TRANSLATIONAL RESEARCH SYMPOSIUM 2009

Translational research has primarily focused on moving laboratory research into the clinic with the promise of such research far surpassing its success. Several critical gaps have limited this accomplishment, most notably a result of attempting to develop drugs almost independently of the complex clinical presentation of the disease. Additionally, there exists limited ability to accurately diagnose and stage an individual within the disease process in spite of efforts to incorporate genomic analysis. Drug development is also finding that off-target drug effects, unanticipated in research and unobserved in clinical trials, are the source of failure both clinically and financially.

These factors are further embedded in a complex eco-system involving patient and physician understanding of the disease, treatment options and side-effects as well as compliance and insurance re-imbursement. An analytical approach to both enhanced drug development and clinical decision support requires evaluation of the strengths and weaknesses of existing clinical protocols and understanding the diverse factors that may impact their application to the patient. The implications for developing treatment protocols will be presented which require the integration of significantly diverse data types and algorithms in the example of the use of herceptin in breast cancer.

Michael N. Liebman, Ph.D. is the Managing Director of Strategic Medicine, Inc after serving as the Executive Director of the Windber Research Institute since November, 2003. Previously, he was Director, Computational Biology and Biomedical Informatics at the University of Pennsylvania Cancer Center since September, 2000. He served as Global Head of Computational Genomics at Roche Pharmaceuticals and Director, Bioinformatics and Pharmacogenomics at Wyeth Pharmaceuticals. He was also Director of Genomics for Vysis, Inc and Director of Bioinformatics at the Amoco Technology Company. He has served on the faculty of Mount Sinai School of Medicine in Pharmacology and Physiology/Biophysics. He serves on 14 international scientific advisory boards consults for 5 pharma/biotech companies and has been on the economic development programs in the Philadelphia Life Sciences Sector and the State of Illinois Biotechnology Commission. He is an Invited Professor at the Shanghai Center for Bioinformatics Technology and is currently on the Human Health and Medicinal Chemistry Commission of the IUPAC. He received his PhD in protein crystallography and theoretical chemistry. His research focuses on computational models of disease progression stressing risk detection, disease process and pathway modeling and analysis of lifestyle interactions and causal biomarker discovery and focuses on moving bedside problems into the research laboratory to improve patient care and their quality of life. Recent activities also include computational approaches to drug safety and toxicology with specific emphasis on reducing animal testing.

Biomarker Data Mining (BDM) will lead to a better understanding of drug action, an improved ability to understand physiological responses, and better overview of interrelationships between research and clinical data. MedImmune needs to have a platform that can extract, transform, and load (ETL) data from various disparate storage locations and formats such as current databases and spreadsheets, and assemble and present correlated data for exploratory analyses. Currently used ad-hoc BDM methods, while effective, do not support the scale of a comprehensive biomarker discovery panel required to support the research efforts for our growing drug pipeline.

Therefore, we just completed building a BDM system based on 3 components: (1) ETL, (2) search/sub-setting, and (3) analyses. To build the required BDM platform, we put in place a team of consultants comprised of scientists, data model experts, and solution architects to build and implement a flexible data model that supports a scalable, industrial-strength scientific data pipeline. By employing basic functional building blocks, we based our development on the use of a simplistic approach combined with agile methodology. I will present our experiences and impressions gained from building this BDM platform.

Daniel Ingber is a Research Information Systems Sr. Manager at MedImmune LLC (Gaithersburg, MD) leading a team to build and support information systems for Research. Prior to MedImmune he was employed at Celera and subsequently at Applied Biosystems where he led a team of bioinformaticians responsible for maintaining the Variation Annotation database (SNPs, CNVs) and related genotyping products and activities. Daniel has been working in the Biotechnology Industry for the past eight years, supporting large-scale scientific data content pipelines. He has resided with his family in Washington D.C. for the past 20 years.

Genome wide association studies have delivered on the promise of uncovering genetic determinants of complex disease, using high-throughput methods allowing large volumes of SNPs (105-106) to be genotyped in large cohort studies. The GWA approach serves the critical need for a comprehensive and unbiased strategy to identify causal genes related to complex disease and is rapidly replacing the more traditional candidate gene studies and microsatellite-based linkage mapping approaches that have dominated the gene discovery attempts for common diseases in previous years. As a consequence of employing this array-based technology, over the last three years dramatic discoveries of key variants involved in multiple complex diseases and related traits have been reported in the top scientific literature, including over 400 novel loci with multiple replications in over 80 disease areas by independent groups. In this talk, discoveries will be reviewed and large-scale database efforts discussed and their use in complex genetic disorders and genomics of drug response.

Hakon Hakonarson, MD, PhD, is a Board Certified Pediatric Pulmonologist and Associate Professor of Pediatrics at the University of Pennsylvania School of Medicine.

He is the Director of the Center for Applied Genomics (CAG) at the Children’s Hospital of Philadelphia (CHOP), a high-throughput highly automated genotyping facility founded to identify the genetic causes of autism and other complex medical disorders in children with the objective of developing new therapies. The Center represents a $40 million commitment from CHOP to collect and genotype approximately 100,000 children over a 4 year period. Dr. Hakonarson has an extensive track record in human genetics and has developed an international reputation amongst his peers. He has served in several senior posts in the past, including as the Head of Inflammatory and Pharmacogenomics Research and the Vice President of Clinical Sciences and Development and CSO for the biopharmaceutical industry. Dr. Hakonarson has also been the principal and Co-principal investigator on several NIH sponsored grants, and he has published numerous high-impact papers in some of the most prestigious scientific medical journals, including Nature, Nature Genetics, The New England Journal of Medicine, the Journal of the American Medical Association, The Journal of Clinical Investigation, The American Journal of Human Genetics and the Proceedings of the National Academy of Sciences. With over ten years of experience in pioneering genomic research and genome-wide mapping and association studies, Dr. Hakonarson has intimate knowledge of the complexities of large-scale genomics projects and has put together the necessary infrastructure and workflow processes to unravel these complexities in his role as the Director of the Center for Applied Genomics at CHOP.

