteria in the MLN can be seen for about 72 hours, while there are none in the spleen. In contrast, mice receiving an IV injection clear the organisms to the spleen. This shows that there is absolute preservation of the geographical containment of these organisms in the challenge dose unless the animals have the MLN surgically removed, and then the gavage dose appears in the spleen. Upon FACS-sorting of MLN cells, the organisms appear in the DC fraction while surprisingly not in the macrophage fraction. Commensals are not found within macrophage because they are rapidly killed. In contrast, the pathogen Salmonella typhimurium, which can survive intracellularly by subverting bacteriocidal mechanisms, can be found within these cells, suggesting that there is a Trojan Horse effect for the commensals existing within DCs. Using a Thirry Vella loop system, Dr. Macpherson's group has shown that commensal bacteria travel to the MLN within DCs and do not just penetrate as free organisms. The functional effect of DCs carrying bacteria as passengers is shown by stimulating the mucosal immune system with repeated challenges with the organism resulting in substantial IgA induction. The induced IgA has a protective role to limit the penetration of commensals. Interestingly, the CD11c+ DCs may be CD8a+ or CD8a- but in either case the cells have upregulated the co-stimulatory molecule CD86, suggesting that they are activated.
In his conclusion, Dr. Macpherson summarized the following points: 1) separate priming of the mucosal compartment is essential to maintain relative systemic ignorance of the commensals since the consequences of unwanted systemic priming are profound but how they influence classical autoimmune models are yet to be explored; 2) commensals find their evolutionary niche in the lumen and unlike pathogens do not subvert microbiocidal killing mechanisms; and 3) compartmentalisation is achieved by the DCs retaining very low numbers of commensals within the mucosal circuit, allowing them to prime the mucosal compartment selectively and locally, but this can probably be broken to a small extent in immunopathology.
Trafficking of Leukocytes in the Brain: Learning by Watching Leukocyte Behaviour
Mr. Steve Kerfoot on behalf of Dr. Paul Kubes, University of Calgary
Mr. Kerfoot discussed techniques used in Dr. Kubes's laboratory to image inflammation and leukocyte recruitment in vivo. Specifically, he described intravital microscopy of the brain and emerging technology for imaging leukocytes in vivo. Leukocytes are recruited to sites of inflammation through a well-characterized cascade of events. In response to inflammation, endothelial cells are activated to express adhesion molecules, which allows leukocytes in circulation to initially tether to the endothelium followed by rolling. If the leukocyte encounters an appropriate signal, such as a chemokine, the cell will upregulate surface integrins, flatten out and then transmigrate into the tissue. Each step of this cascade is a prerequisite for the next with a few exceptions; for example, a previously activated cell can tether and immediately adhere, bypassing the rolling step. Furthermore, different types of adhesion molecule are important at different steps: selectins are required for tethering and rolling while integrins are necessary for firm adhesion. These steps can be imaged in vivo using intravital microscopy. In the case of the brain, a piece of the skull is removed along with the dura mater to reveal the underlying microvasculature. Rhodamine 6G is then administered intravenously to label all of the circulating leukocytes. Finally, using fluorescence microscopy, it is possible to watch the leukocyte/endothelial cell interactions in the blood vessels. When visualizing the cerebromicrovasculature of a healthy control mouse there is little baseline leukocyte recruitment. This is contrasted when examining a mouse that is developing EAE, where a tremendous number of rolling and adherent leukocytes can be seen. These cells will then transmigrate into the tissue causing demyelination and destruction associated with disease. It is possible to study disease development using an actively induced model of EAE in C57BL/6 mice, whereby the mice are immunized with MOG peptide and pertussis toxin (PTx) resulting in a very predictable disease course. Symptoms begin around day 12 and this developing phase is followed by an acute phase after which point the mice do not improve. However, their condition does not worsen and this is considered to be the chronic phase. Using intravital microscopy to examine leukocyte recruitment at these stages of disease in pre-symptomatic mice, an induction of rolling events which peaks in the acute phase and diminishes in the chronic phase is already seen. Moreover, a very similar pattern is seen for leukocyte adhesion. Interestingly, mice treated with adhesion molecule blocking antibodies against a4-integrin, P-selectin or a combination of both, showed significant reductions in rolling and adhesion events. This is especially important, as anti-a4-integrin is currently in trial in patients with MS with some promising results. It is inferred that the antibody works by preventing leukocyte infiltration into the brain. However, this can not be proven until it is possible to image the process in vivo, in real time, in the target organ.
