Translational Medicine Quality Framework
for Duke Medicine
March, 2011
Translational Medicine Quality
Framework
• This past year Duke became the center of a national
controversy concerning the use of gene expression
data as the basis for clinical trials.
• Questions were raised about the veracity of the data
and the quality of the analyses from one of our
laboratories.
• These laboratory findings were rapidly translated
into decision support instruments for use in human
clinical trials.
TMQF Background
• As the public record now shows, the work of our
investigators was flawed and thus did not
demonstrate that personalized genomic predictors
could be used for creating “tailored” cancer care.
• Problems with data provenance, unexplained errors
related to data management, and implausible data
combinations that escaped detection at two different
levels of independent peer review nullified the work
and led to the trials being closed.
TMQF Committee Formation and
Charge
• Duke Chancellor for Health Affairs Victor Dzau, MD,
convened an internal Duke committee to review
institutional approaches to assure the quality of
discovery science destined for clinical application.
• A panel of external experts was also convened to
review the recommendations of the TMQF
Committee.
• The report is now available at
http://medschool.duke.edu/modules/som_research/i
ndex.php?id=22.
• Faculty and staff comments are welcome at
tmqffeedback@duke.edu
TMQF Committee
Chair: Robert M. Califf, MD
• Donald F. Fortin, MD, Professor of Cardiology
• Vice Chancellor for Clinical Research
• Director, Duke Translational Medicine Institute
Co-Chair: Sally Kornbluth, PhD
• James B. Duke Professor, Department of
Pharmacology and Cancer Biology
• Vice Dean for Research, Duke University School
of Medicine
Faculty Membership
Michael Cuffe, MD, MBA
• Vice President for Medical Affairs, Duke
University Health System
• Vice Dean for Medical Affairs, Duke University
School of Medicine
Elizabeth R. DeLong, PhD
• Professor and Chair, Department of Biostatistics
and Bioinformatics
• Co-Director, Outcomes Research & Assessment
Group, Duke Clinical Research Institute
Henry S. Friedman, MD
• James B. Powell, Jr., Professor of NeuroOncology and Professor of Pediatrics
• Deputy Director, Preston Robert Tisch Brain
Tumor Center
Geoffrey S. Ginsburg, MD, PhD
• Professor of Medicine and Pathology
• Director, Center for Genomic Medicine, Duke
Institute for Genome Sciences & Policy
David B. Goldstein, PhD
• Richard and Pat Johnson Distinguished
University Professor
• Professor of Molecular Genetics and
Microbiology, and Biology
• Director, Duke Center for Human Genome
Variation
Robert A. Harrington, MD
• Richard Stack, MD/Guidant Foundation
Professor of Cardiology
• Professor of Medicine, Cardiology
• Director, Duke Clinical Research Institute
TMQF Committee
Elizabeth R. Hauser, PhD
• Professor and Chief, Division of Medical Genetics,
Department of Medicine
• Director, Duke Center for Human Genetics
Barton F. Haynes, MD
• Frederic M. Hanes Professor of Medicine and
Immunology
• Director, Duke Human Vaccine Institute
Mary E. Klotman, MD
• Professor and Chair, Department of Medicine
Robert J. Lefkowitz, MD
• James B. Duke Professor of Medicine and Professor
of Biochemistry
H. Kim Lyerly, MD
• George Barth Geller Professor of Research in Cancer
• Director, Duke Comprehensive Cancer Center
Ross E. McKinney, Jr., MD
• Professor of Pediatrics
• Director, Trent Center for Bioethics, Humanities, and
History of Medicine
Anthony R. Means, PhD
• Nanaline H. Duke Professor and Chair, Department
of Pharmacology and Cancer Biology
• Professor of Medicine
• Co-Director, Duke Cancer Institute
Christopher B. Newgard, PhD
• W. David and Sarah W. Stedman Distinguished
Professor
• Professor of Pharmacology and Cancer Biology,
Biochemistry and Internal Medicine
• Director, Sarah W. Stedman Nutrition and
Metabolism Center
Bruce A. Sullenger, PhD
• Joseph and Dorothy Beard Professor, Department of
Surgery
• Director, Duke Translational Research Institute
Christopher G. Willett, MD
• Professor and Chair, Department of Radiation
Oncology
• Co-Director, Duke Cancer Institute
Jo Rae Wright, PhD
• Professor of Cell Biology
• Dean of the Graduate School and Vice Provost,
Duke University
TMQF External Review Panel
Eugene Braunwald, MD
• Distinguished Hersey Professor of
Medicine
• Harvard Medical School
• Brigham and Women's Hospital
David L. DeMets, PhD
• Professor, Biostatistics and Medical
Informatics
• University of Wisconsin
David Ginsburg, MD
• James V. Neel Distinguished University
Professor of Internal Medicine and Human
Genetics
• Warner-Lambert/Parke-Davis Professor of
Medicine
• Investigator, Howard Hughes Medical
Institute
• University of Michigan
Xihong Lin, PhD
• Professor of Biostatistics
• Coordinating Director, Program of
Quantitative Genomics
• Harvard School of Public Health
Jeffrey M. Trent, PhD