Scientists working in biomarker discovery, development and validation require access to data from a variety of sources such as gene expression, genomic, bioimaging, and proteomic studies. A useful “Biomarker Workbench” would provide a platform allowing facile integration of these data types with traditional patient clinical data. Implementation of software designed for specific data types (e.g. commercial gene expression analysis and visualization or image analysis software) can be prohibitively expensive and sufficiently complex that the intended user is unable to employ it effectively.

At Daiichi Sankyo Pharma Development we have begun the implementation of a Biomarker Data Repository for data archiving and visualization using the InforSense KDE. We have found that the KDE allows us to develop a set of data access components which provide a common structure for accessing heterogeneous data and minimizes unnecessary complexity, but is flexible enough to allow access to new analyses or visualizations as the need arises.

Dr. Jim McGurk is Director of Informatics at Daiichi Sankyo Pharma Development. In his role Jim is responsible for implementing systems for Translational Medicine and Clinical Pharmacology concentrating on platforms for modeling and simulation, pharmacogenomics and biomarker discovery. Jim has more than 15 years of experience as a laboratory scientist and informatician in the pharmaceutical industry. He has a Ph.D. in neuroscience from The Rockefeller University

The Medical Home Decision Support System (MHDSS) is a web-based tool developed by the Decision Support Center, Department of Health Plan Management at Walter Reed Army Medical Center. The MHDSS enables physicians, nurses, administrators, and technologists to improve medical management of chronic disease patients, and was implemented in the National Naval Medical Center (NNMC) about a year ago. The MHDSS displays HEDIS benchmarks, and shows the percentage met for diabetes metrics, breast cancer screening, PAP smears, colorectal cancer screening, and asthma control. Using a wizard-like interface, the program allows users to create teams comprised of patients empanelled to primary care managers (PCM’s). Aggregate data for teams, as well as for individual PCM’s show the compliance rates for HEDIS benchmarks, along with summary data for hospitalizations, ER visits, medications, laboratory values and tests. Individual patient data can be viewed, phone contacts documented, and alerts created to facilitate partnering with the patient at the point of care. The MHDSS has helped to improve Medical Home Team HEDIS statistics from 50% to 90% for hemoglobin A1C, LDL and mammography testing. NNMC results as a whole improved for mammograms, colorectal cancer screening, and hemoglobin A1C testing. Future enhancements include adding automated letter functionality, additional patient cohorts and re-admission rates for chronic heart failure patients.

Earning her Ph.D. in Experimental Psychology, Dr. Hamm has spent the past 14 years developing web-enabled data systems for population health, medical management, clinical research, and health plan management. She is the Chief of the Decision Support Center, Department of Health Plan Management at Walter Reed Army Medical Center where she creates data warehouses, electronic health records, decision support systems and uses data mining tools to identify opportunities to improve healthcare. The Decision Support Center was a finalist in DM Review’s 2005 World Class Solution Awards in the Business Intelligence category.

The rapidly evolving field of translational research presents a number of significant IT challenges. As the amount of data generated from both clinical trials and clinical practice increases, effective management of this data becomes increasingly important. With the advent of the high throughput biology era, techniques such as gene expression, proteomics and genetics are now commonly used to provide better molecular definition of disease. Such experimental approaches generate ever increasing volumes of data that need to be integrated with each other as well as with the clinical data silos. However, analysis of clinical data is only half of the translational medicine paradigm. A successful informatics infrastructure enables the seamless transition from basic research to decision support within the clinical setting. Beyond just integration, researchers now require more flexible, yet simplified interfaces to data.
This presentation will focus on our experiences of building integrated clinical and molecular data warehouses and associated analysis and research portals including project objectives, road blocks and challenges encountered, and a demonstration of solutions that we have been able to put into productive use.

Mick Correll is the Associate Director of the Center for Cancer Computational Biology (CCCB) at the Dana-Farber Cancer Institute. Before joining the CCCB, he worked in industry where he focused on the design, management, and implementation of informatics solutions for the pharmaceutical, biotech, and healthcare industries, most recently as the Director of Healthcare Product Management at InforSense LLC. Mick was a Bioinformatician at Lion Bioscience Research Inc, where he was the principle architect of a globally distributed gene annotation and analysis platform. Mick also served as the Head of Professional Services for Lion Bioscience Inc in North America. He holds a BS in Computer Science and a BA in Molecular, Cellular, and Developmental Biology from the University of Colorado at Boulder.

Jonathan Sheldon is responsible for directing InforSense R&D activities including managing InforSense product development and delivery to meet customer needs. Prior to InforSense he was Chief Technology Officer for Confirmant where he was responsible for developing the company's proteomics products and services. Previously he established the first bioinformatics group and was Head of Bioinformatics for 5 years at Roche Welwyn, UK participating in a number of global initiatives within the company. Dr. Sheldon holds a PhD in Molecular Biology/Biochemistry from the University of Cambridge.