Mr. Kerfoot then introduced some new and exciting work from Dr. Kubes's laboratory. As mentioned above, PTx is used to induce EAE but the mechanism of action is unknown. Dr. Kubes's laboratory now has evidence that suggests PTx may act like environmental factors in influencing disease induction. PTx alone can induce leukocyte rolling and adhesion in the brains of otherwise untreated mice. Interestingly, in Toll-like receptor 4 (TLR4) deficient mice this recruitment was completely eliminated, implying that PTx induced leukocyte recruitment is mediated through TLR4. A major drawback of current studies using intravital microscopy is that the cell type is unknown. This issue could be addressed by studying subset-specific mechanisms of recruitment using purified, fluorescently labeled cells transferred into mice with EAE. A more elegant system could perhaps employ transgenic mice that have subset-specific expression of a fluorescent protein. Importantly, Dr. Kubes's laboratory has learned from looking at various animal models of autoimmunity that it is vital to study the target organ as different mechanisms of leukocyte recruitment predominate in different tissues.
A new technique that is currently in development for use in imaging leukocyte recruitment in vivo is magnetic resonance imaging (MRI). MRI is used clinically to identify demyelinating lesions in patients with MS and now a number of small animal facilities are available. In this strategy, leukocytes are labeled with magnetic agents, which would then appear on MRI scans where they accumulate. Significant advantages of this technique include the ability to perform longitudinal studies, and localization of leukocytes throughout the body and to lesions specifically. In his concluding remarks Mr. Kerfoot stated that in vivo imaging is a powerful tool to understand leukocyte recruitment by permitting direct observation of the process. Moreover, new technologies will permit subset-specific studies. Ultimately, the goal would be to use our knowledge regarding leukocyte recruitment to aid in design of anti-inflammatory therapies.
Co-Chairs: Dr. Steve Collins and Dr. Pere Santamaria
| Topic | Presenters |
| Epidemiological Prognostic Models | Dr. Claire Bombardier and Dr. Sheilah Hogg-Johnson |
| The Application of Proteomics to the Study of Human Disease | Dr. John Wilkins |
| Emerging Genomic Tools to Study Autoimmune and other Complex Diseases | Dr. Alexandre Montpetit |
| Towards an Integrated and Intelligent Molecular Medicine | Dr. Igor Jurisica |
Epidemiological Prognostic Models
Dr. Claire Bombardier (in association with Dr. Sheilah Hogg-Johnson), Toronto General Research Institute
As described by Dr. Bombardier, assembling cohorts for study is a challenging task. Population epidemiologists look at healthy population cohorts with the goal of identifying risk factors associated with disease onset, while clinical epidemiologists deal with people who already have disease. The challenge in studying a healthy population is that not many people actually get disease so the denominator is huge. However, once people have the condition the challenge is to design clinical cohorts that will be useful for recognizing prognostic factors. Using the example of rheumatoid arthritis (RA), Dr. Bombardier discussed classic predictors of disease outcome which include the presence of rheumatoid factor (RF) as a prognostic factor; disease activity at presentation; spread in terms of structural damage; functional ability; and variables such as socio-economic status apart from biological factors that have impact on disease outcome. Some of the problems that clinical epidemiologists are faced with involve assembling and maintaining the cohort. In order to test prognostic factors, an inception cohort is needed. This cohort is defined as a group of patients who are recruited at a uniform point in the course of disease and followed from that point onward. Ideally, the point of inception would be the onset of first symptoms. In practice this is not possible as most patients tend to delay a visit to their doctor. Hence, an alternative inception cohort could be followed from the first visit to the doctor. Similarly, another inception cohort could be formed following referral to a specialist, such as a rheumatologist, and finally the point at which the first diagnosis is made could comprise yet another inception point. Most studies will use the point of diagnosis as time zero, so it becomes evident that the cohort looks very different with respect to length of follow-up. It is important to note the implications with respect to times of follow-up, characteristics of disease, and prognostic factors if an earlier inception cohort was selected, and how different studies could be based on inception point. A common mistake a physician will make when designing a cohort is to see a patient at the time of diagnosis and obtain details on when his or her symptoms first started and use this information to set time zero. It is not a good practice to go backwards because this biases the cohort with patients with chronic disease. In reality, a cohort of patients that is assembled at the onset of first symptoms will not all progress to disease. In order to design more upstream studies, such as at the time of first symptoms, we need a better understanding of the magnitude of these populations.