• President and Research Director
• Translational Genomics Research
Institute (TGen)
• Senior Investigator, Genetic Basis of
Human Disease Division
• Head, Melanoma Therapeutics Lab
Elias A. Zerhouni, MD
• President, Global Research and
Development
• Sanofi-Aventis
TMQF Committee Staff
Ann E. Bradley, BS, BA, JD
• Associate University Counsel, Office of Counsel
• Duke University/Duke University Health System
Paul P. Lee, MD, JD
• James P. Gills, III, MD and Joy Gills Professor of Ophthalmology
• Senior Advisor to the Chancellor and
• Director, Applied Health Systems Research, Duke University Health System
• Vice Chairman, Duke Eye Center
Jonathan McCall
• Senior Editor, Writing Services
• Communications Group, Duke Clinical Research Institute
Deborah A. Roth, MS
• Associate Dean for Clinical Research Administration, School of Medicine
• Chief Operating Officer, Duke Translational Medicine Institute
Landscape of Research
• Basic science has been advanced by focusing on
independent work of individuals and small teams.
– Theories and concepts evolve through serial hypothesis
testing and replication/refutation by independent
laboratories.
– In the past, the route from a basic discovery to clinical
application was long, with many steps in which different
people could test concepts.
Current Landscape
• Systems biology is evolving as a concept.
– Requires multidisciplinary teams
– Complex data sets and multiple measures
• The route from discovery to human application is
becoming shorter due to increasing discovery
research with human samples and tissues.
• Administrative structures of academic health and
science systems are more complex.
Recommendation 1:
Data Provenance
Prior to application of research to humans in a manner that
directs diagnostic testing or therapy either in clinical studies
or in commercialization, findings must be reproducible (both
internally and externally) based on additional documented
data sets. Documentation of internal reproducibility should
extend in a readily traceable manner back to the
fundamental source of the original data, a concept known
as data provenance. External reproducibility should be
demonstrated in data external to those used to produce the
original findings. Optimally, this could include a requirement
that the data analysis be reproduced in an independent
facility in addition to being replicated in second data sets.
Recommendation 2:
The Imperative to Include Adequate
Quantitative Expertise
Scientific studies must be conducted with adequate
quantitative expertise, applied under proper standards
to the design, conduct, analysis, and presentation of
results.
Recommendation 3:
Accountability for Quality
Governance and oversight of science with direct
implications for patient care can no longer be seen as
residing solely within the purview of any one
laboratory. Given consideration of public investment,
public trust, and direct impact on the wellbeing of
patients, a chain of accountability extending beyond
the individual laboratory to scientific leaders at local
and institutional levels must be established.
Recommendation 4:
Special Review of Conflicted Research
In the specific case of diagnostic or prognostic tests
developed within the institution for which financial gain
for scientists and/or the institution is possible or other
forms of conflict are evident, a panel of independent
experts should be convened to review the science
underlying the test before the test is used in human
studies or licensed.
Conclusions of the TMQF Committee
•
•
•
•
This is only the beginning of what we believe will be a constantly evolving
scientific framework, one characterized by increasing transparency and
global sharing to produce new knowledge at accelerating rates
Understanding that data provenance, quantitative expertise,
accountability, use of genomic tools by scientists with deep understanding
of the underlying biology, and external review are all critical elements of
this process should improve the capacity to move from discovery to
improved human health.
We are equally convinced that shortcuts or lack of attention to the checks
and balances noted in this report would be detrimental and even
potentially damaging, especially in research involving human research
participants.
We will endeavor to work collaboratively with leading institutions to
implement these ideas (and better ideas as they evolve) in an efficient
manner that encourages new advances among the community of
scientists and practitioners.
Plans for Implementation
• After a 1-month period for input from faculty and
staff, the TMQF document will be finalized.
• In parallel, an Implementation Plan is being
developed by the TMQF Committee.
• The Implementation Plan will require:
– A determined effort to foster an institutional culture in which
dissenting opinions and critical evaluation of colleagues are
encouraged; and
– An investment in critical infrastructure support.