Dr. Bombardier next discussed the enormous challenges of retaining the cohort, loss to follow-up, and missing data. Dr. Bombardier's colleagues at the Institute for Work and Health modeled the impact of loss to follow-up on the estimate of odds ratio of how strong the factor of interest is related to the outcome. The model plots % of patients loss-to-follow-up versus % confidence interval including true odds ratio (i.e., is the true estimate within the calculated confidence interval?). There are two types of missing data: random missing data and non-random missing data. With random missing data, the true estimate of the odds ratio will commonly be within the confidence interval. This means that with a substantial amount of random missing data, the confidence interval will be very large but the true odds ratio will be within that confidence interval. Conversely, if there is more than 20% non-random missing data, the true odds ratio lies outside the confidence interval and will result in completely biased information. This implies that it is absolutely necessary to have less than 20% loss to follow up, but this is very difficult to achieve. Dr. Bombardier and colleagues are conducting a study across North America of approximately 1000 early RA patients (<1 yr) who will be followed for five years. At the one-year mark there is already 8% missing data. Over a five-year period this could accumulate to 40% missing data, which exemplifies the importance of maintaining both the cohort and the data.
When trying to predict outcomes for the cohort, the challenge is determining what outcome to examine. With respect to RA, one could predict inflammatory disease activity, structural damage, function, quality of life, or mortality. Studies have suggested when looking at long-term effects it is best to measure a variety of outcomes rather than just one. Furthermore, using the appropriate statistical technique to study the cohort presents a great challenge because there are several variables at baseline. The statistical method chosen will really depend on the variable that is being studied.
To summarize, Dr. Bombardier discussed a conceptual framework to classify prognostic studies. First the internal validity of the cohort must be considered: how it was assembled, the follow-up, the quality of the outcome, and the use of appropriate statistical techniques. Secondly, it is important to bear in mind that the findings may be true for patients in the cohort with very severe disease but may not be applicable to all patients. In addition, the current stage of knowledge in the field should be determined. We should identify whether ongoing and previous studies are generating hypotheses from descriptive studies or whether ongoing and previous studies are testing specific theories.
The Application of Proteomics to the Study of Human Disease
Dr. John A. Wilkins, Manitoba Centre for Proteomics
Dr. Wilkins provided three specific examples of proteomics approaches to the study of human diseases. He also discussed issues of patient selection and sample acquisition. First, Dr. Wilkins described an example of 2D analysis as a tool for obtaining compositional information on clinical samples of interest. In this approach samples are separated based on molecular weight and isoelectric point, and the resolved proteins are excised and digested in a gel with trypsin. Subsequently, the peptides are extracted and analyzed by mass spectrometry, generating a peak list that can be subjected to database searches for protein identification. Using synovial fibroblast lysates, Dr. Wilkins's group extracted 390 spots and identified 328 proteins, which included autoantigens, regulatory and novel species of proteins found in patients with rheumatic diseases. A second approach which shows great promise is a proteomics based approach for defining the specificity of autoantibodies. Serum or antibodies from autoimmune individuals are used to select antigens by immunoprecipitation or western blot and the antigens are identified by mass spectrometry. As a cautionary note, it was pointed out that only about 10 to15% of antibodies will actually blot. Using this method for antigen detection, a large subset of potentially physiologically significant autoantibodies may not be detected. Nevertheless, this approach may be useful for tracking autoantibody repertoires during the course of autoimmune development. The third approach that Dr. Wilkins discussed was chip-based surface-enhanced laser desorption ionization (SELDI) analysis. The feature that makes this approach particularly useful and unique is the use of retentate chromatography in conjunction with mass spectrometry. As an illustration of the application of this technique, Dr. Wilkins presented a specific study examining renal transplant patients. The intent was to identify biomarkers associated with acute transplant rejections. The SELDI profiles of urinary proteins fractionated by retentate chromatography were compared between controls and patients with either stable or rejecting transplants. A set of biomarkers were identified which allowed for differentiating between the two patient groups. Those with stable transplants displayed profiles similar to those of non- transplanted controls. In contrast, those patients with ongoing rejections each had a unique profile. These results indicate that it is feasible to identify biomarker patterns that are associated with acute rejection. This type of marker may be useful to monitor disease progression and predict flares in activity. Identification of these biomarkers may also provide a basis for a better understanding of the pathogenic processes involved in the disease process. Ultimately, this type of approach may be useful in assessing treatment regimens and monitoring a patient's response.
Several issues related to study design and patient selection were presented. These included intra versus inter patient comparisons. Depending on the patient heterogeneity, intra patient comparisons may be more informative. Knowledge of disease stage is critical for comparisons between patients. Unlike the genome, protein expression patterns are dynamic, will change over time, and will be influenced by therapies. The source and accessibility of clinical samples are important considerations, especially if multiple samples are required. Ideally, one would like samples from the affected target organ. In order to establish a sample bank, it is important that the samples are processed and stored in such a way to ensure that their integrity is preserved. Detailed clinical data is an essential component of any medical proteomics initiative, as are data analysis and bioinformatics. The latter are required for data acquisition, protein identification, and analysis.
In his concluding remarks, Dr. Wilkins stated that our proteomic capabilities are rapidly increasing and large/broad scale analysis offers new opportunities. However, the power of the technology also presents new challenges in terms of complexity and quantity of data generated. The take home message from the analyses performed to date is that simple, well-designed studies with the most homogeneous patient populations obtainable are critical. Attention to these parameters offers a reasonable chance of success Achievement of these objectives requires the integrated efforts of clinical and basic scientists. In the context of autoimmune diseases, these issues pose exceptional challenges.
Emerging Genomic Tools to Study Autoimmune and Other Complex Diseases
Dr. Alexandre Montpetit, McGill University and Genome Quebec Innovation Centre
Dr. Montpetit discussed the application of genomic tools to investigate the genetic basis of autoimmune and other complex diseases. The phenotype associated with a given disease and the underlying genetic defect can be studied by linkage analysis or association. However, due to the large size of the human genome, linkage analysis is more appropriate for mapping on a genomic scale. Using SNPs or microsatellite DNA, linkage analysis can be performed. Dr. Montpetit described the application of a DNA chip based technology using SNPs for performing such analyses that affords many advantages over using microsatellite DNA markers. Similar conclusions can be derived whether using SNPs or microsatellite DNA. Most significantly, the SNP chip analysis allows for very rapid genome scans. The usefulness of performing linkage analysis with an SNP DNA chip in familial studies is complicated in part by the fact that a large cohort of families is required to reliably demonstrate linkage of a given phenotype with a genetic locus. Application of linkage analysis using SNPs was exemplified using inflammatory bowel disease (IBD). In IBD a linkage peak associated with the long arm of human chromosome 5 was identified and the SNPs within this region were characterized. This enabled the identification of an IBD susceptibility haplotype. This study demonstrates the utility of linkage analysis using SNPs for the identification of disease susceptibility loci and is the basis for the Haplotype Project currently being developed (www.hapmap.org). The haplotype project involves the identification characterization of SNPs throughout the genome in European, Asian, and African populations. Currently data on the 145, 554 SNPs has been released on the HapMap web site.
Towards an Integrated and Intelligent Molecular Medicine
Dr. Igor Jurisica, University of Toronto and Queen's University
Dr. Jurisica discussed the computational aspects, challenges and possibilities that high throughput data obtained from microarray and protein array analyses afford in the characterization of complex diseases. The identified genetic markers may vary between separate studies of a given complex disease. The reasons for this discrepancy are that many subtypes of disease may exist, most studies involve differential analysis, and the approaches to data analysis may differ between studies. Dr. Jurisica described a multi-faceted approach to data analysis. First, data is analyzed in an unbiased manner allowing new hypotheses to be put forth that require subsequent validation through experimentation or further statistical analyses. Multiple platforms exist to acquire experimental data; however, very little overlap exists between each platform. This was illustrated using as an example a recent analysis of data obtained for lung carcinoma where multiple large microarray studies were employed. The complexity of many diseases is large and to date no platform exists that encompasses all genes. By using multiple platforms the identification of genetic markers can be addressed systematically. Dr. Jurisica stressed the importance of knowing the differences between what is being compared in each study. Molecular profiling was also introduced as a useful analysis for the characterization of disease. In this capacity the underlying disease processes can be identified without overt clinical diagnosis. Molecular profiling may also be useful for establishing stages of disease progression. However, using this technique, it is difficult to get patient samples which give conclusive results. It is important to perform analysis without bias in order to obtain some results but obtaining statistically significant results may not be possible. This means new studies must be developed and re-examined. Molecular profiling attempts to take a large screen and to identify a reduced number of disease associated markers that can subsequently be validated. This is a form of data compression and also a process of visualization.
Another approach to analyze gene or protein expression profiles describes these profiles as a two-dimensional table. The data in the table can be clustered and analyzed either unsupervised or with some degree of bias. Dr. Jurisica's group uses two types of analysis: one combines a modified K-means clustering and SOM organizing maps in a complimentary way. SOMs is an approach that combines vector organization and vector quantification. Statistics can be applied after both dimensions of the table have been clustered (i.e., samples and genes or samples and proteins), because clustering ensures that homogeneous groups are compared.
Chair: Dr. Jack P. Antel
| Topic | Presenters |
| NIAID/NIH Funding and Strategic Planning for Autoimmune Diseases Research | Dr. Daniel Rotrosen |
NIAID/NIH Funding and Strategic Planning for Autoimmune Diseases Research
Dr. Daniel Rotrosen, National Institutes of Health
Dr. Rotrosen presented an overview of NIAID/NIH funding and strategic planning as related to autoimmune disease. In the 2003 fiscal year the NIH awarded an estimated $591 million dollars for autoimmunity research, which represents about 2.2% of the $27 billion dollar NIH budget. This percentage is actually slightly lower than what is spent on autoimmunity in Canada. The NIH consists of 27 institutes and centres. The three institutes with the largest expenditures for autoimmunity research are NIAMS, NIAID and NIDDK, each over $100 million. The majority of the NIH budget is awarded to investigator-initiated research project grants, usually to single investigators; but an increasing portion is now being allocated to "large-science" team projects. Approximately 45% of the autoimmunity expenditures are in projects examining pathogenesis and immune dysfunction, while 3% of the funds are allocated to the development of new animal models.
In 1998, Congress required the NIH to convene what is now known as the Autoimmune Disease Coordinating Committee (ADCC). It is comprised of representatives from 22 NIH institutes, as well as from various other federal agencies and private organizations with research programs and advocacy interests that include autoimmunity. Congress called for the committee to develop a strategic plan for coordination of research across NIH and a plan was submitted to Congress in 2002. In addition to the normally appropriated funds, Congress appropriated $150 million dollars a year for type 1 diabetes research that bypasses the normally appropriated NIH budget. These funds are overseen by NIH under the leadership of NIDDK, and the awards are made by a variety of NIH institutes whose missions include prevention or treatment of complications of type 1 diabetes. The ADCC research plan supports and recommends platforms such as the following: support for basic research and clinical trials, expanded support for research resources, registries, repositories, reagent production and distribution, and core facilities. The research plan also proposed increased support for epidemiology and surveillance studies. Interestingly, one of the things that was very clear after assembling this report is that in the United States very little is known about the real prevalence and incidence of autoimmune disease. The plan supports the development and implementation of public awareness and professional education programs.
Another program the NIH expanded in 2003 is the Autoimmunity Centers of Excellence. Under the RFAs soliciting this program, each of the 9 centres is required to bring together physicians and basic scientists representing 3 or 4 medical disciplines spanning autoimmune diseases, for example, neurology, gastroenterology, and rheumatology. By doing so, a very strong network of centres is now capable of conducting multi-site trials in addition to basic research. Approximately 3/4 of the budget for these centres funds basic research, but there is a pilot clinical trials program that is an important component. NIH also supports an accelerated grants program designed to rapidly review and award proposals for mechanistic studies that are performed in conjunction with industry sponsored clinical trials.
The Immune Tolerance Network (ITN) is another NIAID sponsored research consortium that supports clinical trials in autoimmune disease. The ITN supports trials through a national and international network of clinical trial sites and its budget includes resources for state-of-the-art core facilities. These core facilities permit assay centralization for quality control and data accessibility, rapid acquisition and dissemination of data, and central data collection for higher order analyses. The core facilities have both R and D and reagent production/distribution functions. A few of the current ITN core facilities include the following: a tissue sample repository that can handle up to tens of thousands of samples, MHC class I and II tetramer facilities, genomics (microarray and real-time PCR), proteomics, MHC typing, and a variety of other activities. Sample processing is highly regulated at clinical sites whereby the sample goes through a bar coding process and is immediately linked to protocols in the database. The samples are shipped to a repository and from there they may be sent to one of the core facilities for analysis. From these core labs, the data are deposited in a central data server accessible to both ITN sponsored investigators and, following publication of results, to the general research community to do data mining, analysis and hypothesis generation.
Lastly, Dr. Rotrosen discussed strategic and research planning at the NIH. In order to prepare an RFA, the NIH holds meetings, similar to the current CIHR research symposium, where experts from the academic community and industry are assembled to give their opinions. In general, the entire process takes a year and a half and perhaps even longer to initiate large programs like the ITN. Another important aspect of long-range initiative planning is to enable institutes with shared missions to coordinate activities in order to share costs. For example, many of the programs described above have been co-sponsored by multiple institutes at the NIH. Most of the programs, especially those that involve clinical research, are very heavily dependent on strong partnerships with industry. The ultimate goal of these clinical research programs is to bring new vaccines, immune-based therapies and diagnostics to clinical practice.
During the scientific presentations, participants were asked to consider:
a) supports (e.g., infrastructure, capacity building) and opportunities for a Canadian health research agenda with a focus on integrated discovery platforms in autoimmune diseases
b) key challenges and considerations when developing a framework for a Canadian health research agenda with a focus on integrated discovery platforms in autoimmune diseases over the next ten years
Following the presentations, participants worked in small, mixed groups to discuss these two areas with the goal of selecting strategic research directions. Some groups noted overlap between these two areas. The following is a summary of discussions.
Participants came to an agreement on nine strategic research directions which are listed below in an order that reflects participants' perception of priorities and their enthusiasm in relation to the importance of these directions.
After identifying the above areas, participants were engaged in small group discussion. Each group addressed one of these areas for further exploration and developed a report that described the area, provided examples of new research questions, and suggested supports required to enable implementation of the research.
Given the time available and representation in the symposium, one group focused on three areas: Cohort Methodologies, Biometrics and Methods for Early Case Findings. The report on Biometrics1 from this combined group reflects the clinical perspective of group members and the fact that those with an interest in computational methods selected other topic areas.2
While discussing strategic research directions, participants also identified the following area for cross-CIHR Institute Collaboration: "Development of new animal models for identification of early markers to validate target discoveries across a variety of diseases."
Methods for Early Case FindingsThis strategic research direction addresses the challenge of going upstream with respect to autoimmune diseases. There is a need to
Examples of new research questions that could provide significant value in this research area include:
Supports are required to ensure that the research questions in the previous three sections are addressed. A top priority in terms of capacity development involves the creation of a funded standing group or superstructure that enables integrated approaches among autoimmune disease researchers. Clinical researchers in autoimmune disease face many similar problems across Canada, both within individual diseases and across diseases. The purpose of this standing group would be to identify, clarify and address issues such as information technology, bar codes, privacy, data security, standards for tissue collection and handling, innovative methods for data collection, and access to billing data.
BiometricsThere is a need to include analysis of the determinants of health variables in cohort studies across the autoimmune diseases and across CIHR Institutes (e.g., Institute of Population and Public Health, Institute of Health Services and Policy Research). Currently researchers are collecting the determinants of health differently (e.g., ethnicity, social levels and education), which is causing confusion and potential misinterpretation.
Examples of new research questions that could provide significant value in this research area include:
This strategic direction involves cohort research, including methods of assembling and maintaining cohorts.
Examples of new research questions that could provide significant value in this research area include:
The focus of this strategic research direction is to understand the factors that impede or enhance repair and remodeling as part of the biologic response to tissue damage, and to develop experimental and clinical strategies to regenerate healthy tissue. Research in this area can be organized for both individual and team/program applications.
The following research themes are most relevant to this research direction.
| Basic biomedical, e.g., genetic, molecular, cellular, tissue physiology | Applied clinical, e.g., drugs, devices, social intervention | Health systems, health services, e.g., epidemiology, health care quality, cost-effectiveness | Societal, cultural and environmental influences on health and the health of populations |
| + | + |
In the future, research outcomes in this area should be applied to improve health care delivery and quality of life for patients.
The determinants of health most closely linked to this strategic direction are summarized in the following table.
| Income and Social Status |
Personal Health Practices and Coping Skills | ||
| Social Support Networks |
Healthy Child Development |
+ | |
| Education |
Biology and Genetic Endowment |
+ | |
| Employment/ Working Conditions |
Health Services | ||
| Social Environments | Gender |
+ | |
| Physical Environments | + | Culture |
Examples of new research questions that could provide significant value in this research area include:
Supports required to ensure that these research questions are addressed include:
This strategic research direction involves a multidisciplinary approach to the identification, development, validation and implementation of biomarkers. It covers earliest events and ongoing/stage progression. This area includes:
The following research themes are most relevant to this research direction.
| Basic biomedical, e.g., genetic, molecular, cellular, tissue physiology | Applied clinical, e.g., drugs, devices, social intervention | Health systems, health services, e.g., epidemiology, health care quality, cost-effectiveness | Societal, cultural and environmental influences on health and the health of populations |
| + | + | + | + |
The determinants of health most closely linked to this strategic direction are summarized in the following table.
| Income and Social Status |
+ | Personal Health Practices and Coping Skills | + |
| Social Support Networks |
+ | Healthy Child Development |
|
| Education |
Biology and Genetic Endowment |
+ | |
| Employment/ Working Conditions |
Health Services | + | |
| Social Environments | Gender |
+ | |
| Physical Environments | Culture |
+ |
Examples of new research questions that could provide significant value in this research area include:
Supports required to ensure that these research questions are addressed include:
This strategic research direction includes:
This area overlaps with tissue regeneration and the role of microbes, cohorts, biomarker and genetic studies. Immunoregulation is the central hub linking these areas.
The following research themes are most relevant to this research direction.
| Basic biomedical, e.g., genetic, molecular, cellular, tissue physiology | Applied clinical, e.g., drugs, devices, social intervention | Health systems, health services, e.g., epidemiology, health care quality, cost-effectiveness | Societal, cultural and environmental influences on health and the health of populations |
| + |
Given the makeup of the group working on this area and the nature of the topic, the basic biomedical pillar is seen to be most relevant. However, it is recognized that the other three themes also affect the basic biomedical area.
The determinants of health most closely linked to this strategic direction are summarized in the following table.
| Income and Social Status |
4 | Personal Health Practices and Coping Skills | 3 |
| Social Support Networks |
Healthy Child Development |
2 | |
| Education |
Biology and Genetic Endowment |
1 | |
| Employment/ Working Conditions |
6 | Health Services | 5 |
| Social Environments | 6 | Gender |
2 |
| Physical Environments | 6 | Culture |
Many of the determinants of health affect immunopathogenesis and vice versa. In addition, the determinants of health interact with and affect each other. The basis of the above ranking reflects the belief that understanding the cause of disease will affect all other determinants of health.
Examples of new research questions that could provide significant value in this research area include:
Supports required to ensure that these research questions are addressed include:
This strategic research direction includes the following goals:
The following research themes are most relevant to this research direction.
| Basic biomedical, e.g., genetic, molecular, cellular, tissue physiology | Applied clinical, e.g., drugs, devices, social intervention | Health systems, health services, e.g., epidemiology, health care quality, cost-effectiveness | Societal, cultural and environmental influences on health and the health of populations |
| + | + | + |
The determinants of health most closely linked to this strategic direction are summarized in the following table.
| Income and Social Status |
Personal Health Practices and Coping Skills | + | |
| Social Support Networks |
Healthy Child Development |
+ | |
| Education |
Biology and Genetic Endowment |
+ | |
| Employment/ Working Conditions |
Health Services | ||
| Social Environments | Gender |
+ * | |
| Physical Environments | + | Culture |
* The incidence of AID is much higher in females than in males. This is a major unexplored component of autoimmune pathogenesis, with very little research in this area worldwide.
Examples of new research questions that could provide significant value in this research area include:
Supports required to implement effective research teams include:
Without sustained funding for infrastructure and science, researchers can not build capacity for excellence and clinician scientists will not change the existing paradigm of small laboratories, segregation of samples to single hospitals, etc.
Microbial Autoimmune PathogenesisThis strategic research direction includes:
The following research themes are most relevant to this research direction.
| Basic biomedical, e.g., genetic, molecular, cellular, tissue physiology | Applied clinical, e.g., drugs, devices, social intervention | Health systems, health services, e.g., epidemiology, health care quality, cost-effectiveness | Societal, cultural and environmental influences on health and the health of populations |
| + | + | + |
The determinants of health most closely linked to this strategic direction are summarized in the following table.
| Income and Social Status |
Personal Health Practices and Coping Skills | ++ | |
| Social Support Networks |
Healthy Child Development |
++++ | |
| Education |
Biology and Genetic Endowment |
+++ | |
| Employment/ Working Conditions |
Health Services | ||
| Social Environments | +++ | Gender |
+ |
| Physical Environments | +++ | Culture |
Examples of new research questions that could provide significant value in this research area include:
Supports required to ensure that these research questions are addressed are focused on infrastructure:
Dr. Katherine Siminovitch, symposium co-chair, acknowledged the enthusiastic involvement and commitment of participants throughout the session. Dr. Siminovitch also thanked Dr. Bhagirath Singh, Scientific Director of the CIHR Institute of Infection and Immunity, for initiating the event, and the Organizing Committee, speakers and session chairs for their contributions to achieving the session objectives.
Dr. Diane Finegood, Scientific Director of the Institute of Nutrition, Metabolism and Diabetes (INMD) and a researcher in autoimmune diabetes, expressed her support for the results of the meeting as a foundation for developing new research initiatives on integrated discovery platforms. Dr. Finegood also expressed an interest in exploring the strategic research agenda developed at this meeting as input to the evolving research priorities for INMD over the next few years.
In closing the symposium, Dr. Bhagirath Singh confirmed the cross-cutting nature of autoimmune diseases and the need to have Voluntary Health Organizations (VHO) involved in the development and implementation of research frameworks. He emphasized the benefits of having VHO representatives at the symposium and referred to their remarks on the second afternoon, when VHO participants emphasized the importance of inclusive, collaborative approaches to research that would result in clear health outcomes for both patients and caregivers.
Dr. Singh also commented on the presence of researchers across the four CIHR themes and the importance of following through on new relationships developed at this session. He will be sharing the results of the workshop with the Institute Advisory Board and collaborating with other CIHR Institutes and VHO as partners in following through on the results of this symposium. Dr. Singh will also be holding further discussions with the NIH to follow through on suggestions made regarding possible long-term infrastructure partnerships.
Participants rated the workshop a success: an average of 3.5 on a 4.0 scale. They appreciated the fact that a diverse group of individuals representing different diseases and disciplines were represented at the symposium and that they had an opportunity to develop an agreement on a framework for a Canadian health research agenda in autoimmune diseases.
